2010; 14: 1–122 doi: 10.1142/s108842461000188x contents...gustavo t. ruiz, alexander g. lappin and...

CONTENTS

Journal of Porphyrins and PhthalocyaninesJ. Porphyrins Phthalocyanines 2010; 14: 1–122DOI: 10.1142/S108842461000188X

pp. 1–40The key role of peripheral substituents in the chemistry of phthalocyanines and their analo gsEvgeny A. Lukyanets† and Victor N. Nemykin

General strategies for the preparation of peripherally substituted phtha-locyanines, naphthalocyanines, anthracyanines, aza-analogs of phthalo-cyanines, and tetraazaporphyrins have been discussed. An infl uence of the types and positions of the substituents attached to the macrocyclic core on the optical properties of these compounds has been general-ized. In many cases, poorly known methodologies, published in mostly unavailable Russian journals, were highlighted throughout this review.

See Evgeny A. Lukyanets and Victor N. Nemykin pp 1–40

Phthalocyanines and their structural analogs are well-known for their traditional and contemporary high-tech applications and schematically shown on the background of the illustration. The typical UV-vis spectra and colors of tetraazaporphyrins, phthalo-cyanines, and naphthalocyanines as well as their core structures are also shown.

pp. 41–46Syntheses and structural studies of 5-pentamethyl-cyclopentadienyl rhodium(III) and iridium(III) complexes of a Schiff-base expanded porphyrinLuciano Cuesta, Vincent M. Lynch and Jonathan L. Sessler*

The synthesis of new binuclear rhodium(III) and iridium(III) semi-sandwich complexes of a Schiff-base expanded porphyrin is reported here. Single crystal X-ray diffraction analyses and complementary NMR studies provide evidence for the existence of strong intramolecular hydrogen-bonding interactions be-tween the pyrrolic NH protons and the chloride ligand both in dichloromethane solution and in the solid state.

N

NN

N

N

N

N

N

R2

R1R2

R1

R2

R1R2

R1

M

N

N

N

N

N

N

N

NM

R4R3

R2

R1

R4

R3

R2R1 R4 R3

R2

R1

R4

R3

R2R1

N

NN

N

N

N

N

NM

R1

R2

R3

R4 R5

R6 R1

R2R3

R4

R5

R6

R1

R2

R3

R4R5

R6R1

R2R3

R4

R5

R6

MTAP

MPc

MNc

About the Cover

Review

Articles

CONTENTS

J. Porphyrins Phthalocyanines 2010; 14: 1–122

pp. 55–63Through space singlet energy transfers in light- harvesting systems and cofacial bisporphyrin dyadsPierre D. Harvey*, Christine Stern, Claude P. Gros and Roger Guilard*

This review focuses on the photophysical processes and dynamics in the light harvesting devices, notably LH II, in the heavily investigated purple photosynthetic bacteria, and compares these parameters with our cofacial bisporphyrin dyads build upon octa-etio-porphyrin chromophores and rigid and semi-rigid spacers.

pp. 64–68Photooxidation of phenol derivatives using -(dihy-droxo)dipalladium(II) bisporphyrin complexYuko Takao*, Toshinobu Ohno, Kazuyuki Moriwaki, Fukashi Matsumoto and Jun-ichiro Setsune*

μ-(dihydroxo)dipalladium(II) complex with N21,N22-etheno bridged tetraphenyl-porphyrin ligand was employed in the catalytic photooxidation of phenol de-rivatives in aerated homogeneous solution with visible light irradiation. The Pd complex promoted degradation of p-tert-butylphenol and selective photooxida-tion of some phenol derivatives to afford the corresponding quinones. This Pd complex showed higher light durability compared with ordinary porphyrin free base and its Pd complex.

N

N N

N

Mg

OO

R

O

OO

CH3O

N

N N

N

Mg

O

OO

O

R

N

N N

N

MN

N N

N

M'

Spacer

BChl a Special Pair

Face to face biporphyrinsCH3O

O

pp. 69–80Morphology and radical reactions of Cu(II) and Co(II) sul-fonated phthalocyanines covalently linked to poly (ethylene-amide)Gustavo T. Ruiz, Alexander G. Lappin and Guillermo Ferraudi*

The tetrasulfonated CuII(tspc)4-, tspc = 4,4′,4″,4′″-phthalocyaninetetrasulfonate, and the trisulfonated CoII(trspc)3-, trspc = n,n′,n″-phthalocyanine- trisulfonate where n, n′ and n″ indicate the sulfonated positions of the various isomers, were covalently linked to a poly(ethyleneimine). The redox reactions of the phthalocyanine pendants with pul se radiolytically generated radicals (e-

sol, C•H2OH, (CH3)2COHC•H2, CO2

•-, N3• and SO4

•-)produce phthalocyanine radicals, Cu(I) and CoIII-hydroxyalkyl reaction inter me diates.

pp. 47–54Synthesis of near-infrared absorbed metal phtha-locyanine with S-aryl groups at non-peripheral positionsKeiichi Sakamoto*, Eiko Ohno-Okumura, Taku Kato and Hisashi Soga

The target compounds were synthesized: 15 phthalocyanines from 2, 3-dicyanohydroquinone in 3 steps via 1,2-dicyanobenzene-3,6-bis-(tri fl uorate) and 1,2-dicyanobenzene-3,6-thiophenols. The Q bands of obtained compounds appeared in the near-infrared region. In particular, 1,4,8,11,15,18,22,25-octakis(thiophenylmethyl)phthalocyaninate lead shows a Q band at 857 nm. Furthermore, non-colored transparent fi lms in the visible region can be produced.

CONTENTS


pp. 101–107Synthesis and self-assembly of a novel cobalt(II) porphyrin lipoic acid derivative on goldChristoph S. Eberle, Ana S. Viana, Franz-Peter Montforts and Luisa Maria Abrantes*

A novel cobalt(II) porphyrin lipoic acid derivative was synthesized starting from deuteroporphyrin(IX)bis-alcohol and enantiomerically enriched lipoic acid. The two disulfi de functionalities of the lipoic acid moieties allowed its immobili zation on gold by a self-assembly method. It was found that the Co(II) porphyrin self-assembled monolayer is electroactive and exhibits catalytic activity towards reduc-tion of oxygen.

pp. 108–114Synthesis, characterization and electrochemistry of the novel metalloporphyrazines annulated with tetrathiafulvalene having pentoxycarbonyl subs-tituentsFengshou Leng, Ruibin Hou, Longyi Jin, Bingzhu Yin* and Ren-Gen Xiong*

Three novel tetrathiafulvalene-annulated metalloporphyrazines with electron-withdrawing pentoxycarbonyl groups at the periphery were syn thesized via the cyclotetramerization of dipentyl 6,7-dicyanotetra-thia fulvalen-2,3-dicarboxylate in the presence of corresponding metal salts (Zn(OAc)2·2H2O, Cu(OAc)2·2H2O and NiCl2) and in pentanol. Solution electrochemical data showed one reductive and three oxidati ve processes within a -2000 mV to +2200 mV potential window.

N

N

N

NN

N

N

N

SS

S

S

S S

S

S

SS

S

S

S S

S

S

RO2C CO2R

CO2R

CO2R

RO2C CO2R

RO2C

RO2CM

M = Zn, Cu, NiR = pentyl

S

S

S

S CO2R

CO2R

NC

NC

pp. 81–88Porphyrin self-assembled monolayers and photo dynamicoxidation of tryptophanIsabelle Chambrier, David A. Russell, Derek E. Brundish, William G. Love, Giulio Jori*, M. Magaraggia and Michael J. Cook*

Self-assembled monolayer (SAM) fi lms of purposely synthesized zinc and ma-gne sium diporphyrin derivatives, 1b and 1c, have been deposited on the surface of gold-coated glass substrates. The SAM fi lms were characterized by RAIR and fl uorescence spectroscopy. The potential for the fi lms to promote the oxidation of tryptophan within human serum albumin upon irradiation with white light has been demonstrated and attributed to the porphyrins acting as photosensitizers of oxygen to form oxidizing species.

N N

NNM

RO

RO

RO

RO

OROR

OR

OR

OR

O(CH2)12S2

M= Zn

M= Mg1c

1b

pp. 89–100Effi cient synthesis of an A-B-C-tricycle fragment for a structural model of tolyporphinBing C. Hu*, Wei Y. Zhou, Zu L. Liu, Chao J. Cai and Shi C. Xu

An effi cient stereocontrolled synthesis of an A-B-C-tricycle fragment for a structural model of tolyporphin is described. One of the two key steps is a se lective ring-opening reaction of the lactone cycle which introduces the chirality into synthetic compounds. The other key step is the combination of A ring with B-C-bicycle via a two-time Eschenmoser sulfi de contraction.

NH

CH3

H3COOCCH3CN

HN

N

Br

CH3

CH3

EtO2C

HN

CH3O

CH3

HN

Br

CHO

EtO2C

NH

CH3 S

CH3CNH3CO2C

O

O

O t-Bu

O

+

+

B

C

A

CONTENTS


pp. 115–122Inter-ring interactions in [Fe(TalkylP)(Cl)] (alkyl = ethyl, n-propyl, n-hexyl) complexes: control by meso-substi-tuted groupsMing Li, Teresa J. Neal, Noelle Ehlinger, Charles E. Schulz* and W. Robert Scheidt*

We report syntheses, molecular structures and magnetic susceptibilities of three meso-substituted high-spin iron(III) porphyrinate complexes ([Fe(TEtP)(Cl)], [Fe(TPrP)(Cl)], and [Fe(THexP)(Cl)]). It was determined that the inter-ring in-teractions within each dimeric unit change upon alteration of the alkyl groups at the meso positions. Magnetic exchange couplings between iron centers of the dimers are in accord with the trends in structural inter-ring geometries.

AUTHOR INDEX (cumulative)

AAbrantes, Luisa Maria 101

BBrundish, Derek E. 81

CCai, Chao J. 89Chambrier, Isabelle 81Cook, Michael J. 81Cuesta, Luciano 41

EEberle, Christoph S. 101Ehlinger, Noelle 115

FFerraudi, Guillermo 69

GGros, Claude P. 55Guilard, Roger 55

HHarvey, Pierre D. 55Hou, Ruibin 108Hu, Bing C. 89

JJin, Longyi 108Jori, Giulio 81

KKato, Taku 47

LLappin, Alexander G. 69Leng, Fengshou 108Li, Ming 115Liu, Zu L. 89Love, William G. 81Lukyanets, Evgeny A. 1Lynch, Vincent M. 41

MMagaraggia, M. 81Matsumoto, Fukashi 64Montforts, Franz-Peter 101Moriwaki, Kazuyuki 64

NNeal, Teresa J. 115Nemykin, Victor N. 1

OOhno, Toshinobu 64Ohno-Okumura, Eiko 47

RRuiz, Gustavo T. 69Russell, David A. 81

SSakamoto, Keiichi 47Scheidt, W. Robert 115Schulz, Charles E. 115Sessler, Jonathan L. 41Setsune, Jun-ichiro 64Soga, Hisashi 47Stern, Christine 55

TTakao, Yuko 64

VViana, Ana S. 101

XXiong, Ren-Gen 108Xu, Shi C. 89

YYin, Bingzhu 108

ZZhou, Wei Y. 89

Journal of Porphyrins and PhthalocyaninesJ. Porphyrins Phthalocyanines 2010; 14: 1–122

JPP Volume 14 - Number 1 - Pages 1–122

Aaggregation 108antracyanines 1aza-analogues of

phthalocyanines 1

Bbisporphyrin 55bridged porphyrin 64

CCo(II) porphyrin lipoic acid

derivative 101cobalt 69cofacial dyad 55copper 69crystal structures 115

Eelectrocatalytic activity 101electrochemistry 101, 108energy tranfer 55

Dμ-(dihydroxo)dipalladium complex

64

Ggeneral UV-vis spectra trends 1

Iinter-ring interaction 115intramolecular hydrogen bonds 41iron(III)meso-tetraalkylporphinates

115

LLH II 55

Mmagnetic exchange couplings 115metalloporphyrazines 108

N1,2-naphthalocyanines 12,3-naphthalocyanines 1near-infrared absorption 47non-peripheral position 47

Ppentamethylcyclopentadienyl half-

sandwich complexes 41phenol derivative 64photodynamic surface 81photooxidation 64phthalocyanine radicals 69phthalocyanine-derivatized

poly(ethylenimine) 69

porphyrin self-assembled monolayers 81

pulse radiolysis 69

Rreaction intermediates 69reaction kinetics 69rhodium(III) and iridium(III)

complexes 41

SSAM fi lms on gold 81S-aryl substituent 47scanning tunneling microscopy 101Schiff-base expanded porphyrin 41selective ring-opening reaction 89self-assembled monolayers 101structural model 89substituted phthalocyanines 1sulfi de contraction 89sulfonated phthalocyanine 69

Ttetraazaporphyrins 1tetramerization 108tetrathiafulvalenedinitrile 108tolyporphin 89

KEYWORD INDEX (cumulative)


JPP Volume 14 - Number 1 - Pages 1–122


DOI: 10.1142/S1088424610001799

Published at http://www.worldscinet.com/jpp/

Copyright © 2010 World Scientific Publishing Company

INTRODUCTION

Since their accidental discovery in the early 1900s [1–3] and accurate characterization in 1930s [4–8], phthalocyanines (Fig. 1) have been widely used as blue and green pigments in the paper and textile industries because of their high thermal, chemical, and photochemi-cal stabilities. Almost all early phthalocyanine complexes were unsubstituted on the periphery and (taking into con-sideration their planarity and extended π-system) had low solubility in the majority of known solvents. Indeed, the typical solubility of unsubstituted phthalocyanines in aliphatic and chlorinated hydrocarbons, alcohols, and aprotic polar solvents is less than 10-6–10-7 M [9]. Even in condensed aromatic solvents such as 1-chloro- and 1-bromonaphthalene, as well as quinoline, the solubility of unsubstituted phthalocyanines remains low (~10-5 M). Empirically, it was found that the best solvent for unsubstituted phthalocyanines is sulfuric acid, which easily gives a molar range concentration and can be

used for purifi cation of the target unsubstituted phtha-locyanine complexes [10]. Sulfuric acid, however, can-not be used for many labile phthalocyanine compounds because of the demetalation reaction. In addition, it proto-nates meso-nitrogens in the phthalocyanine core, dramati-cally changing its optical and redox properties [10, 11]. In general, the phthalocyanines can be modifi ed using three major pathways: (i) by changing the central atom, (ii) by changing the meso-atom and (iii) by modifying the periphery of the phthalocyanine core [12–14]. The scope and limitations of the fi rst two modifi cation approaches are rather well explored and thus in this review we will concentrate on the third possibility, i.e. peripheral modi-fi cation of phthalocyanine core. An introduction of the peripheral substituents into phthalocyanine core allows several key advantages [16, 20]. First, it can dramatically increase the solubility of the target phthalocyanine in water or common organic solvents. Peripheral substitution can also be used for an accurate tuning of optical and redox properties of desired phthalocyanine so it can be used in specifi c high-tech applications. Finally, peripheral substit-uents can be used as an anchor or bridging groups for the formation of controlled supramolecular assemblies and similar applications (i.e. heterogenization of catalysts).

The blossoming of peripherally substituted phtha-locyanines started in the middle of the 20th century, when controlled degree of peripheral sulfonation of

The key role of peripheral substituents in the chemistry of phthalocyanines and their analogs*

Evgeny A. Lukyanets†a and Victor N. Nemykinb

a Organic Intermediates & Dyes Institute, 1/4 B. Sadovaya St., Moscow 123995, Russiab Department of Chemistry & Biochemistry, 1039 University Drive, University of Minnesota Duluth, Duluth,MN 55812, USA

Received 3 December 2008Accepted 12 February 2009

ABSTRACT: The preparation and optical properties of peripherally substituted phthalocyanines and their analogs (i.e. naphthalocyanines, anthracyanines, aza-analogs of phthalocyanines, and tetra azaporphyrins) have been widely discussed. This review highlights methodologies that have been published in poorly known and mostly unavailable Russian journals.

KEYWORDS: substituted phthalocyanines, 1,2-naphthalocyanines, 2,3-naphthalocyanines, tetra-azaporphyrins, antracyanines, aza-analogues of phthalocyanines, general UV-vis spectra trends.

SPP full member in good standing

*This review is based on the presentation given by E.V. Lukyanates in dedication for Linstead Career Award in Phthalocyanine Chemistry.

†Correspondence to: Evgeny A. Lukyanets, email: [email protected], tel: +7 (495)-254-9547, fax: +7 (495)-254-1200

Copyright © 2010 World Scientific Publishing Company J. Porphyrins Phthalocyanines 2010; 14: 2–40

2 E.A. LUKYANETS AND V.N. NEMYKIN

phthalocyanine core resulted in production of a variety of –SO2XR substituted soluble phthalocyanines [9, 15, 16]. In addition, in the 1960s, an effective strategy for preparation of soluble in the most common organic solvents tert-butyl substituted phthalocyanines by tetramerization reaction of 4-tert-butylphthalonitrile was proposed [17–19]. Since then, the introduction of peripheral alkyl-, aryl-, or het-aryl substituents, halogens, thiols, amines, ethers, and other groups has been the subject of an enormous number of research papers. As a result, in addition to traditional applications as dyes and pigments [20], peripherally sub-stituted phthalocyanines are currently used as industrial catalytic systems [21]; photosensitizers for photodynamic therapy of cancer [22]; materials for electrophotography, ink-jet printing, semiconductor, chemical sensors, and electrochromic devices; functional polymers and liquid crystals; nanotechnology [23]; and nonlinear optics [24]. The majority of these applications take advantage of the unique optical (specifi cally low-energy Q band) and redox properties of phthalocyanines which can be fi ne-tuned using appropriate peripheral substituents [13, 14, 16, 20].

Peripheral substituents can be introduced into phthalo-cyanine core using one of the two basic methods. The fi rst approach involves the modifi cation of an already existing phthalocyanine core using, e.g. aromatic electrophilic substitution reactions [9, 13, 15, 16, 25]. This approach will only be briefl y discussed in this review. The second basic approach involves tetramerization of the substituted phthalocyanine precursors and leads to a controlled number of sub-stituents in the target phthalocyanine [9, 13, 14, 16, 20, 21]. The useful precursors for this method include but are not limited to substituted derivatives (i.e. anhydrides, imides, amides, and nitriles) of ortho-phthalic acid (Scheme 1). This approach will be primarily discussed below. Since one of the pioneer research laboratories in the fi eld of substituted phthalocyanines is located in Russia, and taking into consid-eration the poor availability of Russian- language journals outside the former Soviet Union countries, we will predominantly try to highlight methodologies developed

by Russian research groups; very accurate descrip-tions of research status in the fi eld of substituted phthalocyanines can be found in recently pub-lished books and reviews [12–14, 16, 20, 26–29].

PERIPHERALLY SUBSTITUTED PHTHALOCYANINES

Alkyl- and arylsubstituted phthalocyanines

Probably one of the most used and most convenient alkyl substituted phthalocyanines is the metal-free 2,9(10),16(17),23(24)-tetra-tert-butylphthalocyanine and its metal complexes (Scheme 2). This phthalocyanine was introduced in the late 1960s when a variety of transition-metal and double-decker lanthanide compounds with this macrocycle was described [17–19, 30–34]. Like other 2,9(10),16(17),23(24)-tetrasubstituted phthalocyanines, this compound consists of four possible constitutional isomers presented in Fig. 2 [13]. Since the infl uence of tert-butyl substituents on the electronic structure and thus on spectroscopic properties of phthalocyanine core is negligibly small, such isomeric mixture is suitable for the majority of applications.

The original [34] (and currently industrial) synthesis of key precursor (4-tert-butylphthalonitrile 7) starts from inexpensive ortho-xylene 1, which undergoes the Friedel-Crafts reaction with formation of 2. An oxidation of 4-tert-butyl-ortho-xylene 2 leads to formation of 4-tert-butylphthalic acid 3, which represents the starting reagent for so-called “acidic route” preparation of substituted phthalonitriles (substituted phthalic acid anhydride

imide amide nitrile). This very time consuming route requires overall six steps and provides the typical yield of the target nitrile 7 of ca. 25%. Not surprisingly,

N

N

N

N

N

N

N

NM

N

N

N

N

N

N

N

NM

1

2

3 4 8 9

10

11

15

16

171822

23

24

25

α

α α

α

α

αα

α

ββ

β

β ββ

ββ

Fig. 1. Numbering scheme for phthalocyanine substituents

O

O

O

R

NH

O

O

R

NH

NH

NH

R

NH2

NH2

O

O

R

CN

CNR

OHOH

O

O

R

CN

NH2

O

R

N

N

N

N

N

N

N

NM

R R

RR

Scheme 1. Synthetic pathways and typical precursors for preparation of substi-tuted phthalocyanines


THE KEY ROLE OF PERIPHERAL SUBSTITUENTS IN THE CHEMISTRY OF PHTHALOCYANINES AND THEIR ANALOGS 3

the large-scale laboratory preparation of nitrile 7 was later simplifi ed by using a two-step procedure, which fi rst requires the regioselective bromination of eas-ily available tert-butylbenzene 8 followed by the classic Rosenmund-von Braun reaction between 1,2-dibromo-4-tert-butylbenzene 9 and CuCN (Scheme 2) [12] An industrial preparation of 4-tert-butylphthalonitrile can be achieved by the direct transformation of anhydride 4 using high-temperature catalytic ammonolysis reac-tion [12]. The presence of bulky tert-butyl substituents

in phthalocyanines 10 results in dramatic increase of solubility of these compounds in various common organic solvents, with-out signifi cant shift of Q band in UV-vis spectra compared to that observed in corre-sponding unsubstituted analogs [35]. Sim-ilarly, preparation of even more sterically crowded 4-adamantylphthalonitriles and corresponding 2(3),9(10),16(17),23(24)-tetraadamantylphthalocyanes 11 (Fig. 3) has been reported in the early 80s [36].

The attempt to control a favorable for-mation of sterically less crowded C4h con-stitutional isomer of nickel 1,3,8,10(9,11),15,17(16,18),22,24(23,25)-octa-tert-butylphthalocyanine 12 (Fig. 4) failed [37]. For instance, the yield of the least sterically crowded C4h isomer is signifi cantly lower from that expected on a basis of statistical distribution and close to that observed for the most sterically crowded D2h isomer. In this case, the isomer distribution pattern was tentatively explained as a result of

the specifi city of template tetramerization of 3,5-di-tert-butylphthalonitrile precursor which leads to the initial formation of sterically unrestricted phthalonitrile-based “dimers”.

Preparation of 2,3,9,10,16,17,23,24-octasubstituted, 1,4,8,11,15,18,22,25-octasubstituted and 1,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-hexadecasubstituted phthalo -cyanines allows to avoid the formation of any constitu-tional isomers shown in Fig. 2 and thus target substituted phthalocyanines can be prepared as a single isomer [13, 20]. One of the possible synthetic pathways requires the catalytic coupling of ortho-dihalobenzenes with appro-priate haloalkanes, followed by bromination and fi nally the Rosenmund-von Braun reactions. Alternatively, cyclic alkyl substituents can be introduced into the phthalo-cyanine core by “acidic” methodology (compounds 13–18)presented in Scheme 3 as was shown in the late 80s [38].

Scheme 2. Synthetic pathway for the preparation of 2,9(10),16(17),23(24)-tetra-tert-butylphthalocyanines 10

N

N

N

N

N

N

N

NMO

O

O

NH

O

O

Br

Br

NH2

NH2

O

O

CN

CN

OHOH

O

O 123

4

5 67

8

9

10

N

N

N

N

N

N

N

NM

R

R

R

R

N

N

N

N

N

N

N

NM

R

R

R

R

N

N

N

N

N

N

N

NM

R

RR

R

N

N

N

N

N

N

N

NM

R

R R

R

CsC4h

D2h C2v

Fig. 2. Four possible constitutional isomers of 2(3),9(10),-16(17),23(24)-tetrasubstituted phthalocyanines

N

N

N

N

N

N

N

NM

11

Fig. 3. Structure of 2,9(10),16(17),23(24)-tetraadamantyl-ph thalo cyanines



Alternative reaction pathway for preparation of this nitrile starts from the bromination reaction of commercially available 19, followed by the Rosenmund-von Braun reaction (Scheme 3) [39]. Phthalocyanines 21 as well as their alicyclic fi ve-membered analogs [40] have comparable with 2,9(10),16(17),23(24)-tetra-tert-butylphthalo-cyanines 10 solubility in common organic sol-vents, while the presence of alicyclic substituents in the molecule does not signifi cantly affect their optical properties. Metal-free phthalocyanine 21can be used as a luminescence sensor for laser- induced singlet oxygen generation in solution [41].

The dichloromaleo(fumaro)nitrile 23 [42] can be used as an universal precursor for preparation of the numerous substituted phthalonitriles 25(Scheme 4). Nitrile 23 can be easily prepared by the direct chlorination of inexpensive succino-nitrile with chlorine gas at elevated temperature. Diels-Alder reaction between 23 and substituted dienes 22 leads to formation of the cyclic inter-mediate 24, which can be easily aromatized into substituted nitrile 25 at elevated temperatures with the formation of 2,3- (25a), 1,4- (25b), and 1,3- (25d) disubstituted phthalonitriles as well as 1-substituted phthalonitriles 25c. The yield of this reaction varies with the nature of substituted diene and has been found to be between 16 and 73% [43].

N

N

N

N

N

N

N

NNi

N

N

N

N

N

N

N

NNi

N

N

N

N

N

N

N

NNi

N

N

N

N

N

N

N

NNi

C4h

D2h

Cs

C2v

Statistical: 12.5%Experimental: 6%

Statistical: 12.5%Experimental: 2%

Statistical: 50%Experimental: 69%

Statistical: 25%Experimental: 23%%

Fig. 4. Statistical and experimental distribution of constitutional isomers in nickel 1,3,8,10(9,11),15,17(16,18),22,24(23,25)-octa-tert-butylphthalocyanine 12 [37]

N

N

N

N

N

N

N

NMO

O

O

NH

O

O

Br

Br

NH2

NH2

O

O

OHOH

O

O

ClCl

1314

15

16 17

19

20

21

+

CN

CN

18

Scheme 3. Synthetic pathway for preparation of alicyclic phthalocyanines 21



Another pair of useful precursors for preparation of the tetrasubstituted phthalonitriles 29 consists of well-docu-mented tetrasubstituted cyclopentadienone (“cyclone”) [44, 45] compounds 26 and chloromaleonitrile 27(Scheme 5) [46]. Diels-Alder reaction between these two precursors leads to formation of bicyclic intermedi-ate 28, which upon heating in high boiling point solvents (i.e. chlorobenzene) eliminates carbon monoxide and hydrogen chloride with formation of the target tetrasu-bstituted nitrile 29 in high yields. Use of easily avail-able tetrasubstituted cyclopentadienone precursors 26allow the preparation of several tetrasubstituted phtha-lonitriles with both aryl- and alkyl-substituents in ben-zene ring. Sterically hindered nitriles 29 can be used for preparation of the low-symmetry phthalocyanine analogs using so-called sterically-controlled method (Scheme 6)

[47–53]. In this method, the low-symmetry phthalocya-nines 30a–30d can be prepared by cross-condensation of sterically hindered and sterically unhindered phthalo-nitriles A and B, respectively. Target compounds can be easily separated by preparative TLC method, while an isolated yield of the reaction products follows the order of increase of steric hindrance, i.e. 30e > 30d > 30c > 30b > 30a [49]. Although the steric interactions between phenyl groups located at 1,4-positions of each isoindole fragment prevent direct formation of hexadecaphenyl-substituted transition-metal phthalocyanines, it has been recently shown that analog zinc complexes substituted at 1,4,8,11,15,18,22,25-positions can be prepared using modifi ed approach in low yield [54].

The preparation of another interesting sterically crowded, so-called “concave” phthalocyanine complexes

R3

R2

R4

R1

R3

R2

R4

R1

Cl

ClCNCN

NCC(Cl)=C(Cl)CN

2322 24

+-2HCl

25a: R2 = R3 = CH3, C6H525b: R1 = R4 = C6H525c: R1 = C6H5, o-CH3C6H4, C(CH3)325d: R2 = R4 = C(CH3)3

CN

CN

R1

R2

R3

R4

25

Scheme 4. Illustration of use of dichloromaleo(fumaro)nitrile for preparation of the substituted phthalonitriles 25a–25d using the Diels-Alder reaction

O

CN

CNH

Ph

Ph R

R

ClO

Ph

Ph

R

R

RCN

CNR

Ph

Ph

CN

CN

Cl

-HCl

-CO

+

26 27 28 29

R = Ph, Me, Et, nPr

Scheme 5. Preparation of tetrasubstituted phthalonitriles 29 using tetrasubstituted cyclopentadienones 26

PhCN

CNPh

Ph

Ph CN

CN N

N

N

N

N

N

N

N

R1

R1

R1 R1 R2R2

R2

R2

R3

R3

R3R3R4R4

R4

R4

M

A B

29

+MXn

30a: R1= R2 = R3 = Ph; R4 = H30b: R1 = R2 = Ph; R3 = R4 = H30c: R1 = R3 = Ph; R2 = R4 = H30d: R1 = Ph; R2 = R3 = R4 = H30e: R1 - R4 = H

Scheme 6. Preparation of the low-symmetry phthalocyanines 30a–30d using sterically hindered tetraphenylphthalonitrile 29



33 also includes the initial Diels-Alder reaction between anthracene and generated in situ 4,5-dibromobenzyne (Scheme 7) [55]. The resultant 2,3-dibromotriptycene 31can be introduced into the Rosenmund-von Braun reaction to afford 2,3-dicyano triptycene 32 in moderate yield. Unlike dibenzobarrelene analog discussed below, compound 32 can be easily cyclized into respective sterically crowded phthalo-cyanine complexes 33 using general methods for the prepa-ration of substituted phthalocyanines. The sterically crowded triptycene carcass creates a double “concave” cavity while not affecting optical properties of phthalocyanines 33.

Catalytic ammonolysis is an important synthetic pathway for the multi-gram preparation of alkyl- and aryl-substituted phthalonitriles in a single step, which is useful for industrial-scale production (Scheme 8) [56]. Reaction is typically conducted using appropriate anhy-drides or imides, gaseous ammonia and borophosphate

catalyst [57] at ca. 400 °C. The yields of target phthalo-nitriles vary between 45 and 80%, which is in a range of yields known for multi-step preparation of corresponding nitriles using the traditional “acidic” or Rosenmund-von Braun reaction pathways.

Overall, introduction of the peripheral alkyl and aryl substituents into phthalocyanine core can be achieved via macrocyclization of the respective 3-, 4-, 4,5-, 3,5- and 3,4,5,6-substituted nitriles prepared by “acidic”, the Rosenmund-von Braun, the Diels-Alder, or catalytic ammonolysis reaction pathways.

Phthalocyanines with electron-withdrawing groups

Introduction of the nitro group into α- or β-positionsof phthalocyanine core (Fig. 1) is a well-known method for preparing transition-metal phthalocyanines with

NN

N

N

N

NMN

N

R

R

R

R

R

RR

R

R

R

Br

Br CN

CN

R

R

Br

Br

R

R

32

+

31

33R = H, t-Bu

Scheme 7. Preparation of sterically crowded “concave” phthalocyanines 33

CN

CNX

O

O

RCN

CNR

CN

CN

CN

CN

CN

CN

CN

CNCl

CN

CN

Cl CN

CN

catalyst

NH3, 400oC

Scheme 8. Catalytic ammonolysis reaction pathways for preparation of the alkyl and aryl substituted phthalonitriles

R = Me, t-Bu, Cl, Br



four electron-withdrawing groups [13, 20, 58–62]. The preparation of both 1,8(11),15(18),22(25)- and 2,9(10),16(17),23(24)-tetranitrophthalocyanine can be easily achieved using prepared by “acidic” method 3- and 4-nitrophthalonitrile, respectively, although the fi rst methods suggested at the end of the 1960s also used more exotic treatment of 1,2-bis(dichloromethyl)-4-nitrobenzene with gaseous ammonia in the presence of transition-metal salts [63, 64]. In 1991, we developed a new methodology for synthesis of transition-metal 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octanitrophthalo-cyanines. The key precursor for preparation of these compounds is 3,5-dinitrophthalonitrile 34, which can be prepared by two different ways using nucleophilic aro matic substitution reactions in either 1,2-dibromo-3,5-dinitrobenzene or 2-hydroxy-3,5-dinitrobenzonitrile (Scheme 9) [65]. Although both methods look similar to each other, it is interesting to note that the nitrile 34can be obtained in high yield only by using 2-hydroxy-3,5-dinitrobenzonitrile but not 1,2-dibromo-3,5-dini-trobenzene because in this case copper 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octanitrophthalocyanine is the major reaction product. The presence of eight strong electron-withdrawing groups at both α- and β-positionsof phthalocyanine ring results in very unusual redox properties of these compounds. Specifi cally, no phthalo-cyanine-based oxidation was found within the solvent range, while reduction potentials of these complexes are ~1 V more positive compared to those observed in alkyl-substituted complexes [66]. In general, transition-metal 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octan-itrophthalocyanines are so easily reduced that special care should be taken for solvent purifi cation in order to avoid any traces of minor reducing agents. For instance, the traces of dimethylamine in regular grade DMF can easily reduce cobalt(II) 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octa nitrophthalocyanine into corresponding cobalt(I) compound, which was confi rmed by UV-vis spectra compared to those obtained using spectroelec-trochemical approach [65, 66]. Electrochemical and

chemical reduction data indeed suggest that the eight nitro groups presented in 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octanitrophthalocyanines have more elec-tron-withdrawing power compared to that observed in 2,3,9,10,16,17,23,24-octacyanophthalocyanines [20, 67, 68]. A large stability toward oxidation makes tran-sition-metal complexes of 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octanitrophthalocyanine potentially useful candidates for a variety of oxidative catalytic reactions, for instance C-H bond activation in organic compounds [69–71].

Low solubility of 1,8(11),15(18),22(25)-tetranitro-phthalocyanines, 2,9(10),16(17),23(24)-tetranitrophtha-locyanines and 1,3,8,10(11,9),15,17(18,16),22,24(25,23)-octanitrophthalocyanines, which are soluble only in dipo-lar aprotic solvents and acids, stimulate exploratory work on the preparation of nitro-groups containing phthalo-cyanines that are soluble in common organic solvents. As a result, a synthetic pathway for the preparation of 1,8(11),15(18),22(25)-tetranitro-3,10(9),17(16),24(23)-tetra-tert-butylphthalocyanines has been developed (Scheme 10) [72, 73]. Nitration of readily available 4-tert-butyl-ortho-xylene results in formation of 5-tert-butyl-1,2-dimethyl-3-nitrobenzene, methyl groups of which can be oxidized into carboxylic one by potassium permanganate. Treating 5-tert-butyl-3-nitro-ortho-phthalic acid sequen-tially with acetyl chloride, urea, concentrated ammonia solution, and fi nally phosphorus oxychloride leads to the target 5-tert-butyl-3-nitro-phthalonitrile 35. Nitrile 35 can also be obtained using 3-nitro-4,5-dibromo-tert-butylbenzene by the Rosenmund-von Braun reaction. Macrocyclization of nitrile 35 results in the formation of 1,8(11),15(18),22(25)-tetranitro-3,10(9),17(16),24(23)-tetra-tert-butylphtha locyanines that are highly soluble in common organic solvents [72, 73]. Similar to nitrile 35, the solubility of 1,8(11),15(18),22(25)-tetrabromo-phthalocyanines can be increased by introduction of tert-butyl groups into β-positions of phthalocyanine core. In this case, a key precursor, 5-tert-butyl-3-bromo-phthalonitrile can be obtained in two steps starting from

NO2

O2N

CN

CN

NO2

O2N

Br

Br

NO2

O2N

OH

CN

NO2

O2N

CN

Br

NO2

O2N

Br

CN

CuCN CuCN CuCN

34

1. Py, EtOH

2. PBr3, DMF

Scheme 9. Synthetic pathways for the preparation of 3,5-dinitrophthalonitrile

NO2 NO2

O

O

O

NO2

NH

O

O

CN

CN

NO2

HNO3 (NH2)2CO1. KMnO4

2. AcCl

1. NH4OH2. POCl3

35

Scheme 10. Synthetic pathway for preparation of 5-tert-butyl-3-nitrophthalonitrile



2-bromo-4-tert-butyl-6-nitrobenzonitrile using fi rst its nitro group reduction by tin(II) chloride, followed by Sandmeyer reaction with cuprous cyanide [74].

Current commercial availability of 3- and 4-nitro-phthalonitriles allows simple route for preparation of respective 3- and 4-aminophthalonitriles. These nitriles are excellent candidates for Sandmeyer reaction, which can easily transform amino group into halogen, cyano, sulfo or thiol substituents in 7–75% yields (Scheme 11) [75, 76]. Cyclization of these nitriles leads to the forma-tion of 1,8(11),15(18),22(25)- and 2,9(10),16(17),23(24)-tetrasubstituted phthalocyanines with corresponding electron-withdrawing groups.

Water-soluble phthalocyanines with peripheral elec-tron-withdrawing groups are very important for photo-dynamic therapy of cancer (PDT) [20, 77]. In particular, metal-free, aluminum, and zinc phthalocyanine sulfoac-ids alkali metal salts are widely investigated photosen-sitizers for PDT. It has been shown [78] that sodium

salt of 2,9(10),16(17),23(24)-tetrasulfophthalocyanine-based photosensitizers prepared from the sodium salt of 4-sulfophthalonitrile has lower activity in PDT compared to similar salts (1–3 sulfogroups) obtained by direct sul-fonation of unsubstituted phthalocyanine core. It was found that direct sulfonation of variety of metal-free or metal-containing phthalocyanines using traditional sul-fonation agents in trichlorobenzene leads to formation of corresponding polysulfonated (n = 2–4) phthalocyanine compounds. In the case when central metal is labile in acidic conditions, central metal can be easily introduced into the phthalocyanine cavity by treating metal-free polysulfonated phthalocyanine or appropriate alkali metal salt with the respective metal precursor (Scheme 12) [79, 80]. Sodium salt of hydroxyaluminum tetrasul-fophthalocyanine (“photosens”) is approved for PDT in Russia and sodium salt of metal-free trisulfophthalo-cyanine (“phthalosens”) [81] is undergoing pre-clinical studies [82–85]. Some results of PDT treatment with

2) BuONO in AcOH+ EtOH

1) H2SO4

or BuONO in HBF4 + EtOH

CN

CNR

CN

CNNH2

CN

CNH2N

CN

CNR

CN

CNN2

CN

CNN2

R = I (20 %), CN (48 %), SO3H (75%), SH (42%)

R = I (34 %), Br (25 %), CN (32 %), p-NO2C6H4 (30 %), SO3H (45%), SH (7%)

Scheme 11. 3- and 4-substituted phthalonitriles prepared using Sandmeyer reaction

N

N

N

N

N

N

N

N

MN

N

N

N

N

N

N

N

M

SO3H(Na)

SO3H(Na)(Na)HO3S

(Na)HO3S

HSO3Cl

trichlorobenzene

N

N

N

N

N

N

NH

HN

SO3H(Na)

SO3H(Na)(Na)HO3S

(Na)HO3S

N

N

N

N

N

N

N

N

M

SO3H(Na)

SO3H(Na)(Na)HO3S

(Na)HO3S

MXn

DMF

M = HH, Zn, AlCl, Cu et al.

M = Mg, MnCl, GdCl

Scheme 12. Synthetic pathways for preparation of water-soluble sulfonated phthalocyanines



“photosens” photosensitizer collected over 14 years are presented in Table 1, suggesting that “photosens” is indeed very promising photosensitizer, which can be used for the treatment of cancer of various localizations [86].

Photodynamic activity of sulfonated phthalocyanines can be signifi cantly improved by the targeted delivery of photosensitizer into cancer cells. One of the possi-bilities which has been explored in detail is the use of bio-conjugates of water-soluble aluminum phthalocya-nine with several proteins and growth factors. In partic-ular, the conjugates of aluminium phthalocyanine with oncofetal protein human alpha-fetoprotein and epidermal growth factor exhibit much higher activity against dif-ferent types of tumor cells in comparison with parent water-soluble phthalocyanine [87, 88]. Another possible way to improve the targeted delivery of photosensitizer into cancer cells is to prepare biotin-linked phthalocya-nine assemblies (Scheme 13) [89–92]. Such assemblies

can cause in vivo destruction of Ehrlich carcinoma tumor tissue, with total necrosis of tumor tissue and expressed vascular damage even at very low concentration (0.25 mg/kg) of body weight.

Some conjugates of phthalocyanines with oligonu-cleotides can also be used as reagents for photosensi-tized or catalytic DNA modifi cations. In this case, the oligonucleotide parts of conjugate are responsible for the recognition of selected complementary sequences on the DNA target, while metal phthalocyanine induce the DNA modifi cation [93]. Specifi cally, zinc and aluminum complexes can act as regular photosensi-tizers, while cobalt analog promotes DNA oxidation by molecular oxygen as a catalyst [94]. These results suggest the possible use of phthalocyanine-oligonucle-otide conjugates as novel artifi cial regulators of gene expression and therapeutic agents for treatment of cancer [95].

NN

N

N

N

NHOAlN

N

SO2NH

SO3H

HNO2SNHHN

S

O

H H

H

HN

O

NH

O

HN NH

S

O

HH

H

N

N

N

N

N

NHOAlN

N

SO2Cl

SO2Cl

ClO2S

NHHN

S

O

H H

H

HN NH2

O

+

DMSO/NEt3

Scheme 13. Preparation of biotinylated aluminum sulfophthalocyanine

Table 1. Results of PDT treatments with “photosens” based on investigations in Russia (1994-2008)

Cancer localization Patients Full regression Partial regression Stabilization No effect

Radical treatment 1509 1014 (67%) 414 (27%) 72 (5%) 9 (1%) (skin, pharynx, tongue, lip, mucous membrane of oral cavity, early cancer of stomach and esophagus, vulva, urinary bladder etc.)

Palliative treatment 880 209 (24%) 359 (41%) 272 (31%) 40 (4%) (MTS of breast cancer and melanoma, malignant pleuritis, stomach and esophagus cancer of III–IV stages etc.)



Complementary to these water-soluble phthalocya-nines, highly soluble in organic solvents, tetra- and octa-sulfonamide-containing phthalocyanines with –SO2NRR′residues (R = H, R′ = Et, tert-Bu and R = R′ = Me, Et, Bu) can be prepared either from dicyanobenzenesulfon-amides or from the corresponding dibromides and CuCN [96]. The alkylsulfamoyl groups increase the solubility of the phthalocyanines in common organic solvents with-out signifi cantly affecting the absorption spectra [97]. Formation of sulfonamides from corresponding sulfo-chlorides is probably the most useful synthetic pathway for heterogenization of transition-metal phthalocyanines onto anchoring amino group containing silica gels and carbons [98–101]. These heterogeneous catalysts can be further used in a variety of catalytic transformations of organic substrates, for instance in oxidation of organic thiols into corresponding disulfi des (so-called “Merox” process) [102].

Phthalocyanines substituted with carboxylic acid residues represent another large group of macrocycles with electro-withdrawing substituents. For instance,

water-soluble sodium salt of cobalt 2,3,9,10,16,17,23,24-octa carboxyphthalocyanine, so-called “theraphthal” [103–105], is currently under clinical trials for catalytic therapy of cancer. In this approach, molecular oxygen can be involved in specifi cally triggered catalytic reac-tion with transition-metal phthalocyanines as catalysts giving oxygen-reactive forms [106, 107]. Another poten-tial use in cancer treatment therapy are the recently pre-pared water-soluble 2,3,9,10,16,17,23,24-octacarboxy phthalocyanine complexes 36–39, shown in Fig. 5 [108, 109]. Similarly to sulfonated phthalocyanines discussed earlier, anti-cancer activity of cobalt phthalocyanine-2,3,9,10,16,17,23,24-octacarboxylate can be (at least potentially) increased by preparation of the aqua com-plexes of covalent conjugates of appropriate phthalocya-nine and platinum(II) (complex 39 and its analogs) [109, 110]. The pentanuclear tetraplatinate complex 39 can be easily prepared by the reaction between octasodium cobalt(II) phthalocyanine-2,3,9,10,16,17,23,24-octacarboxylate and K2PtCl4. In this case, the catalytic activity of cobalt 2,3,9,10,16,17,23,24-octacarboxyphthalocyanine

NN N

Zn

NN N

N N

COONa

NaOOC

NaOOC

NaOOCN

OONa

O

O

NNaO

O

N

O

O

ONa

NO

ONa

O

NN N

Zn

NN N

N N

COX

COX

COX

COONa

NaOOC

NaOOC

NaOOC

XOC

N

N

N

N

N

NMN

N

R1

R1

R1

R1

R2

R2

R2

R2

M

N

N

N

N

N

N

N

N

OPtO O

PtO

OPt O Pt

O

O

O O

O

O

OO O

O

OH2

H2O

OH2

OH2

OH2

OH2OH2

H2O

X = (CH3)3N+CH2[CH(OH)]4CH2OH;

RN(CH3)[(CH2)2O]3H

3637

R1 = CON(CH2COOH)2, R2 =H;

R1 = R2 = CON(CH2COOH)2;

38

M = AlOH, Zn, Co

39R1 = R2 = COOCH2CH2N+(CH3)2CH2CH2OH

Fig. 5. Water-soluble carboxyphthalocyanine derivatives useful for potential application in PDT and catalytic therapy of cancer



can be enhanced by cytotoxic properties of well-known platinum(II) fragments.

A very convenient method for selective preparation of the esters of phthalocyanine carboxylic acids starts from the esterifi cation of 4,5-dibromophthalic anhydride 40, which then can be introduced into the Rosenmund-von Braun reaction with the formation of nitrile 41. This nitrile can then be easily converted into respective transi-tion-metal phthalocyanines 42, some of which show dis-cotic liquid crystal properties typical of phthalocyanines (Scheme 14) [108, 111–114].

The use of chloromaleonitrile [115] for preparing methyl 2,3,4-trichloro-5,6-dicyanobenzoate is a rare example of the Diels-Alder reaction used for synthesis of precursors for phthalocyanines with electron-withdraw-ing groups (Scheme 15) [46]. A Diels-Alder reaction of commercially available 1,2,3,4-tetrachloro-5,5-dime-thoxycyclopenta-1,3-diene and chloromaleonitrile leads to formation of bicyclic intermediate, which can be aro-matized in two steps: in the fi rst, hydrogen chloride is eliminated; in the second, methyl chloride from a second unstable bicyclic reaction intermediate is eliminated.

N-substituted imides of 2,3,9,10,16,17,23,24-octa-carboxyphthalocyanines can be prepared using two major strategies. The fi rst utilizes transformation of respective octa-4,5-carboxyphthalocyanine anhydrides by reaction with an appropriate primary amino group. We used this methodology for preparing a large number of water-sol-uble anionic and cationic conjugates of phthalocyanines with α-, β-, and γ-aminoacids, di- and tripeptides, tau-rine, and β-diethylmethylammoniummethyl substituents (Scheme 16) [116–118]. In the case of anionic conjugates with α-, β-, and γ-aminoacids, di- and tripeptides, as well as taurine reaction was conducted in N-methylpyrrolidone

and directly results in the target compound 43, while in the case of cationic β-diethylmethylammoniummethylsubstituents the quaternary ammonium salt 44 can be obtained by an additional alkylation step using appropri-ate alkylating agent. The electronic absorption spectra of the quaternary salts indicated their considerable aggrega-tion in aqueous solution and can be controlled by adding polar aprotic solvents such as DMSO. These compounds were or are currently being tested as potential reagents for biological applications.

The derivatives of phthalonitrile-4,5-dicarboxamides can be prepared by similar method starting from 4,5-di-bromophthaloyl chloride using the standard Rosenmund-von Braun reaction (Scheme 17). The terminal ester groups in the target phthalocyanines 44 can be selectively hydrolyzed with formation of water-soluble terminal car-boxylic acids 45. The presence of eight substituents in phthalocyanines 44 improves their solubility compared to that observed for anionic imide analogs 42.

The water-soluble amides of phthalocyanine octacar-boxylic acid 45 are complementary to reported N-mono- and N-disubstituted amide derivatives of phthalocyanine tetra- and octacarboxylic acids 46 that are highly soluble in organic solvents (Fig. 6) [119–121].

An application of phthalocyanines in organic photo-voltaic cells and pH sensors requires the presence of spe-cifi cally designed peripheral substituents [122]. Some of such recent examples are the rhodamine-phthalocyanine conjugates, which have different optical and electron-transfer properties at different values of pH (Fig. 7) [123, 124]. An appropriate Rhodamine B, and 6G substituents can be linked to the phthalocyanine core via electron-withdrawing imide- or sulfonyl groups with the forma-tion of respective N-alkylimide or sulfamoyl bonds.

O

Br

Br

O

O

RO2C

RO2C

Br

Br N

N

N

N

N

NMN

N

CO2R

CO2R

CO2R

RO2C

RO2C

RO2C

RO2C CO2R

ROH

H+

CuCN

R = n-octyl, n-nonyl, n-dodecyl, 2-ethylhexyl, CF3(CF2)8, (CH2CH2O)nCH3

40 41

42

Scheme 14. Preparation of selected esters of transition-metal octacarboxyphthalocyanines

OMeMeO

Cl

ClCl

Cl

CN

CN

Cl

Cl

Cl

Cl

OMe

OMeCl

CN

CN

OMeMeO

Cl

ClCl

Cl

CN

CNH

ClCl

CN

CN

Cl

ClCOOMe

- MeCl- HCl+

Scheme 15. Synthetic pathway for preparation of methyl 2,3,4-trichloro-5,6-dicyanobenzoate



R = -CH2CO2H; -CH(CH3)CO2H; -CH(CO2H)(CH2)3CO2H;-CH(CO2H)CH2CONH2; -CH(CO2H)(CH2)2CO2H;-CH(CO2H)CH2CO2H; -(CH2)3CO2H; -CH2CONHCH2CO2H;-CH(CH2CH2CO2H)CONHCH(CH2SH)CONH(CH2CO2H);-CH2CH2SO3H; -CH2CH2SO2N(CH3)CH2CO2H

M = Cu, Co, Zn, AlOHM

N

N

N

N

N

N

N

N

NR

NRRN

RN

O

O

O

OO

O

O

O

N

N

N

N

N

N

N

N

M

O O

OO

O

O

O

O

O

OO

O

1. NH2R

NH2CH2CH2N(C2H5)2

M

N

N

N

N

N

N

N

N

NR

NRRN

RN

O

O

O

OO

O

O

O

M = Fe, Co, Zn, AlOH

R1X

M

N

N

N

N

N

N

N

N

N

NN

N

O

O

O

OO

O

O

ON+R N+

R

N+R N+R

R = CH3, X = I; R = C2H5, X = C2H5SO4; R = CH3, X = CH3SO4;R = CH3, X = CH3C6H4SO3; R = CH3, X = (CH3)2PO4

2. H+

42

R = -CH2CH2N(C2H5)2

4X-

43

Scheme 16. Preparation of anionic and cationic water-soluble phthalocyanineimides

Br

Br

O

O

Cl

Cl

Br

Br

O

NO

O

N

O

O

O

CuCN/DMF

N

N

N

N

N

NN

N

R

R

R

R

R

RR

R

M

R = -CON(CH3)CH2COOC2H5

R = -CON(CH3)CH2COOH

NH

O

O NC

NC

O

NO

O

N

O

O

O

N

N

N

N

N

NN

N

R

R

R

R

R

RR

R

M

M = Cu, Zn, Co

4445

MXn

hydrolysis

Scheme 17. Preparation of octaamido-substituted phthalocyanines



In neutral and alkaline media these compounds have electronic absorption spectra typical of phthalocya-nines and dominated by Q and B bands; the infl uence of closed form of rhodamine is negligible. In acidic condi-tions, however, the lactame ring of rhodamine substitu-ents transforms into open ionic form, resulting in new, characteristic for rhodamine band at 530 nm in addi-tion to phthalocyanine-based transitions. Taking into account bandwidths of these transitions, the whole range between 480 and 720 nm is covered by rhodamine and phthalocyanine absorption bands. Of course, interplay between open and closed forms of rhodamine substitu-ents depends on the nature of a specifi c rhodamine. For instance, the conjugates with rhodamine B 2-aminoethyl ester perchlorate preserve their spectral characteristics independently of acidity. Another very interesting obser-vation is that the luminescent properties of rhodamine conjugates are predominantly dictated by central metal atom in phthalocyanine core. For instance, the cationic metal-free rhodamine 6G phthalocyanine conjugate exhibits two equal intensity bands in emission spectrum, with wavelength close to absorption maxima of phthalo-cyanine and rhodamine 6G (Fig. 8). On the other hand, an emission spectrum of aluminum complex of rhodamine 6G phthalocyanine conjugate has only a single emission band close to absorption band of phthalocyanine ligand,

suggesting energy transfer from rhodamine substituents to phthalocyanine core.

Fluorinated substituents are one of the most powerful electron-withdrawing groups which can be introduced into phthalocyanine core. The fi rst tetra-, octa-, and dode-casubstituted by trifl uoromethyl groups phthalocyanine compounds (Fig. 9) were reported at the end of the 1970s by Oksengendler et al. [125–127]. The presence of trifl u-oromethyl groups in these complexes results in increase of their solubility in chloro- and nitrobenzene, as well as DMF, while solubility in common organic solvents remains quite low. Trifl uorosulfanyl and trifl uorosulfonyl substituted phthalocyanines can be prepared using the respective 4-trifl uoromethylsulfanyl-, 4-trifl uorometh-ylsulfonyl-, and 4,5-di(trifl uoromethylsulfanyl) phthalo-nitriles. Trifl uoromethylsulfanyl substituted phthalonitrileswere prepared starting from the reaction between 4-iodo- or 4,5-diiodo diethylphthalate and copper trifl uorometh-ylthiolates, followed by amidation reaction with liquid ammonia and fi nally by dehydration reaction with phos-phorus oxychloride. Oxidation of trifl uoromethylsulfanyl group in the abovementioned compounds leads to forma-tion of the 4-trifl uoromethylsulfonyl- and 4,5-di(trifl uoro-methylsulfonyl) diethyl phthalates, of which only the former can be amidated with liquid ammonia and even-tually converted into substituted phthalocyanine. Again,

N

N

N

N

N

N

N

N

R1

R2 R1

R2

R1

R2R1

R2

M

46

R1 = -CONH(t-Bu); -CONEt2 ; R2 = HR1 = R2 = -CONEt2; -CONBu2

Fig. 6. Derivatives of tetra- and octasubstituted phthalocyanine amides that highly soluble in organic solvents

O NHEtNHEt

NNH

SO

O

O

N N

N

N

NN

N

N Zn

R

R

R

R

OHNEt NHEt

O

HNNH

SO

O

Cl-+

R (open form) =

R (closed form) =

Fig. 7. The open and closed forms of conjugate of tetra-4-sulfo substituted zinc phthalocyanine with spirolactame of N-(2-amino-ethyl)rhodamine 6G



the solubility of trifl uoromethylsulfanyl- and trifl uorom-ethylsulfonyl substituted phthalocyanines in common organic solvents remains low. Interestingly, introduction of any of the fl uorinated substituents presented in Fig. 9 into phthalocyanine core has no signifi cant effect on the position of Q band in UV-vis spectra of these complexes compared to unsubstituted phthalocyanine analogs although a small bathochromic effect was observed for both types of substituents.

The solubility of fl uoroalkyl complexes can be sig-nifi cantly increased by replacing trifl uoromethyl sub-stituents with fl uoroalkoxy one (Scheme 18) [128]. The

preparation of target nitriles 47 requires nucleophilic aromatic substitution of nitro group in commercially available 4-nitrophthalonitrile. Since the nucleophilicity of fl uorinated alcohols is very small, it is necessary to transform these into respective alkoxides using appro-priate base prior substitution conducted at elevated tem-peratures in aprotic polar solvents. Phthalocyanines 48prepared from these nitriles are soluble in a variety of organic solvents and have optical properties similar to those of tetra-tert-butylphthalocyanines. The electron-withdrawing nature of fl uoroalkoxy groups can be clearly seen from the fi rst oxidation potential of respective phtha-locyanines 48. For instance, an oxidation potential of zinc phthalocyanine substituted with four –OCH2CF3 groups is ~200 mV higher compared to that observed in zinc 2,9(10),16(17),23(24)-tetra-tert-butylphthalo cyanine [129].

An esterifi cation of 4-chlorosulfonylphthalonitrile with fl uoroalkyl-containing alcohols leads to formation of polyfl uoroalkoxysulfonyl phthalonitriles 49, which can be converted into appropriate 2,9(10),16(17), 23(24)-tetra(fl uoroalkylsulfonyl)phthalocyanines 50 (Scheme 19) [130]. These complexes are also highly sol-uble in a variety of organic solvents, which makes them potential candidates for application in homogeneous catalytic reactions. The polyfl uoroalkoxysulfonyl groups

N

N

N

N

N

N

N

NM

R1

R2

R3 R4 R1R2

R3

R4

R1

R2

R3R4R1R2

R3

R4

M = 2H, Cu

R1 = CF3, R2-R4 = HR2 = CF3, R1 = R3 = R4 = HR1 = R4 = CF3, R2 = R3 = HR1 = R4 = H, R2 = R3 = CF3R1 = R2 = R3 = CF3, R4 = HR1 = SCF3, R2-R4 = HR1 = R3 = R4 = H, R2 = SCF3R1 = R4 = H, R2 = R3 = SCF3R1 = R3 = R4 = H, R2 = SO2CF3

Fig. 9. Trifl uoromethyl-, trifl uoromethylsulfanyl-, and trifl uo-romethylsulfonyl substituted phthalocyanines

N

N

N

N

N

NMN

N

OR

OR

OR

OR

O2N

CN

CN

RONa or ROH

HMPA, DMSO RO

CN

CN

R = CF3CH2, H(CF2)2CH2, H(CF2)4CH2, H(CF2)6CH2, (CF3)2CH, (CF3)2CF, (CF3)2CCl, (CF3)2COH, (CF3)3CM = Cu, Co

47

48

Scheme 18. Preparation of fl uoroalkoxy-substituted phthalocyanines

Fig. 8. Room-temperature emission spectra of metal-free (1) and aluminum chloride (2) phthalocyanine conjugates with rhodamine 6G in ethanol-chloroform mixture



in these complexes were shown to stabilize both HOMO and LUMO of the complexes in comparison with alkoxy- or polyfl uoroalkoxy- substituted analogs. Although never observed (because of the solvent electrochemical win-dow limitations) the fi rst oxidation potential for zinc 2,9(10),16(17),23(24)-tetra(fl uoro alkyl)sulfonylphthalo-cyanines was estimated at between 1.47 and 1.65 V.

Phthalocyanines with electron-donating groups

The majority of methods for the preparation of phtha-locyanines with electron-donating groups require useful synthetic pathways for the preparation of respective phtha-lonitriles bearing electron-donating groups [13, 20]. One of the most convenient methods which results in formation of target phthalonitriles with electron donating-substitu-ents utilizes nucleophilic aromatic substitution reaction with nitro- or bromide-leaving group proven to be the best in this reaction. Aromatic nucleophilic substitution of the bromine atom in 4-bromophthalonitrile with second-ary amines was discussed in the mid-1970s (Scheme 20) [131–133]. Similarly, nucleophilic aromatic substitution of bromine in the 3-bromo-5-tert-butylphthalonitrile leads to formation of 3-(N,N-dimethylamino)-5-tert-butylphthalo-nitrile [132]. Alternatively, 3- and 4-(N,N-dimethylamino)-phthalonitrile can be prepared by methylation of easily available 3- and 4-aminophthalonitrile with dimethylsulfate in aprotic polar solvents. Similar results can be obtained for the nucleophilic substitution reactions of 3- or 4-nitrop-hthalonitrile with dialkylamino precursors in polar aprotic solvents (Scheme 20). Substituted phthalonitriles obtained this way can be easily introduced into reactions with appro-priate metal salts to form respective tetrasubstituted phtha-locyanines with electron-donor groups which are soluble in various organic solvents.

Although solubility of parent 1,8(11),15(18),22(25)- and 2,9(10),16(17),23(24)-tetraaminophthalocyanines is limited to polar aprotic and halogenated aromatic solvents, these macrocycles have received much atten-tion because they are potentially useful for preparation of chemically modifi ed electrodes for electrocatalytic reactions and also are good candidates for immobiliza-tion of chemically modifi ed surfaces [134–136]. Typi-cally, tetraaminophthalocyanines can be prepared by hydrolysis of appropriately protected amino groups (i.e. acetylamide) or by reduction of nitro group in tetra-substituted phthalocyanines [137–139]. In the mid-1980s, we did report, however, an attempt for direct tetrameriza-tion of 3- and 4-aminophthalonitriles or corresponding phthalimides into respective transition-metal phthalo-cyanines (Scheme 21) [140]. It has been found that 3-aminophthalo nitrile can indeed, upon heating with respective metal salt with or without urea, form tar-get transition-metal 1,8(11),15(18),22(25)-tetraamino-phthalocyanines. Reaction of 4-aminophthalonitrile with transition-metal salts produces uncharacterized black polymers, while similar reaction of 4-aminophthalo-nitrile in the presence of urea results in the formation of 2,9(10),16(17),23(24)-tetraureidophthalocyanines. It seems that the difference in reactivity of 3- and 4-amino-phthalonitriles can be explained on the basis of steric restrictions observed in the case of 3-aminophthalo-nitriles, which cannot form ureido derivatives as well as cannot easily participate in the intermolecular nucleo-philic attack of nitrile group. The ureido derivatives of phthalocyanines are stable in diluted alkaline solutions but partially hydrolyzed in concentrated acids. Complete hydrolysis of these complexes with alkali metal hydrox-ides in triethylene glycol results in formation of the tetrac-arbamic acid derivatives, which are stable in solution but could be thermally decomposed with elimination of CO2.

It is interesting to compare the reactivity of the nitro groups in 3,5-dinitrophthalonitriles toward nucleophilic aromatic substitution reactions with dialkylamines as well as other nucleophiles [141]. Reaction of the 3,5-dinitrophthalonitrile with piperidine in DMF in the pres-ence of potassium carbonate as a base leads to formation of all three expected nucleophilic substitution products

CN

Lv CN

R2NH

DMF

CN

CNR2N

HN HN OLv = Br or NO2 R2NH = (CH3)2NH,

Scheme 20. Synthetic pathway for preparation of dialkylamino substituted phthalonitriles

CN

CN

ClO2S

ROHNEt3

CH2Cl2

CN

CN

ROO2S N N

N

N

NN

N

N M

SO2OR

SO2OR

ROO2S

ROO2S

+

M = Co, Zn

R = CH2CF3, CH2CF2CF2H, CH2(CF2CF2)2H, CH2(CF2CF2)3H, CH(CF3)2

49

50

Scheme 19. Synthetic pathway for the preparation of polyfl uoroalkoxysulfonyl phthalocyanines



with one or two piperidine residues (Scheme 22). The monosubstituted 3(5)-nitro-5(3)piperidinatophthalonitriles are interesting examples of precursors, which lead to prep-aration of octasubstituted phthalocyanines with push-pull substituents in each isoindole subunit. It was shown that an unreacted nitro group in 3(5)-nitro-5(3)-piperidinato-phthalonitriles can be further substituted with a different nucleophile, i.e. phenylthiolate with formation of 3-pipe-ridino-5-phenylsulfanylphthalonitrile (Scheme 22), which is a useful precursor for the preparation of 1,8(11),15(18),22(25)-tetrapiperidinato-3,10(9),17(16),24(23)-tetra-phenyl sulfanylphthalocyanines.

Nucleophilic aromatic substitution of the chlorine atoms in tetrachlorophthalonitrile by phenols, thiols, and amines is relatively new but very promising method of introducing electron-donating groups into phthalocya-nine core [142–145]. It was found that the selectivity and degree of substitution predominantly depends on the nature of nucleophile. For instance, it is possible to

substitute all chlorine atoms in tetrachlo-rophthalonitrile with alkyl- and aryl thiols, while only three chlorine atoms can be sub-stituted by aryloxy and pyridyloxy groups. Moreover, only two chlorine atoms can be substituted by cyclic secondary amines, while only one chlorine atom can be replaced by monoalkyl-, monoaryl-, and dialkylamine nucleophiles (Scheme 23). The fi rst point of substitution is indepen-dent of the type of nucleophile and always leads to the formation of 2,4,6-trichloro-5-nucleophilophthalonitrile. When nucleo-philes are aryl and alkyl thiols, phenols and cyclic secondary amines, an introduction of the second group results in the formation of 3,6-dichloro-4,5-dinucleophilophtha-lonitrile, while in the case of pyridin-3-ol as nucleophile both 3,6-dichloro-4,5-di-(3-pyridyloxy)oxyphthalonitrile and 3,5-dichloro-4,6-di(3-pyridyloxy)oxyphtha-

lonitrile were observed in the reaction mixture. Finally, when three chlorine atoms are substituted by nucleo-philes, only 3-chloro-4,5,6-trinucleophilophthalonitrile was observed in the reaction mixture.

Similar nucleophilic substitution in polychlorophtha-lonitriles already substituted by electron-donor groups can result in substituted phthalonitriles with two differ-ent electron-donating groups [142]. For instance, room-temperature reaction of 3,4,6-trichloro-5-(phenylamino)phthalonitrile with one, two, or three equivalents of phenyl or butyl thiols results in formation of pure 3,6-dichloro-4-(aryl- or alkylsulfanyl)-5-(phenylamino)phthalonitrile, 3-chloro-5,6-di(aryl- or alkylsulfanyl)-4-(phenylamino)phthalonitrile, and 3,4,6-tri(aryl- or alkylsulfanyl)-5-(phenylamino)phthalonitrile, respectively. Another example of such a clean transformation is the reaction of 3,6-di-chloro-4,5-di(piperidin-1-yl)benzene-1,2-dinitrile with phenylthiol, which exclusively leads to formation of 3,6-di(phenylsulfanyl)-4,5-di(piperidin-1-yl)benzene-1,2-dinitrile. If an electron-donating group already present

R1 R2

pip NO2

NO2 pippip pip

PhS PhS

pip PhS

NO2

O2N

CN

CN

PhSNa or C5H10NH

K2CO3, DMF

R1

R2

CN

CN

Cu(OAc)2

N

N

N

N

N

N

N

N

R1

R2

R2

R1

R1

R2

R1

R2

Cu

220-230oC

Scheme 22. Aromatic nucleophilic substitution reactions in 3,5-dinitrophthalonitrile

CN

CNH2N

CN

CN

NH2

N

N

N

N

N

NMN

N

NHCONH2

NHCONH2

HNOCH2N

HNOCH2N

N

N

N

N

N

NMN

N

NH2

H2N

H2N

NH2

NH2CONH2

NH2CONH2

MXn

MXn

Scheme 21. Preparation of amino- and ureido substituted transition-metal phthalocyanines



in the substituted polychlorophthalonitrile is 3-pyridy-loxy or phenylsulfanyl one, an additional nucleophilic substitution leads to a variety of products which origi-nate from scrambling reaction and thus are less use-ful from the synthetic point of view. 3,6-dichloro-4,5-di(phenyl-or alkyl sulfanyl)benzene-1,2-dinitriles, 3,4,5,6-tetra(phenylsulfanyl)benzene-1,2-dinitrile, 3,6-dichloro-4,5-di(aryloxy)benzene-1,2-dinitriles, 3,5-dichloro-4,6-di(3-pyridyloxy)benzene-1,2-dinitrile, 3,4,6-trichloro-5-(3-pyridyloxy)benzene-1,2-dinitrile, and 3,4,6-trichloro-5-aminobenzene-1,2-dinitrile can be readily tetramerized into corresponding transition-metal or main group phtha-locyanines, which in a majority of cases are only slightly soluble in aprotic dipolar solvents. On the other hand, purifi cation of transition-metals 2,3,9,10,16,17,23,24-octa(substituted amine)-1,4-8,11,15,18,22,25-octachlo-rophthalocyanines was not successful because of the low stability of target phthalocyanine complexes.

Aromatic nucleophilic substitution of nitro groups in 3- and 4-mono, as well as 3,5-disubstituted phtha-lonitriles by phenols and thiophenols was reported in 1980 [146–149]. This reaction, which has become a standard method for introducing ArO and ArS substitu-ents into phthalonitriles and respective phthalocyanine complexes, proceeds smoothly in aprotic polar solvents in the presence of large excess of potassium carbonate

as a base, affording phthalonitrile precursors in high yields (Scheme 24) [146, 150]. The Q band in phe-nylsulfanyl derivatives located at longer wavelengths compared to corresponding phenoxy complexes. 1,8(11),15(18),22(25)-tetraphenylthio phthalocyanines have Q band in NIR region and are potential candidates for PDT application. Indeed, the liposomal form of one of these complexes (so-called “thiosens”) is currently undergoing preclinical trials as a PDT photosensitizer. Similar to het-erobimetallic complex 39, the electron-donating pyridy-loxy- substituents in zinc 2,9(10),16(17),23(24)-tetrakis(3-pyridyloxy)phthalocyanine can be used as anchor groups in preparation of tetraplatinum(II) derivatives with PtCl2NH3 and PtCl2(dmso) fragments connected to the phthalocyanine core via 3-pyridyloxy substituents [151, 152]. The platinum fragments in these heterobimetallic complexes are stable in solution up to 90 °C, which makes them excellent potential candidates for combined PDT/chemotherapy treatments. Another interesting feature of transition-metal 1,8(11),15(18),22(25)-tetramethoxy-phthalocyanines is the prominent basicity of meso- nitrogen atoms. For instance, meso-nitrogen atoms in zinc 1,8(11),15(18),22(25)-tetramethoxyphthalocyanine can be protonated even by water compared to basicity of unsubstituted analog, which can be protonated by only strong acids [153].

CN

CNO2NN N

N

NNN

N

N M

R1

R1

R1

R1

R2

R2

R2

R2

R3

R3

R3

R3

CN

CN

NO2

R3

CN

CNPhX

CN

CN

XPh

R3

PhOH or PhSH

K2CO3/DMF

PhOH or PhSH

K2CO3/DMF

X = O, S R1 = OPh, SPh; R2 = H; R3 = H or tert-BuR1 = R3 = H; R2 = OPh, SPh

Scheme 24. Examples of aromatic nucleophilic substitution of nitro group in 3- and 4-phthalonitriles by phenols and thiophenols

ClCl

ClCl

CN

CNCl

Cl

RCl

CN

CN

ClR

RCl

CN

CNR

Cl

RCl

CN

CNR

R

RR

CN

CNR

R

RCl

CN

CN

DMF+ + + +

Nu

Nu = PhSH:tert-BuSHn-BuSH:n-C10H21SH:PhONa:3-PyONa:PhNH2:n-C8H17NH2:Et2NH:Bu2NH:piperidine:morfoline:

XXXXXXXXXXXX

XXXXXX----XX

----XX------

XXXXXX------

XXXX--------

Scheme 23. Nucleophilic substitution reactions of tetrachlorophthalonitrile



The other useful precursors for preparation of dial-kylamino substituted phthalocyanines are 3- and 4-dialky-mamino substituted phthalimides, which can be easily prepared in high yield by aromatic nucleophilic substitu-tion of bromine or nitro groups in 3- and 4-substituted phthalimides [154]. The Rosenmund-von Braun reaction is a very simple reaction pathway for preparing substi-tuted by electron-donating groups phthalonitriles which was successfully exploited from the mid-1970s (Fig. 10) [155, 156]. The reaction yields depend on the number and type of substituents present in dibromo presursors and it seems that in general, preparation of substituted phthalonitriles bearing electron-donating groups is more challenging compared to those substituted with electron-withdrawing groups.

Chloromethylation of the phthalocyanine core results in formation of tetra- and octachloromethylphthalocya-nine macrocycle as a mixture of unknown positional iso-mers. These chloromethylphthalocyanines can be used as the universal precursors for preparation of a variety of functionalized phthalocyanine derivatives. For instance, phosphonate groups can be introduced into phthalocya-nine core using tetrachloromethylphthalocyanine deriva-tives and trialkylphosphite followed (if necessary) by hydrolysis (Scheme 25) [157, 158]. Similarly, eight phos-phonate groups can be introduced into phthalocyanine core using octachloromethylphthalocyanine precursors. Sodium salts of anionic phosphomethyl Pc derivatives absorb at longer wavelengths than sulfo phthalocyanines, making them more suitable photosensitizers for PDT to treat deeper located tumors.

The direct introduction of carbon-phosphorous bond into phthalocyanine core can be achieved either by direct

tetramerization of 4-phosphonophthalic acid in the pres-ence of transition-metal salt and urea or by interaction of unsubstituted aluminum or gallium phthalocyanine with dialkyl phosphonites followed by basic hydrolysis [159, 160]. Obtained in this manner tetra(phosphono)phthalo-cyanine salts are soluble in water and are potentially use-ful candidates for PDT.

Similarly to the preparation of anionic phosphomethyl phthalocyanines, water-soluble cationic phthalocyanines can be prepared using octachloromethyl phthalocyanine precursor (Scheme 26). In this case, the reaction of ter-tiary aliphatic amine or pyridine leads to formation of quaternary ammonium or pyridinium salt which, when other tertiary amino groups are present, can be trans-formed into dicationic (per substituent) derivative [161]. One of these derivatives, so-called “cholosens” is very active in cancer PDT as well as in antimicrobial photody-namic treatment of biologically relevant substances and water [162–164]. It is currently undergoing preclinical tests for application as an antibacterial compound.

Use of tetra- and octachloromethyl phthalocyanine precursors also provides a possibility for introducing carborane cages useful in boron neutron capture ther-apy. Covalent conjugate of cobalt phthalocyanine with carborane closocarboranyl phthalocyanine was pre-pared earlier starting from dimethyl(3,4-dicyanophenyl)(o-carboranyl methyl)malonate [165] and other [166–169] precursors. We have demonstrated that tetrakis and octakis (chloromethyl)phthalocyanines when treated with B12H11NH2

2- result in sodium salts of tetrakis and octakis(undecahydro-closododecaboranylaminomethyl) phthalocyanines which are soluble in water because of the presence of anionic boranyl moieties [170].

N

N

N

N

N

N

N

N

M

CH2Cl

CH2ClClH2C

ClH2C

1. (AlkO)3P

NN

N

N

N

NN

NM

CH2PO(ONa)2

CH2PO(ONa)2(NaO)2OPH2C

(NaO)2OPH2C

2. NaOH

Scheme 25. Preparation of phosphomethyl phthalocyanine derivatives using tetrachloromethyl phthalocyanine precursor

Me2N

CN

CN t-Bu

CN

CN

NMe2

H2N

CN

CN

O2N

Me2N

CN

CN

Me2N

MeO

CN

CN

CN

CNOMe

CN

CN

O

O

CN

CN

O

O

Fig. 10. Examples of substituted with electron-donating groups phthalonitriles prepared using Rosenmund-von Braun reaction



Peripherally substituted naphthalocyanines and larger annulated analogues

Unsubstituted 2,3-naphthalocyanine complexes were fi rst reported at the end of the 1930s; their stability and optical properties were investigated at the end of the 1960s [171, 172]. Since it was found that the solubility of unsubstituted metal-free and transition-metal 2,3-naph-thalocyanines is very low in common organic solvents, the early efforts on substituted 2,3-naphthalocyanines were focused on the preparation of soluble 2,3-naphthalocy-anine derivatives [173]. One of the most useful methods for preparing 6-substituted 2,3-dicyanonaphthalenes was explored at the beginning of 70s and used for the synthesis

of 6-tert-butyl-2,3-dicyanonaphthalene (Scheme 27) [174]. The reaction pathway starts from preparation of 4-alkyl-ortho-xylene 51, which can be further brominated by radical mechanism using N-bromosuccinimide to result in formation of 4-tert-butyl-1,2-bis(dibromomethyl)-benzene 52 in high yield. This compound as well as many other substituted 1,2-bis(dibromomethyl)benzenes (for instance 4-adamantyl-1,2-bis(di bromomethyl)benzene) which can be prepared in a similar way are extremely useful precursors for the Diels-Alder reaction between 4-tert-butyl-1,2-bis(bromomethylene)cyclohexadiene-3,5 53 generated from them in situ and appropriate diene (i.e. fumaronitrile 54). The reaction intermediate 55 can

CHBr2

CHBr2

CHBr

CHBr

BrCN

CNBr

CN

CN

N

N

N

N

N

N

N

NM

NC

CNNBS NaI

DMF

51 52 53

54

56

55

57

hν

-2HBr

MXn

Scheme 27. Synthetic pathway for preparation of tetra-tert-butyl-2,3-naphthalocyanines

Scheme 26. Preparation of water-soluble cationic phthalocyanines using octachloromethyl phthalocyanine precursors

amine

Cl-N+

NN N

N

NNN

N M

Cl

Cl

Cl

Cl

ClCl

ClCl

NN N

N

NNN

N M

R

R

R

R

RR

RR

NN N

N

NNN

N M

Et3N, AlCl3

M = AlCl, ZnR =

(ClCH2)2O

MeI

*

N+

OH

Cl-N+

NN+

N+

Cl-, I-

N+

OH

OHOH

*

*

*

*

Cl-



be easily aromatized by elimination of two hydrogen bromide molecules with the formation of 6-tert-butyl-2,3-dicyanonaphthalene 56, which under regular template high-temperature reaction generates target tetra-tert-butyl-2,3-naphthalocyanines 57 [129, 174–176].

Similarly to the strategy presented in Scheme 27, vari-ous 6-mono- and 6,7-disubstituted-2,3-naphthalonitriles have been prepared (Scheme 28) [38, 40, 129, 177–179]. This strategy can also be used in the case when 1,2-dicyano precursor is located on the “diene” com-ponent of the Diels-Alder reaction, while dienophile component provides substituents at 6 and 7 positions of 2,3-dicyanonaphthalene ring (Scheme 29) [180–183]. The last approach is very useful for preparing 2,3-naph-thalocyanine octa-6,7-carboxylic acid derivatives. As expected, an introduction of chlorine, bromine, nitro, cyano, and trifl uoromethyl groups into 2,3-naphthalocy-anine core does not improve their solubility in common

organic solvents. More interestingly, the best solvent for tetra(perfl uoro)-tert-butyl-2,3-naphthalocyanines is concentrated sulfuric acid; their solubility in common organic solvents remains very low [184].

Although in general substituted 2,3-naphthalocyanine complexes have lower stabilities compared to similarly substituted phthalocyanines in solution, it is possible to hydrolyze terminal cyano or ester groups in octacy-ano- and octa(N-carboxymethylcarbamoyl)-2,3-naph-thalocyanine to the respective terminal octacarboxylic acids [185]. 1-Substituted 2,3-dicyanonaphthalenes and respective tetrasubstituted 2,3-naphthalocyanines can be prepared in two ways. The fi rst method is based on the Diels-Alder reaction between 1,2-bis(dibromomethyl)benzene and chloromaleonitrile, which leads to for-mation of 2,3-dicyanonaphthalene and 1-bromo-2,3-dicyanonaphthalene (Scheme 30) [185]. The later can not only be used for preparation of tetrabromo 2,3-naphthalocyanines but also can be introduced in aro-matic nucleophilic substitution reaction with secondary amines, which results in formation of 1-dialkylamino-2,3-naphthalenes. These nitriles can produce tetrasub-stituted by dialkylamino groups 2,3-naphthalocyanines, which have a very low energy Q band in UV-vis spectra [35]. A second method for preparation of 1-substituted 2,3-dicyanonaphthalonitriles and respective 2,3-naph-thalocyanine derivatives is based on a high-temperature catalytic transformation of anhydride of 1-phenyl-2,3-naphthalic acid or respective imide into 1-phenyl-2,3-dicyanonaphthalene [186].

CHBr2

CHBr2

NC

NC

CN

CN

R

R

RCH=CHR

O

O

O

N

O

O

R = -COOH; -CN; -CONHCH2COOC2H5; ;

Scheme 29. Use of 4,5-bis(dibromomethyl)benzene-1,2- dinitrile for preparation of 6,7-disubstituted derivatives of 2,3-dicyanonaphthalene

CHBr2

CHBr2

CN

CNCl

CN

CN

CN

CN

Br

CN

CN

NR2

CN

CN

Br

DMF

HNR2

+ +

Scheme 30. Preparation of 1-bromo- and 1-dialkylamino-2,3-dicyanonaphthalenes

CN

CN

Br

Br

R1

R2

CHBr

CHBr

R1

R2

CN

CN

R1

R2

CHBr2

CHBr2

R1

R2

CN

NCNaI

DMF

O

O

O

N

O

O

- 2 HBr

R1 = R2 = H, Cl, Br, CN, , , ,

R1 = H; R2 = tert-Bu, NO2, CN, -CO2R, -CF3, -C(CF3)3

Scheme 28. Synthetic pathways for preparation of 6-mono and 6,7-disubstituted 2,3-dicyanonaphthalenes



5-Substituted 2,3-dicyanonaphthalenes can be pre-pared from parent 2,3-dicyanonaphthalene using aromatic electrophilic nitration reaction as a fi rst step (Scheme 31) [187]. The resultant 5-nitro-2,3-dicyanonaphthalene can be reduced to 5-amino-2,3-dicyanonaphthalene, which can be further converted into 5-bromo-2,3-dicyanon-aphthalene using diazotization reaction or acylated by acetyl chloride with formation of 5-acetylamide-2,3-dicyanonaphthalene.

Commercially available 1,3-diphenylbenzofuran is another useful precursor for preparation of 1,4-diphe-nyl-2,3-dicyanonaphthalene using the Diels-Alder reac-tion with fumaronitrile following elimination of water molecule from bicyclic intermediate in acidic condi-tions [188]. Easily available substituted benzophenones

were also found to be an excellent starting material for preparation of 1,6-disubstituted 2,3-dicyanonaphthalenes (Scheme 32) [189]. Reaction sequence starts from radical bromination of methyl group in appropriate benzophe-none derivative, which follows intramolecular cycliza-tion with formation of 1,6-disubstituted benzofuran. This furan can be introduced into the Diels-Alder reaction with respective dienophile to form bicyclic intermediate, which can be easily aromatized into 1,6-disubstituted derivatives of naphthalene 2,3-dicarboxylic acid.

Diels-Alder reaction again is very useful for prepara-tion of 5,6,7,8-tetrasubstituted 2,3-dicyanonaphthalenes and respective substituted 2,3-naphthalocyanines, with two examples shown in Scheme 33 [180]. One interest-ing feature of these reactions is the high-temperature

Scheme 32. Synthetic pathway for preparation of 1,6-disubstituted derivatives of naphthalene 2,3-dicarboxylic acid

CH3

R1 R2

O

CH2Br

R1 R2

OR1

OX

X

R2R2

R1

X

X- H2OR1

O

R2

H+

O

O

O

HN

O

O

CHX = CHXBr2 or NBS

CCl4, hν

R1 = R2 = tert-Bu, Ph; X = -CN; ;

- HBr

Scheme 33. Synthetic pathways for preparation of 5,6,7,8-tetrasubstituted 2,3-dicyanonaphthalenes

Cl

Cl

ClCl

CH2Br

CH2Br NC

CN

RCN

CN

R

RR

RCN

CN

R

R

RPhPh

PhPh

OCN

CN

-CO

-H2

C6H3Cl3

+

+

R = Cl, Ph

reflux

NaI, DMF

CN

CNNO2

CN

CNNH2

CN

CN

NHCOCH3

CN

CN

AcCl

CN

CNBr

KNO3

H2SO4

H+

SnCl2

94% 84 %

77 %

1 HNO2, HBr

2 CuBr, HBr

28 %

Scheme 31. Preparation of 5-substituted 2,3-dicyanonaphthalenes



aromatization of tetrahydro intermediates, which can be conducted in refl uxing trichlorobenzene.

Unlike linear 2,3-naphthalocyanines, macrocycl-ization reactions leading to formation of angular 1,2-naphthalocyanines result in four possible geometrical isomers, presented in Fig. 11 [190–194]. The key pre-cursor for preparation of 1,2-naphthalocyanines is 1,2-dicyanonaphthalene, which can be prepared in two steps: fi rst, by transformation of 2-amino-1-naphthalene sulfo-nic acid into 2-cyano-1-naphthalene sulfonic acid using Sandmeyer reaction, followed by cyanodesulfonation reaction with potassium ferrocyanide [194, 195]. All four pure isomers of magnesium or lutetium 1,2-naphthalocy-anines can be separated using chromatography methods [192, 193]. Spectroscopic data collected for individual isomers show that, as expected, the degree of steric inter-actions increases in the following order: D2h > C2v ~ Cs > C4h. The individual isomers mentioned above were used for preparation of unsymmetric double-decker (1,2-Nc)Lu(Pc) complexes, which were characterized by various spectroscopic techniques.

Preparation of a large variety of ortho-dicyano aro-matic compounds with extended π-system (Fig. 12) and respective complexes of phthalocyanine analogues was explored in detail in the 1970s [43, 196]. Nitriles 58 and 59 can be prepared in high yield using relatively

easy oxidative photodehydrogenation reactions of 4,5-diphenylphthalonitrile and 1,2-(p-tert-butylphenyl)male-onitrile, respectively [197], while the initial synthetic routes for the preparation of 9,10-bis(4-tert-butylphenyl)anthracene-2,3-dicarbonitrile 60 [198] and 5,11-bis(4-tert-butylphenyl)-2,3-dimethyl-6,12-diphenyl tetracene 61 [199] were quite complicated. For instance, synthesis of 61 starts from condensation of p-tert-butylphenylacet-ylene with benzophenone, which results in formation of 3-(4-tert-butylphenyl)-1,1-diphenylprop-2-yne-1-ol. This compound can be cyclized into 5,11-bis(4-tert-butylphenyl)-6,12-diphenyltetracene upon treatment with thionyl chloride in 35% yield. Diels-Alder reaction of 5,11-bis(4-tert-butylphenyl)-6,12-diphenyltetracene with dicyanoacetylene leads to formation of bicyclic interme-diate, the -CH=CH- bridge of which can be fi rst reduced by hydrogen and then -CH2-CH2- fragment eliminated at elevated temperature (320–330 °C) to form target 5,11-bis(4-tert-butylphenyl)-6,12-diphenyltetracene-2,3-di-carbonitrile 61.

An alternative synthetic pathway for preparation of 9,10-diphenyl- and 1-methyl-9,10-diphenyl-2,3-dicyanoanthra-cenes was explored recently which involves the Diels-Alder reactions between 1,3-diphenylisobenzofuran and appro-priate dienophile, followed by aromatization step with the yields of target nitriles ca. 55% (Scheme 34) [200].

CN

CN

CN

CN

CN

CN

CN

CN

CN

CN

58 59

60 61

62

Fig. 12. Examples of ortho-dinitriles with extended π-system prepared in 70s

N N

N

N

NN

N

N M

N N

N

N

NN

N

N M

N N

N

N

NN

N

N M

N N

N

N

NN

N

N M

C4h D2hC2v Cs

Fig. 11. Individual isomers of 1,2-naphthalocyanines



Peripherally substituted nitrogen analogues of phthalo-cyanines

Although fi rst 2,3- and 3,4-tetrapyridoporphyrazines were reported by Linstead in the late 1930s [201, 202], an extensive study of preparation of unsubstituted as well as alkyl substituted derivatives of 2,3- and 3,4-pyridin-edicarbonitrile and investigation of optical properties of respective metal-free and transition-metal pyridinopo-rphyrazines were conducted only in the early 1970s [173]. The majority of the initial nitriles were prepared by the classic “acidic” method used earlier for prepa-ration of substituted phthalonitriles, i.e. transformingunsubstituted or alkyl substituted pyridine 2,3- and pyri-dine 3,4-dicarboxylic acid into respective anhydrides, imides, diamides, and fi nally dinitriles [173, 203, 204]. An interesting reaction pathway for preparation of 5,6-dimethypyridine-3,4-dicarbonitrile and 2,5,6-trim-ethylpyridine-3,4-dicarbonitrile, which was discovered in 1970 utilizes the Diels-Alder reaction between substituted oxazoles and fumaronitrile, which results in formation of bicyclic intermediate [205]. This bicyclic intermediate can be easily aromatized by elimination of water mol-ecule into 5,6-dimethylpyridine-3,4-dicarbonitrile and 2,5,6-trimethylpyridine-3,4-dicarbonitrile (Scheme 35). Heating of 5,6-dimethypyridine-3,4-dicarbonitrile and VCl3 in quinoline 8 h at 150–165 °C results in forma-tion of vanadyl complex of tetrakis(5,6-dimethyl-3,4-pyridino)porphyrazine in 52% yield. Interestingly, that similar reaction with more sterically crowded 2,5,6-trim-ethylpyridine-3,4-dicarbonitrile does not result in forma-tion of corresponding porphyrazine.

The most common strategy for preparation of alkyl- and arylsubstituted 2,3-dicyanopyrazines and respective pyrazinoporphyrazines includes the condensation reac-tion of commercially available diaminomaleodinitrile and α-dicarbonyl component in acetic acid or acetic acid-etanol mixture as a solvent (Scheme 36) [201, 202, 206, 207].

Similar condensation reactions can also be used for preparation of linearly and angularly annulated deriva-tives of quinoxaline-2,3-dicarbonitrile [208, 209]. In this case, the appropriate aromatic diamino component can be condensed with methyl or ethyl esters of 2,3-dioxosuccinic acid. Formed in this way methyl or ethyl esters of quinoxaline-2,3-dicarboxylic acid (Fig. 13) can be transformed into respective dinitriles using usual “acidic” synthetic pathway (ester to diamide to dinitrile) [210].

Peripherally substituted tetraazaporphyrines

There are eight peripheral positions in tetraazapor-phyrin which can be occupied by various substituents (Fig. 14). Depending on the type of substituents, octasu-bstituted tetraazaporphyrins can be symmetric (R1 = R2)or consist of a mixture of four possible positional isomers (R1 ≠ R2).

Synthetic pathway for preparation of octaaryl sub-stituted tetraazaporphyrins was fi rst developed by Lin-stead in 1937 and remains the most useful strategy [211]. The reaction starts from preparation of aryl substituted acetonitrile, which can be coupled sequentially with iodine and sodium methoxide and leads to formation of target 1,2-diarylmaleo(fumaro)nitrile in moderate to low yields (Scheme 37). These nitriles can be further tetramerized into tetraazaporphyrins using magnesium alkoxides or transition-metal salts with relatively high yields despite the presence of both Z and E isomers of nitriles in the reaction mixture [212–216].

N

N CN

CNO

O

H2N

H2N CN

CN

HOAc

EtOH+

Scheme 36. Typical strategy for preparation of substituted 2,3-dicyanopyrazine

N

CN

CNN

O NC

CN

NCN

CN

O-H2O

+

Scheme 35. Synthetic pathway for preparation of 2,5,6-trimethylpyridine-3,4-dicarbonitrile

O

Ph

Ph

CN

CN

Ph

Ph

CN

CN

Ph

Ph

CN

CN

Ph

Ph

CN

CNO

BF3.Et2O

NBSOR1

R1

R2

Ph

Ph

R1

R1

R2

OO

Ph

Ph

R1

R1

R2

O

O

O

N

O

O

O

O

O

>140 ˚C

<140 ˚C

68%80%

hν

R1 = -CN; R1-R1 =

R2 = H, CH3

Scheme 34. Synthetic pathways for preparation of 9,10-diphenyl- and 1-methyl-9,10-diphenyl-2,3-dicyanoantracene



CH2Br

t-Bu

CH2CN

t-Bu

I2

MeONaN

N

N

N

N

NMgN

N

R

R

R

R

R

R

R

R

t-Bu

CN-Mg(OR)2

NC

CN

t-Bu

R =

Bu-t

Scheme 37. Synthetic pathway for preparation of octaaryl substituted tetraazaporphyrins

Scheme 38. Preparation of tetra-tert-butyl tetraazaporphyrins

O PCl5 ClCl KOH Br2

Br Br

CuCN

NC CN

Mg(OBu)2

N

N

N

N

N

N

N

NMg

t-Bu t-Bu

t-But-Bu

N

NH

N

N

N

HN

N

N

t-Bu t-Bu

t-But-Bu

N

N

N

N

N

N

N

NM

t-Bu t-Bu

t-But-Bu

HOAc

hν

MXn

N

N CO2R

CO2R N

N CO2R

CO2R

O2N

N

N CO2R

CO2R

ClCl

ClCl

N

N CO2R

CO2R

N

N CO2R

CO2R N

N CO2R

CO2R

O

ON

N CO2R

CO2R

O

O

Fig. 13. Examples of esters of quinoxaline-2,3-dicarboxylic acid obtained by condensation between aromatic 1,2-diamines and esters of 2,3-dioxosuccinic acid

The preparation of octaalkyl substituted tetraazapor-phyrins was developed by Fitzgerald in the 1990s [217–219]. The reaction pathway utilizes halogenation of dialkylacetylenes followed by the Rosenmund-von Braun reaction with formation of 1,2-dialkylfumaronitriles. Since 1,2-dialkylfumaronitriles cannot directly form tar-get tetraazaporphyrins, they should be photochemically interconverted into 1,2-dialkylmaleonitriles in moderate

yields, which then can be tetramerized into respective octaalkyltetraazaporphyrins. Similarly, iodination of phenyl- and ferrocenylacetylenes leads to formation of both Z and E isomers of 1-phenyl-1,2-diiodoethylene [220, 221] and 1-ferrocenyl-1,2-diiodoethylene [222], which can be transformed into corresponding nitriles under the Rosenmund-von Braun reaction conditions. The Z isomers both of 1-phenyl-1,2-dicyanoethylene and 1-ferrocenyl-1,2-dicyanoethylene can be introduced into reaction with magnesium alkoxide with formation of magnesium 2,7(8),12(13),17(18)-tetraphenyl- and 2,7(8),12(13),17(18)-tetraferrocenyltetraazaporphyrin, respectively, while corresponding E isomers cannot form target macrocycles [221, 223, 224]. Synthesis of tetra-tert-butyl substituted tetraazaporphyrin and its tran-sition-metal complexes requires preparation of tert-butyl-maleonitrile as key precursor for macrocyclization reaction (Scheme 38). The synthetic strategy for preparation of

N

N

N

N

N

N

N

N

R1

R2 R1

R2

R1

R2R1

R2

M

Fig. 14. Structure of substituted tetraazaporphyrins



this compound was developed in the late 1970s [225]. In the fi rst two reaction steps, 3,3-dimethylbutan-2-one was converted fi rst into 2,2-dichloro-3,3-dimethylbu-tane, which was transformed into tert-butylacetylene. The photochemical bromination of carbon-carbon triple bond leads to formation of 1-tert-butyl-1,2-dibromoethylene, which can be transformed into tert-butylmaleonitrile at the standard Rosenmund-von Braun reaction conditions. Template condensation of this nitrile using magnesium butoxide leads to formation of the target magnesium tet-ra-tert-butyl tetraazaporphyrin. The central magnesium atom in this compound can be easily removed by the treatment of this macrocycle with acetic acid to result in metal-free tetra-tert-butyl tetraazaporphyrin [225–227]. Finally, transition-metal ion can be introduced into mac-rocyclic cavity by treatment of metal-free tetra-tert-butyl tetraazaporphyrin with respective transition-metal salts (Scheme 38).

The smallest “conclave” type magnesium dibenzobar-relenotetraazaporphyrin can be prepared by tetramer-ization of 2,3-dicyano[5,6:7,8]dibenzobarrelene in the presence of magnesium alkoxide in low yield. Its demeta-lation in acetic acid provides access to corresponding metal-free compound, while its further treatment with transition-meta salts results in formation of transition-metal complexes [228, 229]. 2,3-dicyano[5,6:7,8]diben-zobarrelene can be prepared using three different ways. The fi rst method is a single step direct reaction between anthracene and dicyanoacetylene at 120 °C [229–231], which has limited use because of the diffi culties of preparation of dicyanoacetylene and its high toxicity. The second approach includes a single reaction between anthracene and more stable and easier available (com-pared to dicyanoacetylene) chloromaleo(fumaro)nitrile at 200 °C [232]. The third reaction pathway is prob-ably the simplest method for preparation of target nitrile

although it includes three steps. First, the Diels-Alder reaction between anthracene and dimethyl ester of acet-ylene dicarboxylic acid leads to formation of dimethyl ester of [5,6:7,8]dibenzo barreleno-2,3-dicarboxylic acid, which upon treatment with ammonia can be transformed into respective diamide in excellent yield. The diamide of [5,6:7,8]dibenzobarreleno-2,3-dicarboxylic acid then should be converted into target dinitrile using thionyl chloride in DMF [233].

The parent tetra-tert-butyltetraazaporphyrin and its transition-metal complexes can be introduced into elec-trophilic aromatic substitution reactions with appro-priate electrophiles. For instance, metal-free as well as copper and cobalt tetra-tert-butyltetraazaporphy-rins can be nitrated by: (i) the mixture of fumic nitric and glacial acetic acids; (ii) nitronium tetrafl uorobo-rate in sulfolane; (iii) dinitrogen tetroxide in hexane [234, 235]. The ratio of mono-, di-, tri-, and tetra nitro tetra-tert-butyltetraazaporphyrins in reaction mix-ture depends on the specifi c reaction conditions. For instance, nitration of metal-free tetra-tert-butyltet-raazaporphyrin in 1:1 mixture of fumic nitric acid and acetic acid for 3–4 minutes at 0 °C results in formation of di-, tri-, and tetranitro tetra-tert-butyltetraazapor-phyrin in 26, 23, and 37% yields, respectively, while increase of the reaction time leads to further increase of the yield of tetranitro product in reaction mixture (Scheme 39). Metal-free 2,7(8),12(13),17(18)-tetra-tert-butyl-3-nitrotetraazaporphyrin can be obtained in 6% yield only when nitration is conducted using 2% nitric acid in acetic acid. Nitration of cobalt and copper 2,7(8),12(13),17(18)-tetra-tert-butyltetraazaporphyrins results in pure tetranitro derivative (52% yield) in the case of cobalt, and the mixture of tri- and tetranitro products (14 and 45% yields, respectively) in the case of copper tetraazaporphyrin.

Scheme 39. Nitration of metal-free tetra-tert-butyl tetraazaporphyrin

N

N

N

N

N

N

NH

HNt-Bu

t-Bu

t-Bu

t-Bu

N

N

N

N

N

N

NH

HNt-Bu

t-Bu

t-Bu

t-Bu

NO2

NO2

O2N

O2N

N

N

N

N

N

N

NH

HNt-Bu

t-Bu

t-Bu

t-Bu

NO2

O2N

N

N

N

N

N

N

N

Nt-Bu

t-Bu

t-Bu

t-Bu

NO2

NO2

O2N

O2N

Co

20 ˚C

TCBCoCl2

72%

+HNO3-AcOH (1:1)

, 2 h

5%



Similarly, the bromination of the metal-free tetra-tert-butyl tetraazaporphyrin with N-bromosuccin-imide in chloroform leads to formation of a mixture of respective mono-, di-, tri-, and tetrabromo deriva-tives, which can be separated using thin-layer chroma-tography [235]. Use of 10–20 times excess of NBS in this reaction leads to selective formation of metal-free 2,7(8),12(13),17(18)-tetra-bromo-3,8(7),13(12),18(17)-tetra-tert-butyltetraazaporphyrin in 80% yield. The recently solved crystal structures of two individual posi-tional isomers of copper mono- and tetrabromo-tetra-tert-butyl tetraazaporphyrins (Fig. 15) confi rm that the bromine atoms substitute only tetraazaporphyrin core.

Bromine atom in the bromo-substituted tetraazapor-phyrins can be introduced into the numerous nucleophilic substitution reactions with copper cyanide, secondary

amines, phenolates, and thiolates as nucleophiles as well as coupling reactions with alkenes and cop-per alkynes to result in respective products shown in Scheme 40 [223–235]. Similar nucleophilic substitution or coupling reactions with dibromo tetra-tert-butyl tet-raazaporphyrins and respective reagents (copper cyanide, sodium thiolate, or copper ferroceneacetylenide) again lead to the formation of expected disubstituted reaction products [223].

Reaction of succinonitrile with phosphorous pentox-ide at elevated temperatures leads to formation of chloro-maleonitrile [115]. This nitrile can be used either for direct macrocyclization reaction or as starting material for nucleophilic substitution reaction with appropriate nucleophiles. In the case of direct template tetrameriza-tion reaction, the nature of substituents in tetraazapor-phyrin core is dictated by the reaction conditions. For instance, when VCl3 was used as a template and trichlo-robenzene as a solvent, vanadyl 2,7(8),12(13),17(18)-tetrachloro tetraazaporphyrin is formed, while if the reaction was conducted with magnesium amylate as a template and amyl alcohol as a solvent, magnesium 2,7(8),12(13),17(18)-tetraamyloxy tetraazaporphyrin is the only reaction product [236]. Another interesting observation is that the nucleophilic substitution of chlo-rine atom in initial chloromaleonitrile can result in for-mation of not only expected substituted maleonitrile, but also (in some cases) of substituted fumaronitriles with product E:Z ratio depending on the nature of nucleophile (Scheme 41) [237, 238]. Target substituted nitriles then can be used for template preparation of corresponding tetrasubstituted tetraazaporphyrins, which have different optical properties and oxidation potentials as discussed below. Because of the presence of bulky substituents in tetraazaporphyrin core, the solubility of these complexes in common organic solvents is much higher compared to those in unsubstituted analogs.

N

N

N

N

N

NMN

N

t-Bu

Br

t-Bu

t-Bu

t-Bu

C5H10NH, DMF (79%)

CuCN, DMF (75%)

PhONa, DMF (26.5%)

PhSNa, DMF (26%)

PhCH=CH2, PPh3

Pd(OAc)2, Et3N (41%)

N

N

N

N

N

NMN

N

t-Bu

R

t-Bu

t-Bu

t-Bu

FcC CCu, DMF (32%)

M R

Cu -CN

2H -NC5H10

2H -OPh

2H - SPh

2H -C=CPh

Cu -C CFc

PhC CCu, DMF (81%)Cu -C CPh

Scheme 40. Nucleophilic substitution and coupling reactions based on monobromo-tetra-tert-butyl tetraazaporphyrin complexes

Fig. 15. X-ray crystal structure of C2v isomer of copper 2,7,13,18-tetrabromo-3,8,12,17-tetra-tert-butyl tetraazaporphyrin [224]



Radical photochlorination of succinonitrile at 140 °C leads to formation of dichloromaleonitrile, which is another excellent precursor for preparation of octasub-stituted tetraazaporphyrins. Template condensation of this precursor leads to formation of vanadyl octachloro-tetraazaporphyrin. It was found that during the template condensation of dichloromaleonitrile with magnesium alkoxide in amyl alcohol, however, only two out of eight chlorine atoms are substituted by amyloxide substituents and the positions of these groups remains unclear [236]. Nucleophilic substitution of chlorine atoms in dichlo-romaleonitrile can result in substitution of one or both chlorine atoms depending on reaction conditions and nature of nucleophile (Scheme 42) [237]. Similarly to nucleophilic substitution reactions of chloromaleonitrile, usage of tert-amylthiolate or tert-butyl amine leads to formation of substituted derivatives of maleo- or fuma-ronitriles. It is interesting that when thiolate is used as a nucleophile, the disubstituted maleonitrile remains major reaction product, while in the case of tert-butylamine, disubstituted fumaronitrile is the major reaction prod-uct. Nitriles received using this reaction pathway can be used for preparation of corresponding octasubstituted tet-raazaporphyrins with high solubilities and specifi c redox and optical properties.

Reaction of dinitrogen tetroxide and phenyl- and p-tert-butylphenylmaleonitrile leads to formation of 1-nitro-2-phenyl- and 1-nitro-2-(p-tert-butylphenyl)-1,2-dicyanoethylenes, which are potentially useful

precursors for preparation of 2,7(8),12(13),17(18)-tet-ranitro-3,8(7),13(12),18(17)-tetra(aryl) tetraazaporphy-rins [239]. The most common pre cursors for preparation of 2,7(8),12(13),17(18)-tetracyano-3,8(7),13(12),18(17)-tetra-R (R ≠ CN) octasubstituted tetraazaporphyrins are substituted tricyanoethylenes. There are two major synthetic routes for preparation of substituted tricya-noethylenes. The fi rst one requires preparation of ben-zylidenemalononitriles by condensation of appropriate benzaldehyde with malononitrile. This intermediate can be further converted into substituted tricyanoethane using potassium cyanide in acetic acid. Finally, treat-ment of substituted tricyanoethane with bromine in pyridine leads to formation of respective substituted tri-cyanoethylenes (Scheme 43) [240, 241]. As usually, the magnesium complexes of 2,7(8),12(13),17(18)-tetracya-no-3,8(7),13(12),18(17)-tetra-R tetraazaporphyrins can be prepared by template cyclization with magnesium alkoxide or hydroxide. Copper and vanadyl complexes can be made by direct tetramerization with appropriate transition-metal salts, while preparation of cobalt com-plex requires in situ demetalation of magnesium precur-sors with acetic acid in the presence of cobalt chloride.

Another common method for preparation of substi-tuted tricyanoethylenes is the direct tricyanovinylation reaction between tetracyanoethylene and N,N-dialkylani-lines, tert-butylamine, chloromercurioferrocene, or fer-rocene (Scheme 44) [242–245]. In the fi rst three cases, reaction can be conducted at very mild conditions in

Nu

R

CN

CNEtOH

Cl CN

CN

CN

RNC

Mg(OR)2 N

N

N

N

N

N

N

NMg

R

R

R

R

Nu Rt-BuONat-AmSNaC6F5ONa(CF3)2CHONaPhC(CF3)2ONat-BuNH2

Et2NH

+

pure not observedpure not observedpure not observedpure not observedpure not observed 1 : 1 2 : 1

Rt-BuO-t-AmS-C6F5O-(CF3)2CHO-PhC(CF3)2O-t-BuNH-Et2N-

Scheme 41. Preparation of tetrasubstituted tetraazaporphyrins using nucleophilic substitution reaction of chloromaleonitrile

Nu

R

CN

CN

Cl CN

CN

CN

RNC

Nu REtONa EtO- pure not observed not observedC6F5ONa C6F5O- pure not observed not observedNaSCH2CH2SNa -SCH2CH2S- not observed pure not observedt-AmSNa t-AmS- not observed 3 : 1t-BuNH2 t-BuNH- not observed 2 : 3

+

Cl R

CN

CN

Cl R R+

Scheme 42. Preparation of disubstituted maleonitriles using dichloromaleonitrile



DMF or acetonitrile with medium to high yield of target compounds. Direct reaction between ferrocene and tetra-cyanoethylene can be conducted in sulfolane at elevated temperatures and proceeds via single-electron transfer mechanism [244]. Template cyclization of obtained by this method nitriles results in formation of respective tetraazaporphyrins.

Finally, a large group of peripherally substituted reduced tetraazaporphyrins (i.e. chlorins, bacteriochlo-rins, and isobacteriochlorins) has been reviewed by Makarova and Lukyanets [246] and thus will not be dis-cussed in this review.

General trends in optical spectra of substituted phthalo cyanines and their analogs

Tuning the optical properties of substituted phthalo-cyanines and their analogs is one of the most important

tasks when optical limiting, PDT, and passive qual-ity switching and mode locking of lasers applications are considered [247–250]. Since in the predominant majority of these applications the only long-wavelength Q band position is exploited, the infl uence of differ-ent substituents on the energy of this band is discussed below. An introduction of the classic (-CN, -NO2, -SO2Ph,-NHCOCH3, -CF3, etc.) electron-withdrawing groups into phthalocyanine core has little effect on the optical properties of corresponding phthalocyanines [20, 35, 126, 141, 251]. For instance, the position of Q band (or Qx and Qy bands in the case of metal-free compound) in zinc and metal-free 1,8(11),15(18),22(25)-tetrasub-stituted phthalocyanines is almost the same compared to parent unsubstituted phthalocyanine with the largest deviation of ~5 nm; a similar trend was observed for other tetrasubstituted at α-positions (Fig. 1) phthalocya-nines [20, 35, 252]. The presence of electron-withdrawing

NC

CN

CN

NC

N

N

N

N

N

NMgN

N

R

R

R

R CN

CN

NC

NC

Mg(OR)2

NAlk

Alk

t-BuNH2

Fe

HgClFe

NAlk

Alkt-BuNH- Fe

NC

CN

CN

R

R =

Scheme 44. Tricyanovinylation of organic and organometallic compounds by direct reaction with tetracyanoethylene

R

CN

CN R

CN

CN

NC

R

CN

CN

NC

N

N

N

N

N

NMN

N

R

R

R

R CN

CN

NC

NC

N

N

N

N

N

NMgN

N

R

R

R

R CN

CN

NC

NC

N

N

N

N

N

NCoN

N

R

R

R

R CN

CN

NC

NC

KCN

AcOH

Br2

Mg(OAm)2

Mg(OH)2

CoCl2

AcOH

TCB

CF3

FF

F

F F OCF2CHFCF3

NF2HCF2CH2C

Py

MXn

M = Cu, VO

R =

Scheme 43. Preparation of selected fl uorine-substituted tetracyanotetraazaporphyrins



substituents at β-positions of phthalocyanine core results in ~10–15 nm low-energy shift of appro-priate Q band. For instance, Qx-band position in metal-free 2,9(10),16(17),23(24)-tetranitrophtha-locyanine is observed at ~15 nm longer wave-length compared to unsubstituted phthalocyanine complex [35, 141, 252]. It is interesting that Q band position in 2,9(10),16(17),23(24)-tetrani-trophthalocyanines and 1,3,8(11),10(9),15(18),17(16),22(25), 24(23)-octanitro phthalocyanines is the same (within an experimental error) confi rm-ing negligible effect of the nitro groups located at α-positions [141, 252]. An infl uence of the substituents with electron-withdrawing inductive effect, but positive mesomeric effect (–I/+M substituents, i.e. chlorine atom) on the position of Q band in UV-vis spectra of phtha-locyanines is more complex (Fig. 16) [253]. Indeed, introduction of four chlorine atoms into β-positions (64)results in a small (6 nm) hypsochromic shift of Q band, while introduction of eight chlorine atoms into the same β-positions (68) leads to small (8 nm) red shift of Q band. Substitution of hydrogen atoms by chlorine atoms at α-positions of phthalocyanine core has more consistent bathochromic effect (5 nm for tetra- and 22 nm for octa-substituted vanadyl complexes 65 and 67, respectively). Finally, hexadecachloro complex 69 has the lowest energy Q band at 745 nm. Similar trends were also observed in the case of fl uorosubstituted copper phthalocyanine com-plexes although stronger electron-withdrawing nature of fl uorine substituents cause larger bathochromic shift of Q band [35, 125].

Similarly to ortho-substituted biphenyl compounds, it can be expected that the phenyl substituents in phe-nylsubstituted phthalocyanines will not be coplanar with phthalocyanine core and thus position of Q band in UV-vis spectra of these compounds should not be signifi cantly affected. The infl uence of coplanarity on the position of Q band in tetraphenylsubstituted phthalocyanines was fi rst investigated at the end of the 1960s (Fig. 17) [35, 254]. First, when π-conjugation between phthalocyanine core and aromatic substituents is broken because of pres-ence of chlorine atom in ortho-position of phenyl ring, which brings the substituent’s and macrocyclic phenyl rings to nearly perpendicular positions (complexes 72

and 75), Q band position is almost indistinguishable from that in unsubstituted vanadyl phthalocyanine 63. The partial π-orbital overlap between aromatic substituents and phthalocyanine core in complexes 71, 72, and 74,however, results in appreciable (11–27 nm) red shift of Q band. Similarly to chlorine substituted phthalocyanines discussed above, the infl uence of aromatic substituents at β-positions is signifi cantly smaller compared to those located at α-positions. In the case of 2,3,9,10,16,17,23,24-octaphenylphthalocyanine complexes two phenyl groups are located next to each other, forcing these substitu-ents from the phthalocyanine plane. As a result, it can be expected that Q band in these compounds will be close to the position of Q band in 2,9(10),16(17),23(24)-tetraphenyl phthalocyanines. Indeed, Q band is located at 691 nm in copper 2,9(10),16(17),23(24)-tetraphe-nylphthalocyanine, while it is observed at 698 nm in cop-per 2,3,9,10,16,17,23,24-octaphenylphthalocyanine.

The infl uence of non-planarity of phthalocyanine macrocycle on observed position of Q band was the sub-ject of investigations starting from the early 1990s and can be illustrated using low-symmetry complexes 76–80(Fig. 18) [255]. These low-symmetry phthalocyanines have unexpected accidental fi rst excited state degen-eracy, which results in a singe Q band in their UV-vis spectra. Such degeneracy allows direct observation of infl uence of non-planarity of phthalocyanine core on position of Q band. Indeed, position and intensity of Q band in planar complexes 76–78 are almost the same. The presence of two or three adjusted tetraphenylisoin-dole fragments in complexes 79 and 80, respectively, cause the deviation of the phthalocyanine core from pla-

narity as a result of steric interactions between neighboring phenyl substituents at different isoin-dole units. Because of such deviations, Q band gradually shifted to the lower energy and loses its intensity. The loss of Q band intensity can be eas-ily explained on the basis of worsening of π−πoverlap within phthalocyanine core, while its red shift refl ects smaller HOMO-LUMO energy gap in these complexes.

Unlike introduction of electron-withdrawing groups, which have little effect on the posi-tion of Q band in substituted phthalocyanines,

N

N

N

N

N

N

N

NV

R1

R2

R1R2

R1

R2

R1R2

R1 R2 Q-band, nm

70 H C6H5 71571 H p-ClC6H4 71672 H o-ClC6H4 70773 C6H5 H 72374 p-ClC6H4 H 73175 o-ClC6H4 H 704

O

Fig. 17. Infl uence of position and coplanarity of substituents aromatic ring on position of Q band

N

N

N

N

N

N

N

NV

R1

R2

R3 R4 R1R2

R3

R4

R1

R2

R3R4R1R2

R3

R4

O

R1 R2 R3 R4 Q-band, nm

63 H H H H 70464 H Cl H H 69865 Cl H H H 70966 Cl Cl H H 71467 Cl H H Cl 73668 H Cl Cl H 71269 Cl Cl Cl Cl 745

Fig. 16. Wavelength of Q band in chlorinated vanadyl phthalocyanines as a function of number and positions of chlorine substituents



peripheral electron-donating groups can signifi cantly affect its energy [35, 252]. For instance, the pres-ence of four alkoxy or aryloxy groups at α-positionsof 1,8(11),15(18),22(25)-tetrasubstituted phthalocya-nines results in ~15–20 nm red shift of Q band com-pared to unsubstituted analogs [252, 256, 257]. Adding four more alkoxy groups at α-positions leads to forma-tion of 1,4,8,11,15,18,22,25-octaalkoxyphthalocyanine, in which Q band is 60–90 nm red-shifted compared to unsubstituted analogs [35, 252, 257–259]. Not surpris-ingly, these complexes are green rather than blue in color. Similar introduction of four or eight alkyl- or phe-nylsulfanyl groups into α-positions of phthalocyanine macrocycle results in even larger red shift of Q band position [35, 146, 252]. For instance, Q band was observed at 670, 709, and 780 nm in the case of zinc phthalocyanine, 1,8(11),15(18),22(25)-tetrabutylsulfanylphthalocya-nine, and 1,4,8,11,15,18,22,25-octabutylsulfanylphthal-ocyanine, respectively, with the last macrocycle having Q band located in spectrum region traditional for 2,3-naph-thalocyanines. The presence of alkyl- and aryloxy as well as alkyl- and arylsulfanyl substituents at β-positions of phthalocyanine core has signifi cantly smaller effect on the position of Q band [146, 260–262]. For instance, both zinc 2,9(10),16(17),23(24)-tetrabutoxy- and 2,3,9,10,16,17,23,24-octabutoxyphthalocyanines have Q band located at 674 nm, just 5 nm lower in energy compared to that in unsubstituted analog [35, 252]. Again, alkyl- or arylthio substituents shift Q band to lower energy more prominently compared to alkyl- and aryloxy groups. For instance, Q band in zinc 2,9(10),16(17),23(24)-tetrabu-tylsulfanyl- and 2,3,9,10,16,17,23,24-octabutylsulfanyl phthalocyanine is red-shifted for 17 and 37 nm, respec-tively [35, 252]. It is interesting that when comparison between alkylthio and arylthio substituents located at the same positions of the phthalocyanine ring is possible, Q band position is located at the same posi-tion (within experimental error), suggest-ing that electron density shift originating from alkyl- or aryl groups attached to the sulfur atom is negligibly small. The infl uence of phenylsulfanyl substitu-ents in 1,3,8(11),10(9),15(18),17(16),22(25),24(23)-octaphenylsulfanylphthalo-cyanines can be seen as an accumulative

effect of the four substituents at α- and four substituents at β-positions of phthalocyanine core resulting in Q band position at 750 nm [35]. Finally, it seems that introduc-tion of sixteen phenylsulfanyl substituents into phthalo-cyanine core results in “saturation” of substituent’s donor effect. Indeed, Q band position in copper hexadecaphe-nylsulfanylphthalocyanine (783 nm) is almost the same as in 1,4,8,11,15,18,22,25-octabutylsulfanylphthalocya-nine (780 nm) [145].

Probably the most dramatic changes in the position of phthalocyanine Q band can be observed in the case of amino and dialkylamino substituted phthalocyanines (Fig. 19) [35, 132, 140, 263]. Similar to the other elec-tron-donating groups discussed above, an introduction of amino or substituted amino groups into α-positionsof phthalocyanine core has signifi cantly more prominent infl uence on Q band position compared to the introduc-tion of the same number of substituents into phthalocya-nine β-positions. For instance, in the case of complex 81Q band was observed at ~50 nm longer wavelength com-pared to that in β-substituted complex 83 and at ~120 nm longer wavelength compared to unsubstituted vana-dyl phthalocyanine. It is interesting that sterically more crowded octasubstituted complex 84 actually has a blue-shifted Q band compared to tetrasubstituted analogue 83and the same situation was observed for the respective copper compounds. Introduction of the bulky tert-butyl groups (complex 82) does not signifi cantly affect posi-tion of the Q band compared to that observed in the par-ent complex 81 (7 nm difference). Similarly, it seems that inductive electron-donor effect of N-alkyl groups is relatively small (13 nm for complexes 83 and 88 as well as 15 nm for complexes 82 and 86). An infl uence

N

N

N

N

N

N

N

NV

R1

R2

R3 R1R2

R3

R1

R2

R3R1R2

R3

O

Q, nm

81 R1 = NMe2 R2 = R3 = H 82882 R1 = NMe2 R2 = H R3 = t-Bu 82183 R1 = R3 = H R2 = NMe2 77584 R1 = H R2 = R3 = NMe2 76085 R1 = NH2 R2 = H R3 = t-Bu 80686 R1 = R3 = H R2 = NHCOCH3 70487 R1 = R3 = H R2 = NHCONH2 73088 R1 = R3 = H R2 = NH2 762

Fig. 19. Structures and Q-band positions of selected amino and dialkylamino vanadyl phthalocyanines

N

N

N

N

N

N

N

NCu

R

R

R

R

R

R

R

R

PhPh

Ph

Ph

SPh

R

R

R

R

= A

=

=

= B Q, nm(log ε)

76 AAAA 694(5.24)77 AAAB 696(5.20)78 ABAB 696(5.23)79 AABB 715(5.08)80 ABBB 741(4.98)

Fig. 18. Low-symmetry phthalocyanines with accidental S1 state degeneracy and non-planar distortions



of the substituents directly attached to the nitrogen atom can be further explored using complexes 83 and 86–88 as an example. Indeed, Q band position in tetra-N-acyl complex 86 is exactly the same as in unsubstituted vanadyl phthalocyanine. Q band in tetraureido phthalocyanine 87is ~25 nm red-shifted, suggesting overall electron-donor effect of these substituents, which is stronger than alkoxy but weaker than alkylthio groups located at the same positions. Another ~30 nm red shift of Q band can be achieved by complete depro-tection of amino groups (88), while additional ~10–15 nm red shift can be gained by alkylation of remaining NH protons (83). As it can be expected, the protonation of peripheral amino groups in complexes 81–85 and 88eliminates nitrogen atom lone-pair-macrocycle conjuga-tion and results in a blue shift of Q band.

One of the examples of substituted phthalocyanines in which both electron-donating and electron-with-drawing groups are present in each isoindole fragment is copper 1,8(11),15(18),22(25)-tetrapiperidino-3,10(9),17(16),24(23)-tetranitrophthalocyanine [35, 252]. This complex has Q band observed at 759 nm, suggesting (as expected) that the energy of this π−π∗ transition is dominated by the electron-donor effect of substituents located at α-positions of phthalocyanine ring, while the infl uence of electron-withdrawing nitro groups located at β-positions is negligibly small.

The infl uence of the chlorine atoms present in amino-, aryloxy-, and alkyl(aryl)thio substituted phthalocya-nines is very interesting (Fig. 20) [142–145]. In the case of tetrachloro dodecaphenoxy zinc phthalocyanine 89 Q band is slightly blue-shifted (7 nm) compared to that observed in 1,8(11),15(18),22(25)-tetramethoxy-phthalocyanine analogue. Since Q band in vanadyl 1,4,8,11,15,18,22,25-octachlorophthalocyanine and 2,3,9,10,16,17,23,24-tetra(N,N-dimethylamino)phtha-locyanine are signifi cantly red-shifted (32 and 54 nm, respectively) compared to unsubstituted phthalocya-nine, it can be ex pected that cumulative effect of both chloro and substituted amino groups will result in large red shift of Q band in phthalocyanines 90 and 91.The observed Q band, however, again is blue-shifted (16 nm) compared to that observed in transition-metal2,3,9,10,16,17,23,24-tetra (N,N-dimethylamino)phthalo-cyanines. In both cases, it seems that σ-acceptor prop-erties of chlorine substituents dominate over π-donatingabilities, resulting in overall electron-withdrawing effect. This situation is quite different from that observed in regular chlorinated phthalocyanines 64–69 in which it seems that π-donor properties of chlorine substituents play a signifi cant role in observed position of Q band.

An angular benzannulation of phthalocyanine core leads to formation of 1,2-naphthalocyanine derivatives. Q band position in these complexes is almost the same

compared to corresponding phthalocyanine compounds [35, 192, 193]. For instance, Q band in zinc phthalo-cyanine and 1,2-naphthalocyanine was observed at 687 and 691 nm, respectively. More interestingly to follow Q band position and its splitting into Qx and Qy components when it is possible to separate four individual positional isomers of 1,2-naphthalocyanines (Fig. 11). Following fi rst-order perturbation theory, it is expected that Q band splitting in the individual isomers of 1,2-naphthalocya-nine will follow the following order: D2h > Cs ~ C2v > C4h (no splitting), while its gravity center should remain the same [20, 264]. In excellent agreement to this predic-tion, it was found that in the case of individual positional isomers of magnesium 1,2-naphthalocyanine Q band is centered at 684 ±1 nm, while Q band splitting follows the order D2h (13 nm) > Cs (5 nm) > C2v = C4h (no splitting).

Unlike in 1,2-naphthalocyanines, the linear ben-zannelation of phthalocyanine results in formation of 2,3-naphthalocyanines, the Q band position of which follows the well-recognized “100 nm” rule: each suc-cessive linear benzan nelation of tetraazaporphyrin core leads to ~100 nm low-energy shift of Q band [20, 35, 173]. Indeed, Q band in vanadyl 2,3-naphthalocyanine is observed at 820 nm, which is ~100 nm lower in energy than Q band observed in phthalocyanine analogue (704 nm) [172, 265]. Again alkyl or dibenzobarrelene substitu-ents located at sixth and seventh positions of naphthalene rings do not signifi cantly affect position of Q band in UV-vis spectra [35, 174, 178]. The infl uence of substituents located at α-positions of 2,3-naphthalocyanine core fol-lows similar trend discussed for phthalocyanine analogs, i.e. electron-donating groups lead to signifi cant red-shift of Q band position, while electron-withdrawing as well as alkyl groups have only a little infl uence (Fig. 21) [35, 173, 187, 258]. Indeed, Q band position in complex 99is close to that observed in unsubstituted complex 93,while the presence of aryl groups at the same position (complexes 95 and 96) brings Q band to ~850 nm region similarly to the infl uence in α-phenyl substituted phthalo-cyanines. Primary (98) and secondary (97) amine groups located at the same position signifi cantly shift Q band to the low-energy region, with the shift magnitude being large for primary amines (160 vs. 90 nm, respectively). The presence of substituents at fi fth and sixth positions of

N

N

N

N

N

N

N

NM

R1

R2

R3 R1R2

R3

R1

R2

R3R1R2

R3

Cl

Cl

Cl

Cl

Q, nm

89 R1 = R2 = R3 = OPh M = Zn 70090 R1 = R2 = Cl R3 = NHPh M = Zn 71491 R1 = Cl R2 = R3 = NEt2 M = Zn 71492 R1 = Cl R2 = R3 = SPh M=Cu 74693 R1 = Cl R2 = R3 = SBu M = Cu 741

Fig. 20. Q-band position in selected examples of chlorinated phthalocyanines with electron-donating groups



naphthalene ring has less effect on the position of Q band in UV-vis spectra of corresponding 2,3-naphthalocyanine complexes. For instance, primary amine substituents at fi fth position of naphthalene ring result in only 70 nm red shift of Q band (complex 101), while the same substitu-ents located at fi rst position of naphthalene ring provide 160 nm (complex 98) red shift of Q band. It is interesting to note that the amino group at sixth and fi fth positions of the naphthalene ring is close to each other (complexes 100 and 103 as well as 102 and 104). Finally, electron-withdrawing substituents located at fi fth, sixth, and sev-enth positions of naphthalene ring (complexes 105–107)have a negligible infl uence on the position of Q band.

In agreement with the “100 nm” rule, substituted transition-metal anthracyanine complexes have Q band at 930–950 nm area [266, 267]. Almost all known anthra-cyanine complexes have cobalt, copper, vanadium, or aluminum ions as a central metal, while anthracyanine complexes with other transition-metals are not stable enough in solution. Where direct comparison is possible (Fig. 22) vanadyl complexes always have red-shifted Q band compared to corresponding copper compounds, while as expected, introduction of tert-butyl substituents into phenyl rings located at ninth and tenth positions of anthracene ring does not affect signifi cantly the Q band position (compounds 108 and 109). Naturally, elimi-nation of aromatic conjugation in anthracenes leads to signifi cant blue shift of Q band. For instance, Q band in

vanadyl tetra-2,3-(9,10-anthraquinono)tetraazaporphyrin 110 was observed at 758 nm, suggesting that there is still some additional conjugation provided by anthraquinone group [268] and similar changes in UV-vis spectra of anthracyanines were observed upon their photochemical degradation in the presence of oxygen.

As discussed above, UV-vis spectra of angularly annelated 1,2-naphthalocyanines are close to those of their parent phthalocyanines. Similarly, the posi-tion of Q band (and UV-vis spectra in general) in tetra-(2,3-phenantro)-tetraazaporphyrins and tetra-4,5-(9,10-phenantro)-tetraazaporphyrins are close to those observed in corresponding 2,3-naphthalocyanine complexes [35, 197, 212]. For instance, Q band was observed at 700 nm in the case of copper tetra-4,5-(9,10-phenantro)-tetraazaporphyrin, which is red-shifted for only 29 nm compared to that in copper tetra-tert-butylphthalocyanine.

Aza-substitution of one or two carbon atoms in phtha-locyanine, naphthalocyanine, or anthracyanine core leads to formation of corresponding pyridino-, pyrazino-, quinoxalino-, and benzo[g]quinoxalinoporphyrazines,respectively (Fig. 23) [35, 173, 268–272]. In all of these complexes, Q band is blue-shifted compared to respec-tive carbon analogs and the degree of this shift depends on the nature of the substituents. Thus, Q band position in tetra-3,4-pyridoporphyrazines 112 is close to that observed in parent phthalocyanine complexes, while it is

N

N

N

N

N

N

N

NM

R

R

R

R

R

R

R

R

O O

108935 nm878 nm

109936 nm890 nm

MV=OCu

110758 nm

Fig. 22. Q band position in selected anthracyanines

N

N

N

N

N

N

N

NV

R

R

R

R

R

R

R

R

O

Cl Cl

t-Bu

Ph

pip

pip

AcHN

O2N

O2N

H2N

H2N

Me2NRHN

C6H11NHt-Bu-C6H4

t-Bu

93(820 nm)

94(826 nm)

95(845 nm)

96(850 nm)

97(910 nm)

98(980 nm)

99(832 nm)

100(886 nm)

101(890 nm)

102(845 nm)

103(870 nm)

104(870 nm)

105(820 nm)

106(815 nm)

107(825 nm)

Fig. 21. Position of Q band in selected vanadyl 2,3-naphthalocyanines



~40 nm blue-shifted in the case of tetra-2,3-pyridoporph-yrazine complexes 111. Further ~20 nm blue shift of Q band can be achieved in tetra-2,3-pyrazinoporphyrazines 113 in which two aromatic carbon atoms in parent phtha-locyanine molecule are substituted by nitrogens. Q band position in quinoxalinoporphyrazines is strongly infl u-enced by the orientation of the heteroaromatic ring with respect to tetraazaporphyrin core. In the case of tetra-[b]quinoxalinoporphyrazines 114, Q band position is indeed closer to that observed in respective phthalocyanines rather than that observed in 2,3-naphthalocyanines. On the other hand, Q band position in tetra-[g]quinoxalinop-orphyrazines 115 is only ~20 nm blue-shifted compared to that in copper 2,3-naphthalocyanine. Finally, Q band in both vanadyl and copper complexes of tetra- benzo[g]quinoxalino-porphyrazines 116 is blue-shifted (80–95 nm) compared to that observed in respective anthracya-nine analogs.

Tuning of optical properties of tetraazaporphyrins is probably the most interesting and challenging task in the chemistry of phthalocyanine analogs [20, 35, 173]. Unsubstituted magnesium tetraazaporphyrin has Q band located at 591 nm, in agreement with “100 nm” rule [35, 273]. Similarly to phthalocyanines and related compounds discussed above, an introduction of tert-butyl groups into tetraazaporphyrin core results in a small (7 nm) batho-chromic shift of Q band [225], while stepwise adding of four [221] and eight [215] phenyl groups leads to a gradu-ated red-shift of Q band on ~40 and ~60 nm, respectively making latter complexes close (from absorption point of view) to pyrazinoporphyrazines. Unlike the majority of phthalocyanines and tetraazaporphyrins, UV-vis spec-trum of sterically hindered “conclave’ dibenzobarreleno-tetraazaporphyrins has very close shapes in solution and in a solid state, suggesting negligible aggregation in both cases [274]. An introduction of single ethynylferrocenyl group into magnesium tetra-tert-butyltetraazaporphyrin results in small shift of Q band from 591 to 602 nm, probably because of weak electron-donating effect of this substituent or/and its extended π-system [224]. Tet-raazaporphyrins with electron-withdrawing substituents have similar to parent macrocycle position of Q band [35]. For instance, introduction of four carboxylic acid or cyano groups into vanadyl 2,7(8),12(13),17(18)- tetra-tert-butyltetraazaporphyrin leads to small (8–12 nm) shift of

Q band [240]. In analogy, vanadyl 2,7(8),12(13),17(18)-tetrachloro- and octachlorotetraazaporphyrins have 12 and 18 nm red-shifted Q band, respectively [236]. Similarly to phthalocyanines, an introduction of electron-donating groups into tetraazaporphyrin core can result in large red shift of Q band. Again, red shift follows the order: alkoxy < thio < amine [237]. For instance, magnesium 2,7(8),12(13),17(18)-tetra-tert-butoxytetraazaporphyrin, 2,7(8),12(13),17(18)-tetra-tert-amylthiotetraazaporphy-rin, and 2,7(8),12(13),17(18)-tetra-tert-butylaminotetra-aza porphyrin have Q bands located at 612 (28 nm red shift), 649 (65 nm red shift), and 740 nm (156 nm red shift), respectively. Moreover, an introduction of the electron-withdrawing trifl uoromethyl groups into alkoxy substituents results in the blue shift of Q band compared to its position in traditional tetraalkoxytet-raazaporphyrins. Thus, Q band was observed at 612, 595, and 598 nm in magnesium 2,7(8),12(13),17(18)-tetra-tert-butoxytetraazaporphyrin, 2,7(8),12(13),17(18)-tetra-(hexafl uoro-iso-propoxy)tetraazaporphyrin, and 2,7(8),12(13),17(18)-tetra-(1,1,1,3,3,3-hexafluoro-2-phenyl-iso-propoxy)tetraazaporphyrin, respectively [238]. It is interesting to compare Q band energies in tetra- and octasubstituted by electron-donating groups tetraazapor-phyrins [237]. In the case of tert-amylthio substituents, an introduction of the additional four electron-donating groups is almost negligible (Q band is observed at 649 nm for tetra- and 651 nm for octasubstituted magnesium complexes). In the case of magnesium tetraazaporphyrins substituted by four or eight amino groups, an effect similar to that discussed for aminosubstituted phthalocyanines was observed, i.e. introduction of additional four tert-butylam-ino groups into tetraazaporphyrin core results in signifi cant (17 nm) blue shift of Q band. One of the most intriguing groups of tetraazaporphyrin complexes belongs to “push-pull” macrocycles in which each tetraazaporphyrin core has four electron-withdrawing and four electron-donating substituents (Fig. 24) [240, 241]. In this case, taking into account that four electron-withdrawing groups can be con-sidered as a constant, the position of Q band depends on the electron-donor ability of the second set of substituents. For instance, in the case of magnesium 2,7,12,17-tetra-cyano-3,8,13,18-tetraphenyltetraazaporphyrin Q band is located at 666 nm, while its position for fl uorine-contain-ing complexes 118–120 clearly correlated with relative

N

N

N

N

N

N

N

NM

R

R

R

R

R

R

R

R

NN

NN NN NN

NN

111663 nm649 nm

113645 nm631 nm

MV=OCu

112700 nm

114715 nm712 nm

115

755 nm

116830 nm810 nm

Fig. 23. Q-band position in selected examples of aza-analogs of phthalocyanine



electron-withdrawing properties of fl uorine-containing groups. Introduction of additional electron- donating dim-ethylamino group into para-position of phenyl ring (com-plex 121) results in dramatic 168 nm red shift of Q band position, while less electron-donor tert-butylamino group (complex 122) shift Q band to 756 nm region.

The longest intense absorption band observed in tetraazaporphyrins probably belongs to magnesium or metal-free 2,7(8),12(13),17(18)-tetracyano-3,8(7),13(12),18(17)-tetraferrocenyltetraazaporphyrins [224, 245]. In this case, strong broad peak was observed at ~950 nm region following another sharp and intensive peak at ~620 nm. The nature of this long wavelength band was discussed on a basis of comparison with that observed in 2,7(8),12(13),17(18)-tetraferrocenyltetraazaporphy-rin (which has two of these bands located at ~722 and ~610 nm) and further confi rmed using MCD, electro-chemistry, spectroelectrochemistry as well as DFT and TDDFT calculations. It seems that the NIR band con-sists of several metal-to-ligand charge transfer transi-tions, while more intense and narrow band located at ~620 nm is indeed Q band (π−π* transition). Overall, the ease of electronic structure perturbation in tetraazapor-phyrins allows accurate tuning of the Q band position within a large range of energies starting from ~580 nm and continuing up to NIR area of 834 nm, which indeed

cover the entire range available for phthalocyanines and naphthalocyanines.

CONCLUSION

The preparation and optical properties of peripherally substituted phthalocyanines and their analogs (i.e. naph-thalocyanines, anthracyanines, nitrogen analogs of phtha-locyanines, and tetraazaporphyrins) have been discussed. Specifi cally, an infl uence of the electron-withdrawing or electron-donating nature of peripheral substituents, ben-zannulation of tetraazaporphyrin core, and introduction of the heteroatoms into π-system of specifi c macrocycles were discussed. In many cases, poorly known method-ologies published in mostly unavailable Russian journals were highlighted throughout the review.

Acknowledgements

Generous support from the NSF (CHE-0809203), Research Corporation (Cottrell College Science Award CC6766), and Petroleum Research Foundation (grant 45510-GB3) to VN is greatly appreciated. The recent part of EL work was supported fi nancially by the Moscow Government.

REFERENCES

1. Braun A and Tscherniac J. Chem. Ber. 1907; 40:2709.

2. De Diesbach H and von der Weid E. Helv. Chem. Acta 1927; 10: 886.

3. Danbridge AG, Drescher HA and Thomas J. GBPatent 322169, 1928.

4. Linstead RP. J. Chem. Soc. 1934; 1022. 5. Dent CE, Linstead RP and Lowe AR. J. Chem. Soc.

1934; 1033. 6. Robertson JM, Linstead RP and Dent CE. Nature

1935; 135: 506. 7. Robertson JM. J. Chem. Soc. 1935; 615. 8. Robertson JM and Woodland I. J. Chem. Soc. 1937;

219. 9. Moser FH and Thomas AL. Phthalocyanine Com-

pounds, Reinhold Publ. Corp: New York, 1963. 10. Berezin BD. Coordination Compounds of Por-

phyrins and Phthalocyanines, Wiley: Chichester, 1981.

N

N

N

N

N

N

N

NMg

R

NC

NC

R

R

CN

R

CN

NF

FF

FF F3C HNOCF2CHFCF3

117666 nm

119624 nm

118634 nm

120641 nm

121834 nm

122756 nm

Fig. 24. Q-band position in selected examples of “push-pull” type magnesium tetraazaporphyrins

Fig. 25. UV-vis and MCD spectra of magnesium complexes of 2,7(8),12(13),17(18)-tetracyano-3,8(7),13(12),18(17)-tetra-ferrocenyl- (blue) and 2,7(8),12(13),17(18)-tetraferrocenyl-tetraazaporphyrins (red)



11. Iodko SS, Kaliya OL, Kondratenko, NV, Lukyanets EA, Popov VI and Yagupol’skii LM. Zh. Obshch. Khim. 1983; 53: 901.

12. Sharman WM and van Lier JE. In The Porphyrin Handbook, Vol. 15, Kadish KM, Smith KM and Guilard R. (Eds.) Academic Press: New York, 2003; pp 1–60.

13. McKeown NB. In The Porphyrin Handbook, Vol. 15, Kadish KM, Smith KM and Guilard R. (Eds.) Academic Press: New York, 2003; pp 61–124.

14. Rodriguez-Morgade MS, de la Torre G and Torres T. In The Porphyrin Handbook, Vol. 15, Kadish KM, Smith KM and Guilard R. (Eds.) Academic Press: New York, 2003; pp 125–160.

15. Linstead RP and Weiss FT. J. Chem. Soc. 1950;2975.

16. Moser FH and Thomas AL. The Phthalocyanines,Vols. 1,2, CRC Press: Boca Raton, Fla., 1983.

17. Mikhalenko SA, Barkanova SV, Lebedev OL and Lukyanets EA. Zh. Obshch. Khim. 1971; 41: 2735.

18. Subbotin NB, Tomilova LG, Kostromina NA and Lukyanets EA. Zh. Obshch. Khim. 1986; 56: 397.

19. Dolotova OV, Bundina NI, Derkacheva VM, Negri-movsky VM, Minin VV, Larin GM, Kaliya OL and Lukyanets EA. Zh. Obshch. Khim. 1992; 62: 2064.

20. Phthalocyanines: Properties and Applications,Vol. 1–4, Leznoff CC and Lever ABP. (Eds.) VCH Publishers: New York, 1990–1996.

21. McKeown NB and Budd PM. Chem. Soc. Rev. 2006; 35: 675.

22. Bonnett R. Chemical Aspects of Photodynamic The-rapy, Gordon and Breach Sci. Publ.: 2000.

23. Elemans JAAW, van Hameren R, Nolte RJM and Rowan AE. Adv. Mater. 2006; 18: 1251.

24. Chen Y, Hanack M, Blau WJ, Dini D, Liu Y, Lin Y and Bai J. J. Mat. Sci. 2006; 41: 2169–2185.

25. Hedayatullah M. Compt. Rend. Acad. Sci. Ser. II C 1983; 296: 621.

26. Rodriguez-Morgade MS and Stuzhin P. J. Porphy-rins Phthalocyanines 2004; 8: 1129.

27. De la Torre G, Claessens CG and Torres T. Chem. Commun. 2007; 2000.

28. Rio Y, Rodriguez-Morgade MS and Torres T. Org. Biomol. Chem. 2008; 6:1877.

29. Claessens CG, Uwe H and Torres T. ChemicalRecord. 2008; 8: 75.

30. Hanack M, Metz J and Pawlowski G. Chem. Ber. 1982; 115: 2836.

31. Ponomarenko EP, Gerasimov BG, Molchanova LI, Puchnova VA and Lukyanets EA. Anil.-Kras. Prom.1976; 7: 1.

32. Tomilova LG, Rodionova GN and Lukyanets EA. Koord. Khim. 1979; 5: 549.

33. Vul’fson SV, Kaliya OL, Lebedev OL and Lukya-nets EA. Zh. Obshch. Khim. 1976; 46: 179.

34. Lukyanets EA, Mikhalenko SA and Kovshev EI. Zh. Obshch. Khim. 1971; 41: 934.

35. Electronic Spectra of Phthalocyanines and Rela-ted Compounds, Lukyanets EA (Ed.) NIITEHIM: Cherkassy, 1989; pp 1–95.

36. Gal’pern MG, Dikolenko EG, Donyagina VF, Isaev SD, Kozlov OF, Lukyanets EA, Mikhalenko SA and Yurchenko AG. Zh. Obshch. Khim. 1980; 50: 2390.

37. Rodriguez-Morgade S and Hanack M. Chem. Eur. J.1997; 3: 1042.

38. Mikhalenko SA, Solov’eva LI and Lukyanets EA. Zh. Obshch. Khim. 1988; 58: 2618.

39. Nemykin VN and Struntz N. Unpublished. 40. Mikhalenko SA, Solov’eva LI and Lukyanets EA.

Zh. Obshch. Khim. 1991; 61: 996. 41. Krasnovsky AA Jr, Schweitzer C, Leismann H,

Tanielian C and Lukyanets EA. Quantum Electro-nics 2000; 30: 445.

42. Cass OW. US 2443494, 1948. 43. Mikhalenko SA, Gladyr SA and Lukyanets EA. Zh.

Org. Khim. 1972; 8: 341. 44. Tidwell TT. Science of Synthesis 2006; 23:

679–734. 45. Knoelker H-J. Chem. Rev. 2000; 100: 2941. 46. Mikhalenko SA and Lukyanets EA. Zh. Org. Khim.

1970; 6: 171. 47. Tomilova LG, Ovchinnikova NA, Khused YZ and

Lukyanets EA. 2nd Intern. Sym. Chemistry of Func-tional Dyes, Kobe, Japan, 1992; 17.

48. Kobayashi N, Ashida T and Osa T. Chem. Lett.1992; 2031.

49. Subbotin NB, Nemykin VN and Voloshin YZ. Men-deleev Commun. 1993; 121.

50. Nemykin VN, Subbotin NB, Kostromina NA and Volkov SV. Mendeleev Commun. 1995; 71.

51. Nemykin VN, Subbotin NB, Kostromina NA, Volkov SV and Lukyanets EA. Zh. Neorg. Khim.1995; 40: 1183.

52. Linssen TG and Hanack M. Chem. Ber. 1994; 127:2051.

53. Kobayashi N, Miwa H and Nemykin VN. J. Am. Chem. Soc. 2002; 124: 8007.

54. Fukuda T, Homma S and Kobayashi N. Chem. Eur. J. 2005; 11: 5205.

55. Gal’pern MG, Donyagina VF, Shalaev VK, Skvar-chenko VR, Bundina NI, Mekhryakova NG, Kaliya OL and Lukyanets EA. Zh. Obshch. Khim. 1982; 52: 715.

56. Lukyanets EA, Vazhnina VA, Mikhalenko SA and Solov’eva LI. Anil-kras. Prom. 1976; 1: 1.

57. Hoppf H and Gallegra P. Helv. Chim. Acta 1968; 51:253.

58. Brach PJ, Grammatica SJ, Osanna OA and Wein-berg L. J. Heterocycl. Chem. 1970; 7: 1403.

59. Metz J, Schneider O and Hanack M. Inorg. Chem. 1984; 23: 1065.

60. Somashekarappa MP and Keshavayya J. Synth.React. Inorg. Met.-Org. Chem. 1999; 29: 767.



61. Borodkin VF and Smirnov RP. Izv. Vys. Ucheb. Zaved., Khim. Khim. Tekhnol. 1960; 3: 915.

62. Kropf H and Hofmann H. Tetrahedron Lett. 1967; 659.

63. Shiryaeva LS and Klyuev VN. Tr. Tambovsk. Inst. Khim. Mashinostr. 1969; 3: 101.

64. Klyuev VN and Shiryaeva LS. SU 260044, 1969. 65. Negrimovsky VM, Derkacheva VM, Kaliya OL and

Lukyanets EA. Zh. Obshch. Khim. 1991; 61: 460. 66. Tse Y-H, Kobayashi N and Lever ABP. Coll. Czech.

Chem. Commun. 2001; 66: 338. 67. Giraudeau A, Louati A, Gross M, Andre JJ, Simon

J, Su CH and Kadish KM. J. Am. Chem. Soc. 1983; 105: 2917.

68. Louati A, El Meray M, Andre JJ, Simon J, Kadish KM, Gross M and Giraudeau A. Inorg. Chem. 1985; 24: 1175.

69. Barkanova SV and Kaliya OL. J. Porphyrins Phthalocyanines 1999; 3: 180.

70. Banfi S, Cavazzini M, Coppa F, Barkanova SV and Kaliya OL. J. Chem. Soc., Perkin Trans. 2 1997; 1577.

71. Barkanova SV, Derkacheva VM, Dolotova OV, Li VD, Negrimovsky VM, Kaliya OL and Lukyanets EA. Lett. 1996; 37: 1637.

72. Mikhalenko SA and Lukyanets EA. Zh. Org. Khim. 1975; 11: 2216.

73. Mikhalenko SA and Lukyanets EA. Anilino-kras. prom. 1975; 3: 3.

74. Mikhalenko SA, Surovtseva AP and Lukyanets EA. Anilino-kras. prom. 1976; 11: 1.

75. Mikhalenko SA and Lukyanets EA. Anilino-kras. prom. 1977; 3: 1.

76. Negrimovsky VM, Derkacheva VM, Lukyanets EA, Weitemeyer A, Wöhrle D and Schneider G. Phosph.Sulf. Silicon 1995; 104: 161.

77. Ben-Hur E and Chan W-S. In The Porphyrin Hand-book, Vol. 19, Kadish KM, Smith KM and Guilard R. (Eds.) Academic Press: New York, 2003; pp 1–36.

78. Chan WS, West CML, Moore JV and Hart IR. Brit.J. Cancer 1991; 64: 827.

79. Vorozhtsov GN, Derkacheva VM, Kazachkina NI, Lukyanets EA, Feofanov AV, Fomina GI, Chissov VI and Yakubovskaya RI. RU 2183635, 2002.

80. Derkacheva VM, Vazhnina VA, Kokoreva VI and Lukyanets EA. RU 2181736, 2002.

81. Feofanov A, Grichine A, Karmakova T, Kazachkina N, Pecherskikh E, Yakubovskaya R, Lukyanets E, Derkacheva V, Egret-Charlier M and Vigny P. Pho-tochem. Photobiol. 2002; 75: 527.

82. Vorozhtsov GN, Davydov AB, Kuz’min SG, Losh-chenov VB, Luzhkov YuM, Lukyanets EA, Meero-vich GA and Khromov GL. RU 2187347, 2003.

83. Vorozhtsov GN, Kaliya OL, Luzhkov YuM, Lukya-nets EA, Andronova NV, Bozhevol’nov VE, Gerasi-mova GK, Melikhov IV, Nikolaev AL, Torshina NL and Treshchalina EM. RU 2188054, 2002.

84. Lukyanets EA, Negrimovsky VM, Yuzhakova OA, Kaliya OL, Kuznetsova NA, Pykhtina EV, Ulanova LA, Kovaleva MA, Luzhkov YuM, Vorozhtsov GN, Meerovich GA and Torshina NL. RU 2164136, 2001.

85. Kogan BY, Butenin AV, Torshina NL, Kogan EA, Kaliya OL, Lukyanets EA, Luzhkov YuM, Voro-zhtsov GN, Volkova AI and Meerovich GA. RU 2157268, 2000.

86. Lukyanets EA. J. Porphyrins Phthalocyanines 1999; 3: 424.

87. Severin SE, Posypanova GA, Katukov VU, Shmy-rev II, Luzhkov YuM, Gerasimova GK, Zhukova OS, Vorozhtsov GN, Kaliya OL, Lukyanets EA and Severin ES. Biochem. Mol. Boil. Int. 1997; 43:873.

88. Lutsenko SV, Feldman NB, Finakova GV, Posypa-nova GA, Severin SE, Skryabin KG, Kirpichnikov MP, Lukyanets EA and Vorozhtsov GN. Tumor Biol.1999; 20: 218.

89. Golubeva NA, Demcheva MV, Chernyaeva EB, Yakubovskaya RI, Sukhin GM, Ponomarev GV, Gal’pern MG, Lukyanets EA and Savitsky AP. Proc. SPIE Int. Soc. Optic. Engin. 1996; 2728: 165.

90. Savitsky AP, Lopatin KV, Golubeva NA, Poroshina MYu, Chernyaeva EB, Stepanova NV, Solov’eva LI and Lukyanets EA. J. Photochem. Photobiol. B 1992; 13: 327.

91. Savitsky AP, Meerovich IG, Moser JG, Yakubovs-kaya RI, Kaliya OL and Lukyanets EA. Proc. SPIE Int. Soc. Optic. Engin. 1997; 3191: 343.

92. Meerovich IG, Jerdeva VV, Derkacheva VM, Meerovich GA, Lukyanets EA, Kogan EA and Savitsky AP. J. Photochem. Photobiol. B 2005; 80: 57.

93. Chernonosov AA, Kuznetsov NA, Koval VV, Pysh-nyi DV, Derkacheva VM, Lukyanets EA and Fedo-rova OS. Nucleosides, Nucleotides & Nucleic Acids2004; 23: 983.

94. Koval VV, Chernonosov AA, Abramova TV, Iva-nova TM, Fedorova OS, Derkacheva VM and Lukyanets EA. Nucleosides, Nucleotides & Nucleic Acids 2001; 20: 1259.

95. Chernonosov AA, Koval VV, Knorre DG, Chernenko AA, Derkacheva VM, Lukyanets EA and Fedorova OS. Bioinorg. Chem. Applic. 2006; 63703.

96. Mikhalenko SA, Solov’eva LI, Surovtseva AP and Lukyanets EA. SU 925948, 1982.

97. Solov’eva LI, Mikhalenko SA, Chernykh EV and Lukyanets EA. Zh. Obshch. Khim. 1982; 52: 90.

98. Buck T, Wöhrle D, Schulz-Ekloff G and Andreev A. J. Mol. Cat. 1991; 70: 259.

99. Nemykin VN, Polshyna AE, Borisenkova SA and Strelko VV. J. Mol. Cat. A 2007; 264: 103.

100. Borisenkova SA, Kalashnikova TE, Mokshin VM, Gal’pern MG, Samigulina NN and Lukyanets EA. Zh. Fiz. Khim. 1987; 61: 2995.



101. Akopyants YG, Borisenkova SA, Kaliya OL, Der-kacheva VM and Lukyanets EA. J. Mol. Cat. 1993; 83: 1.

102. Basu B, Satapathy S and Bhutnagar AK. Catal. Rev.1993; 35: 571.

103. Kaliya OL, Lukyanets EA and Vorozhtsov GN. J. Porphyrins Phthalocyanines 1999; 3: 592.

104. Syrkin AB, Zhukova OS, Kikot BS, Gatinskaya LG, Treshchalina EM, Yakubovskaya EYu, Pankratov AA, Mikhailov LM, Zimakova NI, Kolesnikova EYu, Oborotova NA, Polozkova AP, Gerasimova GK, Kaliya OL, Kuz’min SG, Lukyanets EA, Voro-zhtsov GN and Trapeznikov NN. Ross. Khim. Zh.1998; 42: 140.

105. Golub YuM, Dmitrieva ND, Zelikhina VA, Kaliya OL, Kuz’min SG, Lukyanets EA, Mikhalenko SA, Sennikov VA, Solov’eva LI, Teplyakova NV and Yakunina TV. RU 2304582, 2007.

106. Vol’pin ME, Krainova NYu, Moskaleva IV, Novo-darova GN, Vorozhtsov GN, Gal’pern MG, Kaliya OL, Lukyanets EA and Mikhalenko SA. Izv. Akad. Nauk, Ser. Khim. 1996; 2105.

107. Vol’pin ME, Vorozhtsov GN, Kaliya OL and Lukya-nets EA. J. Inorg. Biochem. 1997; 67: 151.

108. Vorozhtsov GN, Kalinichenko AN, Luzhkov YuM, Kaliya OL, Karmakova TA, Lukyanets EA, Morozova ZV, Pankratov AA, Teplyakova NV, Chissov VI and Yakubovskaya RI. RU 2172319, 2001.

109. Dolotova OV and Kaliya OL. Russ. J. Coord. Chem.2007; 33: 111.

110. Mytsyk VM, Nemykin VN and Volkov SV. Ukr. Khim. Zh. 2003; 69: 3.

111. Solov’eva LI and Lukyanets EA. Anil.-kras. Prom.1976; 9: 1.

112. Tylleman B, Gomez-Aspe R, Gbabode G, Geerts YH and Sergeyev S. Tetrahedron 2008; 64: 4155.

113. Sergeyev S, Debever O, Pouzet E and Geerts YH.J. Mat. Chem. 2007; 17: 3002.

114. Dulog L and Gittinger A. Mol. Cryst. Liq. Cryst. A1992; 213: 31.

115. Shevchenko VI and Kukhar VP. Zh. Obshch. Khim.1966; 36: 725.

116. Mikhalenko SA, Solov’eva LI and Lukyanets EA. Russ. J. Gen. Chem. 2004; 74: 451.



119. Solov’eva LI and Lukyanets EA. Zh. Obshch. Khim.1980; 50: 1122.

120. Drechsler U, Pfaff M and Hanack M. Eur. J. Org. Chem. 1999; 3441.

121. Fujiki M, Tabei H and Kurihara T. J. Phys. Chem. 1988; 92; 1281.

122. Dini D and Hanack M. In The Porphyrin Hand-book, Vol. 17, Kadish KM, Smith KM and

Guilard R. (Eds.) Academic Press: New York, 2003; pp 1–36.

123. Derkacheva VM, Mikhalenko SA, Solov’eva LI, Alekseeva VI, Marinina LE, Savina LP, Butenin AV and Lukyanets EA. Russ. J. Gen. Chem. 2007; 77:1117.

124. Kuznetsova N, Makarov D, Derkacheva V, Savvina L, Alekseeva V, Marinina L, Slivka L, Kaliya O and Lukyanets E. J. Photochem. Photobiol., A 2008; 200: 161.

125. Oksengendler IG, Kondratenko NV, Lukyanets EA and Yagupol’skii LM. Zh. Org. Khim. 1977; 13:2234.



128. Kondratenko NV, Nemykin VN, Lukyanets EA, Kostromina NA, Volkov SV and Yagupolskii LM. J. Porphyrins Phthalocyanines 1997; 1: 341.

129. Gal’pern MG, Talismanova TD, Tomilova LG and Lukyanets EA. Zh. Obshch. Khim. 1985; 55:1099.

130. Kondratenko NV, Tretyakova IN, Lukyanets EA, Volkov SV, Orlova RK, Nemykin VN and Yagupols-kii LM. Dyes Pigments 1999; 41: 101.

131. Mikhalenko SA and Lukyanets EA. Zh. Obshch. Khim. 1976; 46: 2156.

132. Mikhalenko SA, Derkacheva VM and Lukyanets EA. Zh. Obshch. Khim. 1981; 51: 1650.

133. Lukyanets EA, Vazhnina VA, Gal’pern MG and Talismanova TD. SU 652173, 1979.

134. Tse Y-H, Janda P, Lam H, Zhang J, Pietro WJ and Lever ABP. J. Porphyrins Phthalocyanines 1997; 1: 3.

135. Tau P and Nyokong T. J. Electroanal. Chem. 2007; 611: 10.

136. Sivanesan A and John SA. Biosens. Bioelectr. 2007; 23: 708.

137. Jung S-H, Choi J-H, Yang S-M, Cho W-J and Ha C-S. Mat. Sci. Engineer., B 2001; B85: 160.

138. Pan Y, Chen W, Lu S and Zhang Y. Dyes Pigments2005; 66: 115.

139. Sorokin AB, Mangematin S and Pergrale C. J. Mol. Cat. A 2002; 182–183: 267.

140. Mikhalenko SA, Solov’eva LI, Ivanova TM and Lukyanets EA. Zh. Obshch. Khim. 1985; 55:1106.

141. Negrimovsky VM, Derkacheva VM and Lukyanets EA. Zh. Obshch. Khim., 1989; 59: 1688.

142. Volkov KA, Negrimovsky VM, Avramenko GV and Lukyanets EA. Russ. J. Gen. Chem. 2008; 78:1794.

143. Volkov KA, Avramenko GV, Negrimovsky VM and Lukyanets EA. Russ. J. Gen. Chem. 2008; 78:1787.





146. Derkacheva VM and Lukyanets EA. Zh. Obshch. Khim. 1980; 50: 2313.

147. Derkacheva VM, Iodko SS, Kaliya OL and Lukya-nets EA. Zh. Obshch. Khim. 1981; 51: 2319.

148. Hall TW, Greenberg S, McArthur CR, Khouw B and Leznoff CC. Nouv. J. Chim. 1982; 6: 653.

149. Oliver SW and Smith TD. Heterocycl. 1984; 22:2047.

150. Meerovich IG, Smirnova ZS, Oborotova NA, Lukyanets EA, Meerovich GA, Derkacheva VM, Polozkova AP, Kubasova IY and Baryshnikov AY. Proc. SPIE Int. Soc. Optic. Engineer. 2005; 5973:59730I.

151. Nemykin VN, Mytsyk VM, Volkov SV and Kobayashi N. J. Porphyrins Phthalocyanines 2000; 4: 551.

152. Nemykin VN and Mytsyk VM. Ukr. Khim. Zh. 2000; 66: 67.

153. Derkacheva VM, Kaliya OL and Lukyanets EA. Zh.Obshch. Khim. 1983; 53: 188.

154. Derkacheva VM and Lukyanets EA. SU 919310, 1981.

155. Kovshev EI, Solov’eva LI, Mikhalenko SA and Lukyanets EA. Zh. Vsesoyuz. Khim. Obs. Mendel.1976; 21: 465.

156. Hu M, Brasseur N, Yildiz SZ, van Lier JE and Leznoff CC. J. Med. Chem. 1998; 41: 1789.

157. Meerovich GA, Lukyanets EA, Yuzhakova OA, Torshina NL, Loschenov VB, Stratonnikov AA, Kogan EA, Vorozhtsov GN, Kunetz AV, Kuvshinov YP, Poddubny BK, Volkova AI and Posypanova AM. Proc. SPIE Int. Soc. Optic. Engineer. 1997;3191: 193.

158. Vorozhtsov GN, Kogan EA, Loshchenov VB, Luzhkov YuM, Lukyanets EA, Meerovich GA, Torshina NL and Yuzhakova OA. RU 2146144, 2000.

159. Meerovich GA, Lukyanets EA, Yuzhakova OA, Kaliya OL, Vorozhtsov GN, Loschenov VB, Torshina NL, Stratonnikov AA and Kogan EA. Proc. SPIE Int. Soc. Optic. Engineer. 1996; 2924:86.

160. Yuzhakova OA and Lukyanets EA. RU 2181735, 2002.

161. Makarov DA, Yuzhakova OA, Slivka LK, Kuznet-sova NA, Negrimovsky VM, Kaliya OL and Lukya-nets EA. J. Porphyrins Phthalocyanines 2007; 11:586.

162. Vorozhtsov GN, Kaliya OL, Kuznetsova NA, Kuz’min SG, Kucherov AG, Lapchenko AS, Luzhkov YuM, Lukyanets EA, Negrimovsky VM,

Slivka LK, Strakhovskaya MG, Yuzhakova OA and Yakubovskaya RI. RU 2282647, 2006.

163. Vorozhtsov GN, Karmakova TA, Luzhkov YuM, Lukyanets EA, Morozova NB, Negrimovsky VM, Pankratov AA, Plyutinskaya AD, Chissov VI, Yuzhakova OA and Yakubovskaya RI. RU 2282646, 2006.

164. Vorozhtsov GN, Kaliya OL, Kuznetsova NA, Luzhkov YuM, Lukyanets EA, Makarov DA, Negri-movsky VM, Slivka LK and Yuzhakova OA. RU 2281953, 2006.

165. Kahl SB and Li J. Inorg. Chem. 1996; 35: 3878. 166. Tsaryova O, Semioshkin AD, Wöhrle D and Bre-

gadze VI. J. Porphyrins Phthalocyanines 2005; 9:268.

167. Giuntini F, Raoul Y, Dei D, Municchi M, Chiti G, Fabris C, Colautti P, Jori G and Roncucci G. Tetra-hedron Lett. 2005; 46: 2979.

168. Friso E, Roncucci G, Dei D, Soncin M, Fabris C, Chiti G, Colautti P, Esposito J, De Nardo L, Rossi CR, Nitti D, Giuntini F, Borsetto L and Jori G. Pho-tochem. Photobiol. Sci. 2006; 5: 39.

169. Zakharkin LI, Guseva VV, Abramov IG and Balagu-rova EV. Russ. J. Gen. Chem. 2000; 70: 1296.

170. Zhizhin KYu, Malinina EA, Goeva LV, Chernyavs-kii AS, Ivanov SV, Lukyanets EA, Solntsev KA and Kuznetsov NT. Dokl. Akad. Nauk 1997; 357: 206.

171. Anderson JS, Bradbrook EF, Cook AH and Linstead RP. J. Chem. Soc. 1938; 1151.


173. Kobayashi N. In The Porphyrin Handbook, Vol. 15, Kadish KM, Smith KM and Guilard R. (Eds.) Aca-demic Press: New York, 2003; pp 161–262.

174. Kovshev EI and Lukyanets EA. Zh. Obshch. Khim.1972; 42: 696.

175. Nemykin VN, Kostromina NA, Subbotin NB and Volkov SV. Izv. Akad. Nauk, Ser. Khim. 1996; 99.

176. Freyer W and Minh LQ. Monatsh. Chem. 1986; 117: 475.

177. Kopranenkov VN, Makarova EA and Lukyanets EA. Zh. Org. Khim. 1981; 17: 358.

178. Shatskaya TA, Gal’pern MG, Skvarchenko VR and Lukyanets EA. Zh. Obshch. Khim. 1987; 57: 2364.

179. Lukyanets EA, Puchnova VA, Mikhalenko SA and Ruzina SG. SU 232963, 1968.

180. Polley R and Hanack M. J. Org. Chem. 1995; 60:8278.

181. Kovshev EI and Lukyanets EA. Zh. Vsesoyuz. Khim. Obshch. Mendel. 1975; 20: 230.

182. Kovshev EI and Lukyanets EA. SU 380641, 1973.

183. Hanack M and Grosshans R. Chem. Ber. 1992; 125:1243.

184. Oksengendler IG, Kondratenko NV, Lukyanets EA and Yagupol’skii LM. Ukr. Khim. Zh. 1980; 46:507.



185. Kovshev EI and Lukyanets EA. Zh. Vses. Khim. Obshch. Mendel. 1975; 21: 231.

186. Goncharova GI, Gal’pern MG and Lukyanets EA. Zh. Obshch. Khim. 1982; 52: 666.

187. Kovshev EI and Lukyanets EA. Zh. Obshch. Khim.1972; 42: 1593.

188. Freyer W and Fink F. Z. Chem. 1988; 28: 67. 189. Donyagina VF and Lukyanets EA. Russ. J. Gen.

Chem. 2005; 75: 795. 190. Bradbrook EF and Linstead RP. J. Chem. Soc. 1938;

1157. 191. Nemykin VN, Subbotin NB, Kostromina NA and

Volkov SV. Ukr. Khim. Zh. 1995; 61: 3. 192. Negrimovsky VM, Bouvet M, Lukyanets EA and

Simon J. J. Porphyrins Phthalocyanines 2001; 5:423.

193. Negrimovsky VM, Bouvet M, Lukyanets EA and Simon J. J. Porphyrins Phthalocyanines 2000; 4:248.

194. Hanack M, Renz G, Straehle J and Schmid S. J. Org. Chem. 1991; 56: 3501.

195. Bradbrook EF and Linstead RP. J. Chem. Soc. 1936; 1739.

196. Blackburn EV and Timmons CJ. J. Chem. Soc. (C) 1970; 175.

197. Mikhalenko SA, Yagodina LA and Lukyanets EA. Zh. Obshch. Khim. 1976; 46: 1598.

198. Kopranenkov VN, Rumyantseva GI and Lukyanets EA. Anilino-kras. Prom. 1974; 2: 6.

199. Kopranenkov VN and Lukyanets EA. Zh. Org. Khim. 1972; 8: 1690.

200. Donyagina VF, Kovshev EI and Lukyanets EA. Russ. J. Gen. Chem. 2006; 76: 654.

201. Scott JE. Histochem. 1972; 30: 215. 202. Linstead RP, Noble EG and Wright JM. J. Chem.

Soc. 1937; 911. 203. Spiessens LIM and Anteunis MJO. Bull. Soc. Chim.

Belg. 1980; 89: 205. 204. Danzig MJ, Liang CY and Passaglia E. J. Am. Chem.

Soc. 1963; 85: 668. 205. Puchnova VA and Lukyanets EA. Khim. Geterotsikl.

Soedin. 1970; 327. 206. Gal’pern EG, Lukyanets EA and Gal’pern MG. Izv.

Akad. Nauk SSSR, Ser. Khim. 1973; 1976. 207. Rothkopf HW, Wöhrle D, Mueller R and Kossmehl

G. Chem. Ber. 1975; 108: 875. 208. Kudrevich SV, Gal’pern MG, Lukyanets EA and

van Lier JE. Can. J. Chem. 1996; 74: 508. 209. Gal’pern MG, Lukyanets EA, Shutova AM and

Ruzina SG. SU 228691, 1968. 210. Gal’pern MG and Lukyanets EA. Khim. Geterots.

Soed. 1971; 7: 280. 211. Cook AH and Linstead RP. J. Chem. Soc. 1937;

929. 212. Marinina LE, Mikhalenko SA and Lukyanets EA.

Zh. Obshch. Khim. 1973; 43: 2025.

213. Nemykin VN, Kostromina NA and Volkov SV. Ukr. Khim. Zh. 1997; 63: 3.

214. Nemykin VN, Cherniy VYa and Volkov SV. Ukr. Khim. Zh. 2001; 67: 3.

215. Khelevina OG, Bubnova AS, Makarova ON, Lukina SA, Vagin SI and Stuzhin PA. Russ. J. Coord. Chem.2006; 32: 451.

216. Camenzind MJ and Hill CL. Inorg. Chim. Acta1985; 99: 63.

217. Fitzgerald JP, Haggerty BS, Rheingold AL, May L and Brewer GA. Inorg. Chem. 1992; 31: 2006.

218. Collman JP, Arnold HJ, Fitzgerald JP and Weissman KJ. J. Am. Chem. Soc. 1993; 115: 9309.

219. Ni Y, Fitzgerald JP, Carroll P and Wayland BB. Inorg. Chem. 1994; 33: 2029.

220. Fitzgerald J, Taylor W and Owen H. Synthesis 1991; 686.

221. Stuzhin PA, Migalov IS, Berezin BD and Lubimov AV. Koord. Khim. 1994; 20: 444.

222. Nemykin VN, Makarova EA, Grosland JO, Hadt RG and Koposov AY. Inorg. Chem. 2007; 46:9591.

223. Kopranenkov VN and Lukyanets EA. Izv. Akad. Nauk, Ser. Khim. 1995; 2320.

224. Nemykin VN, Barrett C, Hadt RG, Rohde GT, Gros-land JO, Galloni P, Bizzari C, Floris B and Loim N. Fifth International Conference on Porphyrins and Phthalocyanines, ICPP-5, Moscow, Russia (July 7–11, 2008), SYM-151.

225. Kopranenkov VN, Goncharova LS and Lukyanets EA. Zh. Obshch. Khim. 1977; 47: 2143.

226. Arabei SM, Kopranenkov VN and Solov’ev KN. Optika Spektros. 1990; 69: 1312.

227. Kobayashi N, Nakajima S and Osa T. Chem. Lett. 1992; 2415.

228. Fitzgerald JP, Lebenson JR, Wang G, Yee GT, Noll BC and Sommer RD. Inorg. Chem. 2008; 47: 4520.

229. Kopranenkov VN, Rumyantseva GI and Lukyanets EA. Zh. Obshch. Khim. 1972; 42: 2586.

230. Weis CD. J. Org. Chem. 1963; 28: 74. 231. Cookson RC and Dance J. Tetrahedron Lett. 1962;

879. 232. Kopranenkov VN, Goncharova LS and Lukyanets

EA. Zh. Obshch. Khim. 1988; 58: 1186. 233. Oliver SW and Smith TD. J. Chem. Soc., Perkin

Trans 2. 1987; 1579. 234. Kopranenkov VN, Mundshtukova ID and Lukyanets

EA. Khim. Geterots. Soed. 1994; 30. 235. Makarova EA, Kopranenkov VN, Shevtsov VK

and Lukyanets EA. Khim. Geterots. Soed. 1994; 1206.

236. Kopranenkov VN, Goncharova LS, Marinina LE and Lukyanets EA. Khim. Geterots. Soed. 1982; 1645.

237. Kopranenkov VN, Goncharova LS and Lukyanets EA. Zh. Org. Khim. 1979; 15: 1076.



238. Kondratenko NV, Movchun VN, Kopranenkov VN, Lukyanets EA and Yagupol’skii LM. Zh. Org. Khim.1992; 28: 2149.

239. Kopranenkov VN and Goncharova LS. Zh. Vse-soyuz. Khim. Obshch. Mendel. 1981; 26: 456.

240. Kopranenkov VN, Goncharova LS and Lukyanets EA. Zh. Obshch. Khim. 1979; 49: 1408.

241. Troitskaya VI, Trushanina LI, Kopranenkov VN, Lukyanets EA and Yagupol’skii LM. Zh. Org. Khim.1989; 25: 571.

242. Sausen GN, Engelhardt VA and Middleton WJ. J. Am. Chem. Soc. 1958; 80: 2815.

243. Rappoport Z (Ed.). The Chemistry of the Cyano Group, Itersci. Publ.: London, 1970.

244. Nemykin VN, Maximov AY and Koposov AY. Organomet. 2007; 26: 3138.

245. Nemykin VN and Kobayashi N. Chem. Commun.2001; 165.

246. Makarova EA and Lukyanets EA. J. Porphyrins Phthalocyanines 2009; 13: 1–15.

247. Lebedev OL, Lukyanets EA and Puchnova VA. Optika Spektr. 1971; 30: 640.

248. Gal’pern MG, Gorbachev VA, Katulin VA, Lebedev OL, Lukyanets EA, Mekhryakova NG, Mizin VM, Nosach VYu, Petrov AL and Semenovskaya GG. Kvant. Elektr. 1975; 2: 2531.

249. Batashev SP, Gal’pern MG, Katulin VA, Lebedev OL, Lukyanets EA, Mekhryakova NG, Mizin VM and Nosach VYu. Kvant. Elektr. 1979; 6: 2652.

250. Balenko VG, Kaliya OL, Kirkin AN, Lukyanets EA, Mizin VM, Popov MB, Teplyakova NV, Tomi-lova LG and Chernykh EV. Kvant. Elektr. 1989; 16:615.

251. Wöhrle D, Meyer G. and Wahl B. Macromol. Chem. 1980; 181: 2127.

252. Kobayashi N, Ogata H, Nonaka N and Lukyanets EA. Chem. Eur. J. 2003; 9: 5123.

253. Mikhalenko SA, Korobkova EV and Lukyanets EA. Zh. Obshch. Khim. 1970; 40: 400.


255. Nemykin VN, Kobayashi N, Nonomura T and Lukyanets EA. Chem. Lett. 2000; 184.

256. Leznoff CC, Hu M, McArthur CR and Qin Y. Can.J. Chem. 1994; 72: 1990.

257. Kobayashi N, Sasaki N, Higashi Y. and Osa T. Inorg. Chem. 1995; 34: 1636.

258. Cook MJ, Dunn AJ, Howe SD and Thomson AJ. J. Chem. Soc. Perkin Trans. 1 1988; 2453.

259. Rihter BD, Kenney ME, Ford WE and Rodgers AJ. J. Am. Chem. Soc. 1990; 112: 8064.

260. Snow AW and Jarwis NL. J. Am. Chem. Soc. 1984;32: 4706.

261. Haisch P and Hanack M. Synthesis 1995; 1251. 262. Azuma N, Kitahara K, Motoi T, Hokita S and Nishi

H. J. Heterocyc. Chem. 1993; 30: 225. 263. Langfi eld HW. Rev. de Chim. 1979; 30: 740. 264. Konami H and Hatano M. Chem. Lett. 1988;

1359. 265. Freyer W and Minh Q. J. Pract. Chem. 1987; 329:

365. 266. Kopranenkov VN and Lukyanets EA. Zh. Obshch.

Khim. 1971; 41: 2341. 267. Freyer W, Stiel H, Hild M, Teuchner K and Leupold

D. Photochem. Photobiol. 1997; 66: 596. 268. Freyer W. Z. Chem. 1986; 26: 216. 269. Gal’pern MG and Lukyanets EA. Anil.-kras. prom.

1972; 5: 22. 270. Gal’pern MG and Lukyanets EA. Zh. Obshch. Khim.

1971; 41: 2549. 271. Gal’pern MG and Lukyanets EA. Zh. Obshch. Khim.

1969; 39: 2536. 272. Gal’pern MG and Lukyanets EA. Zh. Vsesoyuz.

Khim. Obshch. Mendel. 1967; 12: 474. 273. Linstead RP and Whalley M. J. Chem. Soc. 1952;

4839. 274. Kopranenkov VN and Rumyantseva GI. Russ.

J. Gen. Chem. 1975; 45: 1555.


DOI: 10.1142/S1088424610001738



INTRODUCTION

Exploiting the combined properties of metals and hydrogen bonds is a common theme in the chemistry of biological species. For example, the active sites of met-alloenzymes usually exhibit several hydrogen-bonding interactions, which enhance the properties of these natu-ral metal complexes [1]. It is not surprising, therefore, that there is a considerable interest in creating synthetic metal complexes that take advantage of complementary hydrogen-bonding features; such systems are attractive since they could provide the basis for new materials, more effective catalysts or reagents, and novel metal-containing therapeutics. Indeed, over the past few years, as the result of both accident and rational design, several synthetic systems have been reported that underscore how hydrogen-bonding interactions within a metal complex can exert important structural and chemical effects [2]. Nevertheless, additional examples of such complexes are

needed if the potential of this dual functional approach is to be fully realized. We believe that one way this can be done is via the use of core-expanded porphyrin analogs.

Core-expanded porphyrins can be considered as a special class of expanded porphyrins, a class of mac-rocycles that possess an inherently larger core than the better-studied naturally-occurring tetrapyrrole pigments. Over the last three decades, expanded porphyrins have become increasingly recognized as useful metal-com-plexing ligands that have helped to expand the frontiers of “porphyrin-like” coordination chemistry [3]. For instance, expanded porphyrins have been used to stabi-lize non-labile lanthanide complexes [4] and to produce complexes containing multiple cations [5]. Expanded porphyrins also display a rich array of hydrogen-bond-ing donors, typically pyrrolic NH protons, a feature that is largely responsabile for their well-recognized anion binding behavior [6]. Owing to the combination of these attributes, expanded porphyrins are promising platforms for the creation of synthetic metal complexes that incor-porate hydrogen-bonding interactions. Studies in the lit-erature involving very different classes of materials lead

Syntheses and structural studies of 5-pentamethyl-cyclopentadienyl rhodium(III) and iridium(III) complexes of a Schiff-base expanded porphyrin

Luciano Cuesta, Vincent M. Lynch and Jonathan L. Sessler*

Department of Chemistry and Biochemistry and Institute for Cellular and Molecular Biology, University of Texas, 1 University Station A5300, Austin, TX 78712-0165, USA

Received 7 January 2009Accepted 27 January 2009

ABSTRACT: Reported here is the synthesis of new binuclear rhodium(III) and iridium(III) semi-sandwich complexes of a Schiff-base expanded porphyrin. Single crystals of these new complexes were subject to X-ray diffraction analysis. The resulting structures revealed that the Schiff-base macrocycle adopts a V-shape in which two {( 5-C5Me5)MCl} (M = Rh and Ir) fragments are accommodated within the macrocyclic pocket. The coordination environment of the metal centers is typical to that of “piano stool”-type complexes. The X-ray analyses and complementary NMR studies (carried out in CD2Cl2)provide evidence for the existence of strong intramolecular hydrogen-bonding interactions between the pyrrolic NH protons and the chloride counteranion both in solution and in the solid state.

KEYWORDS: Schiff-base expanded porphyrin, rhodium(III) and iridium(III) complexes, pentamethyl-cyclopentadienyl half-sandwich complexes, intramolecular hydrogen bonds.


*Correspondence to: Jonathan L. Sessler, email: [email protected], fax: +1 512-471-7550


42 L. CUESTA ET AL.

us to suggest that complexes incorporating this kind of non-covalent interaction could play a role as functional elements in organometallic chemistry [7].

Recently, we reported the syntheses of arene and cyclopentadienyl ruthenium “piano stool”-type com-plexes based on the Schiff-base expanded porphyrin 1(see Scheme 1) [8], organometallic species characterized by unusual hydrogen bond interactions between the pyr-role NH protons and a metal-bound dioxygen or chloride ligand. This original study did not, however, extend to the corresponding rhodium and iridium half-sandwich com-plexes, even though these latter elements have proved to be extremely useful in catalysis [9], the production and storage of hydrogen [10], electrocatalysis [11], and supramolecular chemistry [12]. Accordingly, we have now prepared, and report here, the synthesis of new dica-tionic binuclear pentamethylcyclopenta dienyl rhodium(III) and iridium(III) Schiff-base expanded porphyrin com-plexes; as with the previously reported ruthenium com-plexes, these new species display strong intramolecular hydrogen-bonding interactions.

RESULTS AND DISCUSSION

As depicted in Scheme 1 and detailed in the Experimen-tal section, the dinuclear complexes [{ 5-C5Me5)M(μ-Cl)-Cl}2] (M = Rh and Ir) undergo a bridge cleavage reaction with the Schiff-base macrocyle 1, in the presence of two equivalents of NaBAr′4 (Ar′: 3,5-bis(trifl uoromethyl)-phenyl), to yield the dicationic complexes 2 and 3 after chromatographic purifi cation. These complexes, soluble in most organic solvents, were isolated as orange air-stable crystalline solids. Both new compounds were characterized by NMR spectroscopy (1H and 13C) and microanalysis. In addition, their structures were eluci-dated using single X-ray diffraction methods.

The ORTEP drawings of the cations present in com-pounds 2 and 3 are shown in Figs 1 and 2, respectively.

The crystal structures revealed that the macrocyclic pocket is large enough to stabilize two {( 5-C5Me5)MCl}(M = Rh, Ir) units simultaneously, affording binuclear complexes in which the metallic centers are coordinated to two nitrogens of the Schiff-base ligand, one chloride anion, and the pentamethylcyclopentadienyl ligand in a

5-fashion ( 5-C5Me5); leading to a “piano stool” struc-ture, wherein the iminic nitrogen and the chloride ligands provide the “legs” and the C5Me5 moiety resemble the “seat”. In both complexes, the macrocycle adopts a char-acteristic V-shape conformation where one of the pen-tamethylcyclopentadienyl ligands points to the interior face of the cavity, and the other one is orientated to the “outside” face. As a consequence, the structure of these complexes is asymmetrical in the solid state.

In complex 2, the bond distances between the rhodium and the chelated nitrogen atoms (2.141(6) Å, 2.148(6) Å, 2.119(6) Å and 2.124(6) Å) are similar to those seen in complexes of the type [(N-N)( 5-C5Me5)RhCl] [13]. Likewise, the Rh-Cl (2.3862(19) Å, 2.4418(19) Å) dis-tances, for which a signifi cant disparity ascribed to a solid state effect is seen, fall within the range observed for other complexes containing {( 5-C5Me5)RhCl} frag-ments [13]. The Ir-N (2.131(7) Å, 2.135(7) Å, 2.098(7) Å, 2.108(8) Å) and Ir-Cl (2.389(2) Å, 2.410(2) Å) bond distances present in complex 3 also fall within the range typically observed for complexes containing a {( 5-C5Me5)IrCl} moiety [14]. The coordination mode of the pentamethylcyclopentadienyl ligands is clearly

5, as evidenced by the planarity of the rings, as well as by the fact that there is no signifi cant difference in the C-C bonds lengths within these latter rings (all are about 1.42 Å), and that the metal atom-to-ring carbon distances are nearly equal.

Interestingly, the chloride ligands are engaged in strong intramolecular hydrogen bonds with the pyrrolic NH protons, as refl ected in the solid state by geometries and distances consistent with such interactions. The N···Cl

Scheme 1. Synthesis of complexes 2 and 3


RHODIUM(III) AND IRIDIUM(III) COMPLEXES OF A SCHIFF-BASE EXPANDED PORPHYRIN 43

distances, which fall in the range 3.087(6)–3.108(7) Å for 2 and 3.047(7)–3.097(8) Å for 3, are exceptionally short when compared to the corresponding distances seen in previous crystallographically characterized adducts of hydrogen-bonded chloride-containing species [15]. Nev-ertheless, no elongation of the M-Cl bonds is observed, as noted above.

In the 1H NMR spectra of 2 (see Fig. 3) and 3, recorded at room temperature in CD2Cl2, the signals for the methyl groups of the C5Me5 ligands appear as two large non-equivalent singlets (at 1.50 and 0.75 ppm in complex 2,and 1.50 and 0.80 ppm in complex 3, respectively). This observation is ascribed to the presence of asymmetrical complexes. Consistent with this interpretation, two differ-ent set of signals are also seen for the methyl groups, the imine protons, and the pyrrolic NH protons. The inferred lack of symmetry is also evident in the 13C NMR spec-trum. Accordingly, the NMR spectroscopic data provide support for the conclusion that the structures of 2 and 3seen in the solid state are retained in solution. Presum-ably, the presence of the bulky 5-C5Me5 ligands serves to lock the conformation of the two dicationic complexes 2and 3, as previously observed for the analogueous ruthe-nium derivative [8].

The iminic protons of complex 2 appear as doublets (JRh-H = 3.1 and 2.8 Hz) in the 1H NMR spectrum due to the coupling with the Rh center. Coupling is also seen with the carbon atoms of the C5Me5 ring bound to the metal center (JRh-C = 8.5 and 8.7 Hz) in the 13C NMR spec-trum. Separate from this, the pyrrolic NH resonance is found to undergo a considerable shift to lower fi eld in the complexes (δ = 11.70 and 11.60 ppm in 2, and 11.50 and 11.30 ppm in 3), relative to the free ligand (δ = 9.83 ppm for 1), indicating the presence of strong hydrogen-bonding interactions between these atoms and the anionic chloride ligands. Again, taken in concert, these observa-tions are consistent with the coordination mode seen in the solid state being retained in solution.

EXPERIMENTAL

General

Prior to use, all glassware was soaked in KOH-sat-urated isopropyl alcohol for ca. 12 h and then rinsed with water and acetone before being thoroughly dried. Dichloromethane was freshly distilled from CaH2.

Fig. 1. Top and side views of the cation present in complex 2 showing a partial atom labeling scheme. Displacement ellipsoids are scaled to the 50% probability level. The BAr′4 counteranions and most hydrogens atoms have been omitted for clarity. Dashed lines are indicative of H-bonding interactions

Fig. 2. Top and side views of the cation present in complex 3 showing a partial atom labeling scheme. Displacement ellipsoids are scaled to the 50% probability level. The BAr′4 counteranions and most hydrogens atoms have been omitted for clarity. Dashed lines are indicative of H-bonding interactions


44 L. CUESTA ET AL.

n-pentane was stirred over concentrated H2SO4 for more than 24 h, neutralized with K2CO3, and distilled from CaH2. The complexes [{ 5-C5Me5)M(μ-Cl)Cl}2] (M=Rh or Ir) were purchased commercially (Strem) and used as received, the macrocycle 1 [16] and NaBAr′4 (Ar′:3,5-bis(trifl uoromethyl)phenyl) [17] were prepared fol-lowing reported procedures. Solutions were stirred mag-netically. Nuclear magnetic resonance (NMR) spectra were obtained on a Varian Mercury 400 MHz. Elemen-tal analyses were performed by Midwest Microlabs Inc., Indianapolis, IN.

Synthesis

Preparation of complex 2. [{ 5-C5Me5)Rh(μ-Cl)Cl}2] (31 mg, 0.05 mmol) and NaBAr′4 (88 mg, 0.10 mmol) were added to a suspension of the free base mac-rocycle 1 (30.0 mg, 0.05 mmol) in dry CH2Cl2 (30 mL) under an argon atmosphere. The mixture was stirred for 8 h affording a dark orange solution. The solvent was evaporated under reduced pressure. The air stable orange solid obtained in this way was purifi ed by column chro-matography (silica gel, 10% MeOH/CH2Cl2). The fi rst orange fraction was collected and the resulting product was crystallized from CH2Cl2/n-pentane to yield 104 mg of complex 2 (71%). mp > 200 °C. Anal. calcd. for 2.C122H90B2Cl2F48N8Rh2: C, 50.91; H, 3.15; N, 3.89. Found: C, 50.83; H, 3.37; N, 4.00. 1H NMR (400 MHz; CD2Cl2;Me4Si): H, ppm 11.70 (s, 2H, NH), 11.60 (s, 2H, NH), 8.68 (d, JRh-H = 3.1 Hz, 2H, CHN), 8.28 (d, JRh-H = 2.8 Hz, 2H, CHN), 7.72 (s, 16H, Ho BAr′4), 7.63 (m, 2H, Ar), 7.55 (s, 8H, Hp BAr′4), 7.50 (m, 2H, Ar), 7.34 (m, 4H, Ar), 7.09 (m, 2H, pyrrolic CH), 7.03 (m, 2H, pyrrolic CH), 6.65 (m, 2H, pyrrolic CH), 6.32 (m, 2H, pyrrolic CH), 2.21 (s, 6H, CH3), 1.86 (s, 6H, CH3), 1.50 (s, 15H, C5Me5), 0.75 (s, 15H, C5Me5).

13C NMR (100.6 MHz,

CD2Cl2): C, ppm 161.9 (q, J = 50.1 Hz), 156.3, 154.5, 153.1, 151.5, 150.5, 148.1, 134.9, 132.4, 131.1, 130.2, 129.2, 128.9, 128.8 (q, J = 31.5 Hz), 124.5 (q, J = 272.7 Hz), 123.1, 118.1, 117.7, 117.2, 111.0, 97.6 (d, J = 8.5), 97.2 (d, J = 8.7), 37.2, 31.0, 27.0, 9.1, 7.7.

Preparation of complex 3. [{ 5-C5Me5)Ir(μ-Cl)Cl}2](40 mg, 0.05 mmol) and NaBAr′4 (88 mg, 0.10 mmol) were added to a suspension of the free-base macrocycle 1 (30.0 mg, 0.05 mmol) in dry CH2Cl2 (30 mL) under an argon atmosphere. After stirring for 8 h, a bright orange-yellow solution was formed. Evaporation of the solvent under reduced pressure and purifi cation by column chro-matography (silica gel, 10% MeOH/CH2Cl2) afforded the compound 3 as an orange-yellow air-stable solid. Crystals grew as orange-yellow cubes by slow difus-sion of n-pentane into a CH2Cl2 solution of compound 3. Yield 128 mg (84%), mp > 200 °C. Anal. calcd. for 3.C122H90B2Cl2F48Ir2N8: C, 47.95; H, 2.97; N, 3.67. Found: C, 47.79; H, 3.08; N, 3.67. 1H NMR (400 MHz; CD2Cl2;Me4Si): H, ppm 11.50 (s, 2H, NH), 11.38 (s, 2H, NH), 8.75 (s, 2H, CHN), 8.34 (s, 2H, CHN), 7.72 (s, 16H, Ho

BAr′4), 7.64 (m, 2H, Ar), 7.55 (s, 8H, Hp BAr′4), 7.48 (m, 2H, Ar), 7.32 (m, 4H, Ar), 7.09 (m, 2H, pyrrolic CH), 7.04 (m, 2H, pyrrolic CH), 6.66 (m, 2H, pyrrolic CH), 6.33 (m, 2H, pyrrolic CH), 2.15 (s, 6H, CH3), 1.86 (s, 6H, CH3), 1.50 (s, 15H, C5Me5), 0.80 (s, 15H, C5Me5).13C NMR (100.6 MHz, CD2Cl2): C, ppm 161.9 (q, J =50.1 Hz), 154.5, 153.4, 153.1, 151.6, 149.2, 134.9, 135.0,133.3, 130.9, 128.9, 128.8 (q, J = 31.5 Hz), 125.9, 124.5 (q, J = 272.7 Hz), 122.5, 117.5, 116.4, 112.4, 111.3, 89.6, 89.1, 37.0, 30.9, 29.9, 8.9, 7.3.

X-ray structure determinations

The data were collected on a Nonius Kappa CCD diffractometer using a graphite monochromator with

Fig. 3. 1H NMR spectrum of complex 2 recorded at room temperature in CD2Cl2. Residual dichloromethane and water signals are indicated by *


RHODIUM(III) AND IRIDIUM(III) COMPLEXES OF A SCHIFF-BASE EXPANDED PORPHYRIN 45

Mo-Kα radiation ( = 0.71073 Å). The data were col-lected at 153 K using an Oxford Cryostream low tem-perature device. Data reduction were performed using DENZO-SMN [18]. The hydrogen atoms on carbon were calculated in ideal positions with isotropic displace-ments parameters set to 1.2 × Ueq of the attached atom (1.5 × Ueq for methyl hydrogen atoms). The function, Σw(|Fo|

2 - |Fc|2)2, was minimized, where w = 1/[(σ(Fo))

2 +(0.0562*P)2 + (18.8245*P)] and P = (|Fo|

2 + 2|Fc|2)/3.

The data were checked for secondary extinction but no correction was necessary. In both cases some of the trif-luoromethyl groups were disordered. In addition, several fl uorine atoms exhibited highly anisotropic displacement parameters. The disordered CF3 groups were all modeled alike. The C-F bond lengths and the F-C-C bond angles were restrained to be equivalent for all CF3 groups. For a given CF3 group, the site occupancy factor for one com-ponent of the disorder was assigned the variable x, while the site occupancy factor for the alternate component was set to (1-x). A common isotropic displacement parameter was refi ned for all six F atoms. Once the variable x was refi ned, the site occupancies were fi xed and the displace-ment parameters were refi ned. Anisotropic displace-ment parameters for all fl uorine atoms were refi ned with restraints to keep them approximately isotropic. One of the dichloromethane molecules was also disordered in both complexes. The disorder was the result of a rotation about one of the Cl-C bonds. The geometry of the dichlo-romethane was restrained to be equivalent to the ordered dichloromethane molecule throughout the refi nement. The hydrogen atoms associated with these molecules were not included in the refi nement model.

Crystallographic data have been deposited at the CCDC, and copies can be obtained on request, free of charge, by quoting the publication citation and the depo-sition numbers indicated below via www.ccdc.cam.ac.uk/conts/retrieving.html or from the Cambridge Crystallo-graphic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44 1223-336-033 or email: [email protected]).

Compound 2. C124H94B2Cl6F48Rh2N8. The data crystal was cut from a larger crystal and had approximate dimen-sions of 0.35 × 0.17 × 0.07 mm; monoclinic, space group P21, a = 13.5078(3) Å, b = 26.2580(8) Å, c = 18.6348(6) Å, α = 90º, β = 104.457(2)°, γ = 90º. V = 6400.2(3) Å3.Z = 2, ρcalcd = 1.582 Mg/m3. μ = 0.505 mm-1. F(000) =3056. A total of 480 frames of data were collected using ω-scans with a scan range of 0.8° and a counting time of 85 seconds per frame. A total of 26,327 refl ections were measured, 26,327 unique. The structure was refi ned on F2

to 0.1425, with R(F) equal to 0.0689 and a goodness of fi t, S = 1.054. The deposition number is 709220.

Compound 3. C124H94B2Cl6F48Ir2N8. The data crystal was cut from a larger crystal and had approximate dimen-sions of 0.20 × 0.13 × 0.10 mm; monoclinic, space group P21, a = 13.5372(2) Å, b = 26.2793(6) Å, c = 18.6279(4) Å, α = 90º, β = 104.594(2)°, γ = 90º. V = 6414.3(2) Å3.

Z = 2, ρcalcd = 1.671 Mg/m3. μ = 2.320 mm-1. F(000) =3184. A total of 397 frames of data were collected using ω-scans with a scan range of 1.1° and a counting time of 102 seconds per frame. A total of 67,201 refl ections were measured, 27,926 unique (Rint = 0.0702). The structure was refi ned on F

2 to 0.135, with R(F) equal to 0.0603

and a goodness of fi t, S = 1.11. The deposition number is 709221.

CONCLUSION

In summary, the Schiff-base tetrapyrrolic macro-cycle 1 has been used to prepare two new organome-tallic binuclear “piano stool”-type complexes 2 and 3,which contain Rh(III) and Ir(III) centers, respectively. Both complexes display strong intramolecular hydrogen-bonding interactions, both in CD2Cl2 solution and in the solid state. The results reported here serve to demonstrate that Schiff-base expanded porphyrins, which allow metal centers to be coordinated in close proximity to a rich array of hydrogen-bond donors, are potential platforms for the synthesis of novel complexes that combine the features of traditional transition metal coordination and ancillary hydrogen-bonding interactions.

Acknowledgements

This work was supported by the National Sci-ence Foundation (grant no. CHE-0749571 to J. L. S.), the Robert A. Welch Foundation (grant no. F-1018 to J. L. S.), and the Spanish Ministry and Fulbright Founda-tion (postdoctoral support to L. C.).

Supporting information

Details of the crystallography (Table S1) are given in the supplementary material. This material is available free of charge via the Internet at http://www.worldscinet.com/jpp/jpp.shtml.

REFERENCES

1. Borovik AS. Acc. Chem. Res. 2005; 38: 54–61, and references therein.

2. a) Mareque-Rivas JC. Curr. Org. Chem. 2007; 11:1434–1449. b) Natale D and Mareque-Rivas JC. Chem. Commun. 2008; 425–437. c) Shook RL and Borovik AS. Chem. Commun. 2008; 6095–6107.

3. a) Shimizu S and Osuka A. Eur. J. Inorg. Chem.2006; 1319–1335. b) Sessler JL and Tomat E. Acc.Chem. Res. 2007; 5: 371–379.

4. Sessler JL, Hemmi G, Mody TD, Murai T, Burrel A and Young SW. Acc. Chem. Res. 1994; 27: 43–50.

5. See reference 3, and for example: a) Givaja G, Volpe M, Leeland JW, Edwards MA, Young TK, Darby SB, Reid SD, Blake AJ, Wilson C, Wolowska J, McInnes EJL, Schröder M and Love JB. Chem.Eur. J. 2007; 13: 3707–3723. b) Sessler JL, Tomat E


46 L. CUESTA ET AL.

and Lynch VM. J Am. Chem. Soc. 2006; 128: 4184–4185. c) Tomat E, Cuesta L, Lynch VM and Sessler JL. Inorg. Chem. 2007; 46: 6224–6226.

6. a) Sessler JL and Weghorn SJ. Expanded, Con-tracted and Isomeric Porphyrins, Pergamon: New York, 1997. b) Sessler JL and Davis JM. Acc. Chem. Res. 2001; 34: 989–997.

7. a) Lee D-H, Patel DP, Clot E, Einstein O and Crabtree RH. Chem. Commun. 1999; 297–298. b) Gruet K, Crabtree RH, Lee D-H, Liable-sands L and Rheingold AL. Organometallics 2000; 19:2228–2232. c) Yamakawa M, Yamada I and Noy-ori R. Angew. Chem. Int. Ed. 2001; 40: 2818–2821. d) Grotjahn DB and Lev DA. J. Am. Chem. Soc. 2004; 126: 12232–12233. e) Appelhans LN, Zucca-ccia D, Kovacevic A, Chianese AR, Miecznikowski JR, Macchioni A, Clot E, Eisenstein O and Crabtree RH. J. Am. Chem. Soc. 2005; 127: 16299–16311. f ) Ohkuma T, Utsumi N, Tsutumi K, Murata K, Sandoval C and Noyori R. J. Am. Chem. Soc. 2006;128: 8724–8725. g) Martínez-García H, Morales D, Pérez J, Coady DJ, Bielwaski CW, Gross DE, Cuesta L, Marquez M and Sessler JL. Organome-tallics 2007; 26: 6511–6514. h) Pérez J and Riera L. Chem. Commun. 2008; 533–534.

8. Cuesta L, Tomat E Lynch VM and Sessler JL. Chem.Commun. 2008; 3744–3746.

9. Selected examples: a) Christine LH, Buriez O, Kerr JB and Fish RH. Angew. Chem. Int. Ed. 1999; 38:1429–1432. b) Kölle U and Fränzl H. Monatsheftefür Chemie 2000; 131: 1321–1326. c) Ogo S, Maki-hara N, Kaneko Y and Watanabe Y. Organometallics 2001; 20: 4903–4910. d) Abura T, Ogo S, Watanabe Y and Fukuzumi S. J. Am. Chem. Soc. 2003; 125:4149–4154. e) Gabrielson A, Leeuwen PV and Kaim W. Chem. Commun. 2006; 47: 4926–4927. f ) Tejel C and Ciriano MA. Top. Organomet. Chem. 2007; 22: 97–124. g) Govindaswamy P, Canivet J, Therrien B, Süss-Fink G, Stepnicka P and Ludvík J. J. Organomet. Chem. 2007; 692: 3664–3675. h) Himeda Y, Onozawa-Komatsuzaki N, Miyazawa S, Sugihara H, Hirose T and Kasuga K. Chem. Eur. J. 2008; 14: 11076–11081.

10. a) Ziessel R. Angew. Chem. Int. Ed. 1991; 30: 844–847. b) Himeda Y, Onozawa-Komatsuzaki N, Sugi-hara H, Arakawa H and Kasuga K. Organometallics 2004; 23: 1480–1483. c) Ogo S, Kabe R, Hayashi H, Harada R and Fukuzumi S. Dalton Trans. 2006;4567–4663. d) Himeda Y, Onozawa-Komatsuzaki N, Sugihara H and Kasuga K. Organometallics 2007; 26: 702–712. e) Fukuzumi S. Eur. J. Inorg. Chem. 2008; 1351–1362.

11. a) Steckhan E, Herrman S, Ruppert R, Dietz E, Frede M and Spika E. Organometallics 1991; 10:1568–1577. b) Kaim W, Reinhardt R, Waldhör E and Fiedler J. J. Organomet. Chem. 1996; 524:195–202. c) Caix C, Chardon-Noblat S, Deronzier A, Moutet J-C and Tingry S. J. Organomet. Chem. 1997; 540: 105–111. d) Song H-K, Lee SH, Won K, Park JH, Kim JK, Lee H, Moon S-J, Kim DK and Park CB. Angew. Chem. Int. Ed. 2008; 47: 1749–1752. e) Hildebrand F, Kohlmann C, Franz A and Lütz S. Adv. Synth. Catal. 2008; 350: 909–918.

12. a) Amouri H, Rager NM, Cagnol F and Vaiseer-man J. Angew. Chem. Int. Ed. 2001; 40: 3636–3638. b) Severin K. Coord. Chem. Rew. 2003; 245: 3–10. c) Severin K. Chem. Commun. 2006; 3859–3867. d) Mirtschin S, Krasniqui E, Scopelliti R and Sev-erin K. Inorg. Chem. 2008; 47: 6375–6381. e) Han Y-F, Lin Y-J, Weng L-H, Berke H and Jin G-X. Chem. Commun. 2008; 350–352.

13. Selected examples: a) Youinou MT and Ziessel R. J. Organomet. Chem. 1989; 363: 197–208. b) Aneetha H, Zacharias PS, Srinivas B, Lee GH and Wang Y. Polyhedron 1999; 18: 1710–1716. c) Brunner H, Köllnberger T and Burgemeister MZ. Polyhedron 2000; 19: 299–307. d) Govindas-wamy P, Mozharivskyj YA and Kollipara MR. Poly-hedron 2005; 24: 1710–1716. e) Scharwitz MA, Ott I, Geldmacher Y, Gust R and Sheldrick WS. J. Organomet. Chem. 2008; 693: 2299–2309. Also see reference 9g.

14. See references 9c,d,g, 10c, and 16a–d. 15. Selected examples: a) Sessler JL, Mody TD, Ford

DA and Lynch V. Angew. Chem., Int. Ed. Engl. 1992; 31: 452–455. b) Gale PA, Sessler JL, Král V and Lynch VM. J. Am. Chem. Soc. 1996; 118: 5140–5141. c) Kang SO, Llinares JM, Powel D, Vander-velde D and Bowman-James K. J. Am. Chem. Soc. 2003; 125: 10152–10153. d) Curiel D, Beer PD, Paul RL, Cowley A, Sambrook MR and Szemes F. Chem. Commun. 2004; 1162–1163. e) Arroyo M, Miguel D, Villafañe F, Nieto S, Pérez J and Riera L.Inorg. Chem. 2006; 45: 7018–7026.

16. Sessler JL, Cho WS, Dudek SP, Hicks L, Lynch VM and Huggins MT. J. Porphyrins Phthalocyanines 2003; 7: 97–104.

17. Nishida H, Takada N, Yoshimura M, Sonoda T and Kobayashi T. Bull. Chem. Soc. Jpn 1984; 57:2600–2604.

18. Macromolecular Crystallography. Part A. Methods in Enzymology, Vol. 276, Carter CW Jr and Sweet RM. (Eds.) Academic Press: New York, 1997.


DOI: 10.1142/S1088424610001726



INTRODUCTION

Phthalocyanines, which are used as important green-to-blue colorants, have attracted attention as functional chromophores for various applications including organic charge carriers in photocopiers, laser light absorbers in data storage systems, photoconductors in photovoltaic cells, electrochromic displays [1–6], and non-colored transparent fi lms in the visible region. For these applica-tions, absorption maxima of phthalocyanines, also called Q bands, are best if moved to the near-infrared region. The Q band absorption at around 650 nm is attributed to the π-π∗ transition. In general, the Q band of phthalocyanines

can be moved to longer wavelengths through extension of π electron conjugation systems such as naphthalocyanines, anthracyanines and anthraquinocyanines [7]. For naph-thalocyanines and anthracyanines, syntheses are diffi cult and yields are low. Therefore, it is necessary to continue investigate the synthesis of novel phthalocyanines of which the Q band shows near-infrared absorption, for example, to introduce an electron-donating substituent at a peripheral (2,3,9,10,16,17,13,24) position [1–6].

Cook and co-workers reported that 1,4,8,11, 15,-18,22,25-octakis(hexylsulfanyl)phthalocyanines were moved in the Q band around 780 nm [8]. The substituents located at a non-peripheral (1,4,8,11,15,18,22,25) posi-tion can show bathochromic effects through comparison with the peripheral (2,3,9,10,16,17,23,24) position. Then, alkyl substituents in phthalocyanines linked through oxy-gen and sulfur atoms have the important effect of moving

Synthesis of near-infrared absorbed metal phthalocyanine with S-aryl groups at non-peripheral positions

Keiichi Sakamoto*a , Eiko Ohno-Okumuraa,b, Taku Katoc and Hisashi Sogaa

a Department of Applied Molecular Chemistry, College of Industrial Technology, Nihon University, 1-2-1 Izumi-cho, Narashino, Chiba 275-8575, Japanb Research Institute of Chemical Science, Technology and Education, 8-37-1 Narashinodai, Funabashi,Chiba 274-0063, Japanc Electronic Materials Research Laboratories, Nissan Chemical Industry, 488-6 Suzumi-cho, Funabashi,Chiba 275-0052, Japan

Received 7 November 2008Accepted 28 January 2009

ABSTRACT: Phthalocyanines are used as various applications, including as organic charge carriers in photocopiers, laser light absorbers in data storage systems, photoconductors in photovoltaic cells and electrochromic displays, and non-colored transparent fi lm in visible region. The absorption maxima of phthalocyanines are best if moved near the infrared region for these applications. The Q band of phthalocyanines can be moved to bathochromic effects through extension of a π conjugation system such as naphthalocyanines and anthracyanines. Yields of naphthalocyanines and anthracyanines are, however, low. To solve the problem, novel metal phthalocyanines having non-peripheral S-aryl substituent were synthesized. The novel phthalocyanines show a high strain structure and no liquid crystal property. The target compounds were synthesized: 15 phthalocyanines from 2,3-dicyanohydroquinone in 3 steps via 1,2-dicyanobenzene-3,6-bis(trifl uorate) and 1,2-dicyanobenzene-3,6-thiophenols. The Q bands of obtained compounds appeared in the near-infrared region. In particular, lead 1,4,8,11,15,18,22,25-octakis(thiophenylmethyl)phthalocyanine shows a Q band at 857 nm. Furthermore, non-colored transparent fi lms in the visible region can be produced.

KEYWORDS: S-aryl substituent, near-infrared absorption, non-peripheral position.


*Correspondence to: Keiichi Sakamoto, email: [email protected], tel: +81 47-474-2572, fax: +81 47-474-2579


48 K. SAKAMOTO ET AL.

the Q band absorption to the near-infrared region: 760 and 800 nm [9, 10]. The wavelength of absorption maximum (λmax) of the Q band in 1,4,8,11,15,18,22,25-octakis(hexylsulfanyl)phthalocyanine is dependent upon its central metal [8]. Especially, lead phthalocyanines show longer λmax of the Q band because of their shuttle-cock-shaped molecule. However, Cook’s phthalocyanines were anticipated to show discotic liquid crystal behavior because of their long alkyl substitution. The phthalocya-nines are thought to lack high heat resistance up to 300°C on a solid body. The Cook’s phthalocyanines show red color because the Q band moved to the near-red region accompanying the Soret band [8].

In 2005, Kobayashi and co-workers reported synthe-sis of non-planar 1,4,8,11,15,18,22,25-octa(p-methoxy-phenyl)phthalocyanines, of which the Q band was shifted beyond 800 nm [11]. Although most metal phthalocya-nines have a planar structure, the bathochromic effect of the Q band observed for the octa(p-methoxyphenyl)phtha-locyanines arises from ligand deformation and the elec-tron-donating property of methoxy groups. Kobayashi’s group also reported that 1,4,8,11,15,18,22,25-octapheny-lated phthalocyanines have a high strain structure caused by the steric hindrance of substituents [12].

Near-infrared absorption materials were fi rst reported in patents [12]. Infrared-absorbed phthalocyanines were obtained as a mixture of a constitutional isomer or regio-isomers. The reason for the mixture of isomers is as follows: raw materials were added onto the mode of prep-aration of materials to decrease manufacturing process-ing; the selected molecular structures were unfortunately designed to include isomers and crude materials that can-not be easily separated into isomers. Even materials of the same series of phthalocyanine isomers are known to show different properties [13–16].

The authors reported a charge-transfer thin fi lm made from organic varnish including aniline oligomer as a charge-transfer material exhibiting organic electrolumi-nescent (EL) properties [17]. Copper phthalocyanine is used as a standard charge transfer material in the fi eld of organic EL; it coats a surface with a thin fi lm using a dry process such as a vacuum method because of the lower solubility of copper phthalocyanine. Copper phthalo-cyanine, having high heat resistance and high light resis-tance, is required for application to wet processes such as spin-coating, spraying, and ink-jet methods because of the demand for large EL display manufacture. Cop-per phthalocyanine presents the disadvantageous point of lower transmittance in the visible region because of its strong absorption maxima.

In this study, to develop new charge transfer mate-rials, we attempt to synthesize novel non-peripheral S-aryl substituted phthalocyanines, 1,4,8,11,15,18,22,25-octakis(thiophenyl)phthalocyanines, which show near-infrared absorption, have high strain structure, with no liquid crystal property and no isomers, linking through sulfur atom electron-donating groups.

EXPERIMENTAL

Equipments

Ultraviolet-visible (UV-vis) spectra were measured on a Shimadzu UV-2400PC spectrometer. Each sample was prepared in toluene at 5.0 × 10-5 M, and in chloroform at 5.0 × 10-5 M. Fluorescent spectra were recorded in DMF on a Nihon Bunko Jasco FP-6600 spectrofl uorometer. Proton magnetic resonance (1H NMR) spectra were mea-sured at 400 MHz on a Bruker Avance 400S in dimethyl sulfoxide-d6 (DMSO-d6) using tetramethylsilane (TMS) as the internal standard. Elemental analysis was carried out using a Perkin-Elmer 2400CHN instrument. Mass spectra were taken with a Nihon Denshi Joel JMS-AX500 mass spectrometer. Melting points were measured with a stanford research system MPA100 optimelt automated system. A differential thermobalance was carried out using a MAC Science Tg-DTA 2000SR instrument. Mea-surement of the ionization potential (Ip) was performed using a Riken Keiki AC-2 instrument. Thickness of fi lm was measured with a Kosaka Laboratories Suppercorder ET4000A instrument.

Materials

All chemicals were purchased from Aldrich or Tokyo Chemical Industry Co. Ltd. They were used as received without further purifi cation. For chromatographic separa-tion, silica gel was used (60, particle size 0.063–0.200 nm, 7734-grade; Merck).

Synthesis

Phthalonitrile-3,6-ditrifl ate (1). 2,3-dicyanohydro-quinone (4.80 g, 30 mmol) in dichloromethane (100 mL) and pyridine (py) (5.93 g, 75 mmol) was treated with trifl uoromethanesulfonic anhydride (21.16 g, 75 mmol) under nitrogen at –78°C. After the reaction, the mixture was allowed to warm slowly to room temperature; stir-ring was continued for 24 h. The mixture was poured into water (600 mL) and the organic layer was extracted using dichloromethane (5 × 100 mL). The extract was washed in turn with water, 2% hydrochloric acid, water, brine and water, and dried on magnesium sulfate (MgSO4).The fi ltrate and the solvent evaporated. The crude prod-uct was recrystallized from dichloromethane to afford 1(6.35 g, 50%) as colorless needles. Found: C, 28.32%; H, 0.48%; N, 6.59%. Calcd. for C10H2F6N2S2O6: C, 28.31%; H, 0.48%; F, 26.87%; N, 6.60%; O, 22.63%; S, 15.12%. IR (KBr): cm-1 3115 (νC-H), 2550 (νC-N), 1601 (νC-C), 1472 (νC-C), 1439 (νC-C), 1134 (νS=O). 1H NMR (400 MHz, DMSO-d6): δ, ppm 8.44 (s, 2H).

3,6-bis(thiophenyl)phthalonitriles (2). In a mixture of 1 (0.85 g, 2 mmol), potassium carbonate (1.16 g) and dim-ethyl sulfoxide (DMSO) (15 mL), thiophenols (4 mmol)such as p-toluenethiol, 4-methoxybenzenethiol and tert-butylthiophenol was added; the mixture was reacted


SYNTHESIS OF NEAR-INFRARED ABSORBED METAL PHTHALOCYANINE 49

at room temperature for 24 h in nitrogen atmosphere. The reaction products were poured into water (300 mL), and the organic layer extracted using dichloromethane (5 × 100 mL), and dried on MgSO4. The fi ltrate and the solvent evaporated. The crude product was washed with methanol (3 × 50 mL) and recrystallized from toluene to afford 2 as a yellow solid. 3,6-bis(thiophenylmethyl)phthalonitrile (2a) (0.27 g, 35%). Found: C, 70.90%; H, 4.30%; N, 7.50%. Calcd. for C22H16N2S2: C, 70.93%; H, 4.33%; N, 7.52%; S, 17.22. IR (KBr): cm-1 3050 (νC-H),2970 (νC-H), 2218 (νC-N), 1600 (νC-C), 1535 (νC-C), 1490 (νC-C), 1435 (νC-C), 1210, 809 (δC-H). 1H NMR (400 MHz, DMSO-d6): δ, ppm 7.54 (d, 4H), 7.46 (d, 4H), 7.35 (s2H), 2.66 (tt, 6H). 3,6-bis(thiophenyl methoxy)phthalonitrile (2b) (0.28 g, 34%). Found: C, 65.30%; H, 4.00%; N, 6.93%. Calcd. for C22H16N2S2O2: C, 65.32%; H, 3.99%; N, 6.93%; S, 15.82%; O, 7.91%. IR (KBr): cm-1 3050 (νC-H), 2970 (νC-H), 2216 (νC-N), 1600 (νC-C), 1540 (νC-C),1487 (νC-C), 1430 (νC-C), 1210, 810 (δC-H). 1H NMR (400 MHz, DMSO-d6): δ, ppm 7.49 (d, 4H), 7.06 (d, 4H), 7.04 (s2H), 3.79 (s, 6H). 3,6-bis(thiophenyl tert-butyl)phtha-lonitrile (2c) (0.38 g, 42%). Found: C, 73.65%; H, 6.18%; N, 6.11%. Calcd. for C28H28N2S2: C, 73.64%; H, 6.18%; N, 6.13%; S, 14.04%. IR (KBr): cm-1 3040 (νC-H), 2960 (νC-H), 2210 (νC-N), 1600 (νC-C), 1500 (νC-C), 1460 (νC-C),1210, 808 (δC-H). 1H NMR (400 MHz, DMSO-d6): δ, ppm 7.50 (d, 4H), 7.46 (d, 4H), 7.26 (s, 2H), 1.28 (s, 18H).

1,4,8,11,15,18,22,25-octakis(thiophenyl)phtha-locyanines (3). A solution of 2 (0.25 mmol), metal chlorides or metal acetate, 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as a catalyst and 1-pentanol (1-PeOH) was refl uxed for 7 h. After cooling to room temperature, the reaction products were poured into methanol to form a precipitate, which was washed with water and methanol, and chromatographed on silica gel with toluene as eluent. 1,4,8,11,15,18,22,25-octakis(thiophenylmethyl)phthalo-cyanines (3a); 3a-Cu (0.02 g, 18%). Found: C, 67.98%; H, 4.51%; N, 7.01%. Calcd. for C88H65N8S8Cu: C, 67.99%; H, 4.21%; N, 7.21%; S, 16.50%; Cu, 4.09%. MS (FAB): m/z 1554, calcd. 1554.59. 3a-Co (0.02 g, 24%). Found: C, 68.10%; H, 4.20%; N, 7.25%. Calcd. for C88H65N8-

S8Co: C, 68.19%; H, 4.23%; N, 7.23%; S, 16.55%; Co, 3.80%. MS (FAB): m/z 1550, calcd. 1549.97. 3a-Ni (0.03g, 28%). Found: C, 68.18%; H, 4.20%; N, 7.13%. Calcd. for C88H65N8S8Ni: C, 68.20%; H, 4.23%; N, 7.23%; S, 16.55%; Ni, 3.79%. MS (FAB): m/z 1550, calcd. 1549.73. 3a-Zn (0.02 g, 24%). Found: C, 67.87%; H, 4.23%; N, 7.11%. Calcd. for C88H65N8S8Zn: C, 67.91%; H, 4.21%; N, 7.20%; S, 16.48%; Zn, 4.20%. MS (FAB): m/z 1556, calcd. 1556.43. 3a-Pb (0.04 g, 34%). Found: C, 62.29%; H, 3.88%; N, 6.40%. Calcd. for C88H65N8S8Pb: C, 62.24%; H, 3.86%; N, 6.60%, S, 15.11%; Pb, 12.19%. MS (FAB): m/z 1699, calcd. 1698.24. 3a-H2 (0.04 g, 19%). Found: C, 70.85%; H, 4.66%; N, 7.51%. Calcd. for C88H66N8S8:C, 70.84%; H, 4.46%; N, 7.51%, S, 17.1%. MS (FAB): m/z 1492, calcd. 1492.04. 1,4,8,11,15,18,22,25-octakis-(thiophenylmethoxy)phthalocyanines (3b);3b-Cu (0.03 g,

27%). Found: C, 62.88%; H, 3.91%; N, 6.65%. Calcd. for C88H65N8S8 O8Cu: C, 62.82%; H, 3.89%; N, 6.66%; S, 15.25%; O, 7.61%; Cu, 3.77%. MS (FAB): m/z 1682, calcd. 1682.58. 3b-Co (0.02 g, 21%). Found: C, 63.00%; H, 3.89%; N, 6.59%. Calcd. for C88H65N8S8O8Co: C, 62.99%; H, 3.90%; N, 6.68%; S, 15.29%; O, 7.63%; Co, 3.51%. MS (FAB): m/z 1678, calcd. 1677.97. 3b-Ni (0.03g, 25%). Found: C, 63.01%; H, 3.92%; N, 6.67%. Calcd. for C88H65N8S8O8Ni: C, 63.01%; H, 3.92%; N, 6.67%; S, 15.29%; O, 7.63%; Ni, 3.49%. MS (FAB): m/z 1678, calcd. 1677.73. 3b-Zn (0.03 g, 29%). Found: C, 62.75%; H, 3.95%; N, 6.65%. Calcd. for C88H65N8S8O8Zn: C, 62.75%; H, 3.89%; N, 6.65%; S, 15.23%; O, 7.60%; Zn, 3.88%. MS (FAB): m/z 1684, calcd. 1684.42. 3b-Pb (0.04 g, 35%). Found: C, 57.88%; H, 3.59%; N, 6.15%. Calcd. for C88H65N8S8O8Pb: C, 57.88%; H, 3.59%; N, 6.14%; S, 14.05%; O, 7.01%; Pb, 11.33%. MS (FAB): m/z 1827, calcd. 1826.23. 3b-H2 (0.02 g, 8%). Found: C, 65.24%; H, 4.11%; N, 6.93%. Calcd. for C88H66N8O8S8:C, 65.24%; H, 4.11%; N, 6.92%, O, 7.90, S, 15.18%. MS (FAB): m/z 1620, calcd. 1620.03. 1,4,8,11,15,18, 22,25-octakis(thiophenyl tert-butyl)phthalocyanines (3c); 3c-Cu (0.02 g, 20%). Found: C, 71.13%; H, 6.05%; N, 5.87%. Calcd. for C112H113N8S8Cu: C, 61.13%; H, 6.02%; N, 5.92%; S, 13.56%; Cu, 3.37%. MS (FAB): m/z 1891, calcd. 1891.22. 3c-Co (0.02 g, 21%). Found: C, 71.24%; H, 6.04%; N, 5.93%. Calcd. for C112H113N8-

S8Co: C, 71.30%; H, 6.04%; N, 5.94%; S, 13.60%, Co, 3.12%. MS (FAB): m/z 1886, calcd. 1886.61. 3c-Ni (0.03g, 23%). Found: C, 71.31%; H, 6.04%; N, 5.89%. Calcd. for C112H113N8S8Ni: C, 71.31%; H, 6.04%; N, 5.94%; S, 13.60%; Ni, 3.11%. MS (FAB): m/z 1886, calcd. 1886.37. 3c-Zn (0.03 g, 28%). Found: C, 71.00%; H, 6.01%; N, 5.92%. Calcd. for C112H113N8S8Zn: C, 71.06%; H, 6.02%; N, 5.92%; S, 13.55%; Zn, 3.45%. MS (FAB): m/z 1893, calcd. 1803.07. 3c-Pb (0.04 g, 34%). Found: C, 66.12%; H, 5.63%; N, 5.53%. Calcd. for C112H113N8S8Pb: C, 66.11%; H, 5.60%; N, 5.51%; S, 12.61%; Pb, 10.17%. MS (FAB): m/z 2035, calcd. 2034.88. 3c-H2 (0.02 g, 9%). Found: C, 73.55%; H, 6.26%; N, 6.13%. Calcd. for C112H114N8S8: C, 73.56%; H, 6.28%; N, 6.13%, S, 14.03%. MS (FAB): m/z 1829, calcd. 1828.68.

Reference compounds (4–6). 1,4,8,11,15,18,22,25-octakis(thiooctyl)phthalocyaninate copper, cobalt, nickel, zinc and lead (4) were synthesized from 3,6-thiooctylphtha-lonitrile in accordance with a description in the litera-ture [4]. 2,3,9,10,16,17,23,24-octakis(thiophenylmethyl)phthalocyaninate copper, cobalt, nickel, zinc and lead (5)were synthesized from phthalonitrile having thiophenyl-methyl at a peripheral site [18]. Tetrakis(tert-butyl)phtha-locyaninate copper (6) was synthesized. 4-Cu. Found: C, 73.90%; H, 7.65%; N, 4.76%. Calcd. for C144H177N8S8Cu:C, 73.91%; H, 7.62%; N, 4.79%; S, 10.96%; Cu, 2.72%. MS (FAB): m/z 2340, calcd. 2340.07. 4-Co. Found: C, 74.06%; H, 7.64%; N, 4.77%. Calcd. for C144H177N8S8Co:C, 74.06%; H, 7.64%; N, 4.80%; S, 10.98%, Co, 2.52%. MS (FAB): m/z 2335, calcd. 2335.46. 4-Ni. Found: C,



74.02%; H, 7.64%; N, 4.77%. Calcd. for C144H177N8S8Ni: C, 74.06%; H, 7.64%; N, 4.80%; S, 10.99%; Ni, 2.51%. MS (FAB): m/z 2335, calcd. 2335.22. 4-Zn. Found: C, 73.85%; H, 7.65%; N, 4.78%. Calcd. for C144H177N8S8Zn: C, 73.85%; H, 7.62%; N, 4.78%; S, 10.95%; Zn, 2.80%. MS (FAB): m/z 2342, calcd. 2341.92. 4-Pb. Found: C, 69.63%; H, 7.18%; N, 4.41%. Calcd. for C144H177N8S8Pb: C, 69.63%; H, 7.18%; N, 4.51%; S, 10.32%; Pb, 8.35%. MS (FAB): m/z 2484, calcd. 2483.73. 5-Cu. Found: C, 68.00%; H, 4.28%; N, 7.29%. Calcd. for C88H65N8S8Cu: C, 67.99%; H, 4.21%; N, 7.21%; S, 16.50%; Cu, 4.09%. MS (FAB): m/z 1555, calcd. 1554.59. 5-Co. Found: C, 68.19%; H, 4.23%; N, 7.24%. Calcd. for C88H65N8S8Co: C, 68.19%; H, 4.23%; N, 7.23%; S, 16.55%; Co, 3.80%. MS (FAB): m/z 1550, calcd. 1549.97. 5-Ni. Found: C, 68.21%; H, 4.24%; N, 7.21%. Calcd. for C88H65N8S8Ni: C, 68.20%; H, 4.23%; N, 7.23%; S, 16.55%; Ni, 3.79%. MS (FAB): m/z 1550, calcd. 1549.73. 5-Zn. Found: C, 67.89%; H, 4.19%; N, 7.24%. Calcd. for C88H65N8S8Zn: C, 67.91%; H, 4.21%; N, 7.20%; S, 16.48%; Zn, 4.20%. MS (FAB): m/z1556, calcd. 1556.43. 5-Pb. Found: C, 62.24%; H, 3.84%; N, 6.57%. Calcd. for C88H65N8S8Pb: C, 62.24%; H, 3.86%; N, 6.60%, S, 15.11%; Pb, 12.19%. MS (FAB): m/z 1699, calcd. 1698.24. 6-Cu. Found: C, 72.03%; H, 6.05%; N, 14.01%. Calcd. for C48H48N8Cu: C, 72.02%; H, 6.04%; N, 14.00%; Cu, 7.94%. MS (FAB): m/z 801, calcd. 800.49.


Syntheses of target compounds

Target compounds, 1,4,8,11,15,18,22,25-octakis (thio-phenyl) phthalocyanines (3) were synthesized in three stepsvia intermediates, phthalonitrile-3,6-ditrifl ate (1) and3,6-bis(thiophenyl)phthalonitriles (2). Intermediate 1

was synthesized from 2,3-dicyanohydroquinon and trif-luoromethanesulfonic anhydride for 24 h in accordance with a description from the literature [5]. Intermedi-ates 2 were synthesized, respectively, from 1 and thio-phenols such as p-toluenethiol, 4-methoxybenzenethiol and tert-butylthiophenol at room temperature for 24 h to obtain 3,6-bis(thiophenylmethyl) phthalonitrile (2a),3,6-bis(thiophenylmethoxy)phthalonitrile (2b) and 3,6-bis(thiophenyl tert-butyl)phthalonitrile (2c). Intermediates 1 and 2 were analyzed using IR and 1H NMR spectros-copy, and elemental analysis. Their analytical data showed good agreement with the proposed structure.

The target compounds 1,4,8,11,15,18,22,25 octa kis-(thiophenylmethyl)phthalocyanines (3a), 1,4, 8,11,15, 18,-22,25-octakis(thiophenylmethoxy)phthalocyanines (3b)and 1,4,8,11,15,18,22,25-octakis(thiophenyl tert-butyl)phthalo cyanines (3c) were synthesized, respectively, from corresponding intermediates 2a, 2b, and 2c and metal salt in the presence of DBU as catalyst in 1-PeOH for 7 h (Scheme 1). As metal salts, chloride or acetate of copper, cobalt, nickel, zinc, and lead were chosen [12]. Metal-free compounds of 3 were obtained directly by refl uxing 2 in 1-PeOH. The products were isolated using column chromatography on silica gel with toluene as eluent. The most readily apparent feature of the 3 compounds is their solubility in various solvents.

The target compounds 3 were analyzed using elemen-tal analysis and MS spectroscopy. The analytical data showed good agreement with the proposed structure.

Syntheses of reference compounds

For comparison of properties, reference compounds of three types were also synthesized. One kind of ref erence compound included 1,4,8,11,15,18,22,25-octakis (thiooctyl)phthalocyaninate copper, cobalt, nickel,

3a: X=CH3

S

S

X

X

CN

CN

CN

CNO

CF3

CF3

OOSO

OSO

OH

OH

CN

CN

SHXOSO

CF3OOSO

F3CMCl or MAc

DBU, 1-PeOH160 °C, 7 h

N

N

N

N

N

N

N

N M

S

S

S

S S

S

S

S

X X

X

X

XX

X

X

3b: X=OCH3

1 2

3c: X=t-butyl

3

M=Cu, Co, Ni, Zn, Pb

2a: X=CH3

2b: X=OCH3

2c: X=t-butyl

K2CO3, DMSOrt, 24 h

Py, CH2Cl2-78 to rt, 24 h

Scheme 1. Synthetic pathway of 1,4,8,11,15,18,22,25-octakis(thiophenyl)phthalocyanine



zinc and lead (4) from 3,6-thiooctylphthalonitrile, in accordance with a process described in the literature [7] (Scheme 2).

Other reference compounds were 2,3,9,10,16,17,23,24-octakis(thiophenylmethyl)phthalocyaninate copper, cobalt, nickel, zinc and lead (5) synthesized from phtha-lonitrile having thiophenylmethyl at peripheral site [15] (Scheme 3). The other reference compound was tetrakis(tert-butyl)phthalocyaninate copper (6), which included four regioisomers [19]. The tetra-tert-butyl phthalocyaninate copper was synthesized from 4-tert-butylphthalonitrile. The reference compounds were ana-lyzed using elemental analysis and MS spectroscopy. The

analytical data showed good agreement of the proposed structure.

UV-vis spectra

In substituted metal and metal-free phthalocya-nines, strong absorption is detected in the visible region between the 650 and 690 nm, termed the Q band, and in UV between 320 and 370 nm, called the Soret band. A typical value for the extinction coeffi cient (ε) of the Q band is around 105 cm2 . mol-1.

The absorption spectra of synthesized compounds 3show typical shapes for phthalocyanine analogs (Figs 1–3).

N

N

N

N

N

N

N

N M

S

S

S

S S

S

S

S

SS

NC CN


MCl or MAc


4

Scheme 2. Synthetic pathway of reference compound 4

N

N

N

N

N

N

N

N M

S

S

S

SS

S

S

S

SNC

NC S


MCl or MAc


5

Scheme 3. Synthetic pathway of reference compound 5



They also displayed strong absorption peaks in the vis-ible region at around 800 nm. The strongest peaks in the visible region are assigned as the Q band, which were attributed to the allowed π−π∗ transition of the phthalocyanine ring. The Q band absorption of syn-thesized compounds 3 shifted by 100–150 nm to a

longer wavelength in comparison to unsubstituted metal phthalocyanine and reference compounds. In the case of 3a, absorption maxima are moved to longer wavelengths in the order of Co, Ni, Cu, Zn, and Pb. Compounds 3band 3c show similar phenomena. In comparison to the same central metal, non-peripheral substituted phthalo-cyanines, 3a, 3b, and 3c show longer wavelengths than peripheral substituted 5. Metal-free 3a, 3b, and 3c were also synthesized. The Q band absorption peaks of 3a, 3b,and 3c respectively appeared at longer wavelengths of 815, 820, and 816 nm. In general, metal-fee phthalocya-nines show shorter wavelengths than corresponding metal phthalocyanines. Metal-free 3a, 3b, and 3c, which have remarkably bulky substituents, increase the distortion of the molecule because the four central cavities cannot be fi xed. Then, the Q band of metal-free 3a, 3b, and 3c were not split. The Q band is known to split into two peaks for high symmetry; the splitting Q band decreases with decreasing symmetry. Metal-free 3a, 3b, and 3c display decreased symmetry as a result of the molecular distor-tion. We leave a detailed discussion about these phenom-ena of Q band of metal-free 3a, 3b and 3c for another opportunity. The Q band absorption data of phthalocya-nines are presented in Table 1.

The Q band shifts depend upon the change in the electron distribution in the phthalocyanine ring caused by substituents and their position. These results sug-gest that the steric hindrance arising from the substituted S-aryl groups appears to be as signifi cant as reported by Kobayashi and co-workers [12]. However, a difference of the Q band in 3 is low between substituents, methyl, methoxy and tert-butyl. In this case, electron-donating substituents only slightly affect the movement to a lon-ger wavelength. Although the effect of the central metal on the energy of Q band is usually small [20], absorp-tion maxima of 3 are moved to longer wavelengths, and apparently increase with the ionic radius of the central metal, particularly lead. Lead complexes of 3 showed amplifi ed structural distortion.

Because non-peripheral S-aryl substituted phthalocya-nines having lead as the central metal have an absorption band near 500 nm, 3a-Pb, 3b-Pb, and 3c-Pb resembles a red solution; the other central metals of 3a, 3b, and 3cshow a slightly reddish solution.

Heat resistance test

The respective heat resistance characteristics of 3and 4 were estimated using a differential thermobalance instrument (Tg-DTA 2000SR; Mac Science). Heat resis-tance testing was the following: the sample temperature was raised from 30 to 400 °C by 1 °C every 1 min; the temperature was recorded at a point to 5% weight less per 400 mg sample weight. The target compounds 3a, 3b,and 3c exhibited high 5% weight less temperature around 300°C, whereas reference compounds 4 showed a very low temperature. Results show that 3a, 3b and 3c possess

Fig. 1. UV-vis spectra of 3a

Fig. 3. UV-vis spectra of 3c

Fig. 2. UV-vis spectra of 3b



very high heat-resistance. All 3 compounds are able to use an organic device and a thin fi lm coated onto a sur-face. The 5% weightless data are presented in Table 2.

Compounds 4 show behavior of anisotropic disc-shaped liquid crystals around 70°C because of their longer alkyl side chains. These phenomena have been reported for long alkyl group substituted phthalocyanines [1, 2, 21].

Preparation of thin fi lms and their ionization potentials

Under a nitrogen atmosphere, 0.3 g of 3 was dissolved in 5.0 g of 1,3-dimethyl-2-imidazolidinone. Cyclohexanol (5.0 g) was added to the 1,3-dimethyl-2-imidazolidinone-solution to afford varnish (including 3% solid content). The varnish was spin-coated onto an indium tin oxide (ITO) glass basal plate, which was washed with ozone for 40 min until immediately before spin-coating. The spin-coated ITO glass basal plates were baked at 200°C for 10 min to obtain 30-nm-thick thin fi lms. The thin fi lm thickness was measured (Supercorder ET4000A; Kosaka Laboratory). Prepared thin fi lms of 3 show no apparent color difference from the solution phase.

Varnish of compound 5 is diffi cult to prepare to the same composition of the solvent to manufacture for 3.For preparation of thin fi lms, the solvent 1,3-dimetyl-2-imidazolidinone is an effi cacious solvent for solid content, while cyclohexanol acts as a poor solvent. The solubility in cyclohexanol of 5 was lower than that of 3.Therefore, the poor solvent was changed to chloroform.

Non-peripheral substituted phthalocyanines show higher solubility than peripheral substituted phthalocyanine because 3a and 5 have identical substituents at differing positions. By improving the poor solvent, 30-nm-thick thin fi lms containing 5 can be made.

The Ip of thin fi lms were measured (AC-2; Riken Keiki Co. Ltd.). In the case of 3, Ip were varied for the central metal, and increased in the order of Cu, Zn, Ni, Co, and Pb, which were independent of the S-aryl group type: thiophenylmethyl, thiophenylmethoxy, and thiophenyl tert-butyl. In comparison to the same central metal of 3,the tendency of Ip shows a similar value for 3a and 3c,and is apparently the highest for 3b. Comparison with Ipof thin fi lms shows that those of 3 are lower than those of 5 by 0.5 eV. Actually, Ip is known to be around 5.0 eV for usually used non-substituted copper phthalocyanine thin fi lms. The Ip of copper phthalocyanine shows a simi-lar value for the peripheral substituted phthalocyanine, such as 5. Because Ip means energy of Highest Occupied Molecular Orbital (HOMO), non-peripheral S-aryl sub-stituted phthalocyanines have the highest HOMO energy. This result suggests that non-peripheral S-aryl substituted phthalocyanines 3 removed a barrier of charge transfer using the electrode.

CONCLUSION

Phthalocyanines having S-aryl at non-peripheral posi-tion can be synthesized. Their strong absorption peaks in the visible region called the Q band show around 800 nm.

Table 1. Q band absorption data of target compounds 3a, 3b and 3c and reference compounds

Solvent Central metal 3a 3b 3c 4 5

λmax, nm log ε λmax, nm log ε λmax, nm log ε λmax, nm λmax, nm log ε

Chloroform metal-free 815 5.22 820 5.33 816 5.08 Pb 857 5.09 862 5.13 855 4.89 818 Cu 805 5.20 808 5.17 805 5.29 783 Ni 788 4.74 790 4.97 787 5.00 Zn 795 4.92 795 5.01 789 5.18 782 Co 779 4.91 783 4.51 780 4.91

Toluene Pb 838 5.15 846 5.18 836 5.08 754 4.89 Cu 791 5.01 796 5.06 790 5.01 716 4.96 Ni 777 5.09 782 5.10 776 4.94 706 4.99 Zn 792 4.20 790 4.34 782 4.86 715 5.11 Co 775 4.71 778 4.89 774 4.89 708 4.70

Table 3. Ionization potential for 3a, 3b and 3c, and reference compound 5

Central metal 3a 3b 3c 5

Pb 4.80 4.90 4.75 5.11Cu 4.65 4.78 4.64 4.98Ni 4.69 4.82 4.68 5.03Zn 4.66 4.80 4.66 5.01Co 4.70 4.84 4.70 5.05

Table 2. The 5% weight less temperature for 3a, 3b and 3c, and reference compound 4

Central metal 3a 3b 3c 4

Pb 298 298 291 73Cu 305 302 300 81Ni 305 301 296 80Zn 312 306 303 75Co 303 301 299 78



The synthesized phthalocyanines have high solubility for organic solvents, high heat resistance without discotic liquid crystal behavior, no absorption in visible region, and a simple compound including no regio isomers.

Thin fi lms of synthesized phthalocyanine on ITO were prepared. The Ip measurements show that those of 3were lower than those of 5 and copper phthalocyanine by 0.5 eV. Synthesized phthalocyanines 3 were removed as a barrier to charge transfer using an electrode.

Acknowledgements

We are pleased to acknowledge the considerable assis-tance of Mr. Tomoyuki Ozawa.

REFERENCES

1. Hirohashi R, Sakamoto K and Okumura E. Phthalo-cyanines as Functional Dyes, IPC: Tokyo, 2004.

2. McKeown NB. Phthalocyanine Materials. Synthe-sis, Structure and Function, Cambridge University Press: Cambridge, 1998.

3. Leznoff CC and Lever ABP. Phthalocyanines Prop-erties and Applications, Vols. 1–4, VCH: New York, 1989–1996.

4. Rio Y, Rodriguez S and Torres T. Org. Biomol. Chem. 2008; 6: 1877–1894.

5. The Porphyrin Handbook, Vols. 15–20, Kadish KM, Smith KM and Guilard R. (Eds.) Academic Press, San Diego, 2003.

6. de la Torre G, Clessens CG and Torres T. Chem.Commun. 2007; 2000–2015.

7. Ohono E and Sakamoto K. Nippon Kagaku Kaishi, J. Chem., Soc. Jan, Chem. Ind. Chem. 1995; 730–735.

8. Burnham PM, Cook MJ, Gerrard LA, Heeney MJ and Hughes DL. Chem. Commun. 2003; 2064–2065.

9. Cook MJ, Dunn AJ, Thomson AJ and Harrison KJ. J. Chem. Soc., Perkin Trans. I 1998; 2453–2458.

10. Ouguchi T and Aihara S. JP 1992–04015265 (Chem.Abstr. 1992; 117: 61660).

11. Fukuda T, Ishiguro T and Kobayashi N. Tetrahedron Lett. 2005; 46: 2907–2909.

12. Kobayashi N, Fukuda T, Ueno K and Ogino H. J. Am. Chem. Soc. 2001; 123: 10740–10741.

13. Japan patent 2004–18561, 2004–309655. 14. Sakamoto K, Kato T and Cook MJ. J. Porphyrins

Phthalocyanines 2001; 5: 742–750. 15. Sakamoto K, Kato T, Kawaguchi T, Ohno-Okumura

E, Urano T, Yamaoka T, Suzuki S and Cook MJ. J. Photochem. Photobiol. A: Chem. 2002; 153:245–253.

16. Sakamoto K, Ohno-Okumura E, Kato T, Kawagu-chi T and Cook MJ. J. Porphyrins Phthalocyanines2003; 7: 83–85.

17. Japan patent 2002–151272. 18. Cosimelli B, Roncucci G, Dei D, Fantetti L, Ferroni

F, Ricci M and Spinelli D. Tetrahedron 2003; 59:10025–10030.

19. Gaspard S and Maillard PH. Tetrahedron 1987; 43:1083–1090.

20. Isago H, Kagaya Y and Nakajima S. Chem. Lett.2003; 32: 112–113.

21. Chambrier I, Cook MJ, Ccknell SJ and McMurdo J. J. Mater Chem. 1993; 3: 841–849.


DOI: 10.1142/S1088424610001702



INTRODUCTION

Since the earlier reports of the X-ray structure of the light-harvesting device II in the purple photosynthetic bacteria (Rhodopseudomonas acidophila and Rhodo-pseudomonas molischianum) [1–3], the number of experimental and theoretical studies on the photophysi-cal properties and excited state dynamics of this supra-molecular biological antenna have increased rapidly. All

in all, both the supramolecular biostructure and ultrafast excited state dynamics turn out to be fairly well under-stood in a general sense.

Over the past seven years, our groups have investi-gated large series of cofacial bisporphyrin dyads espe-cially designed for through space singlet-singlet (S1) and triplet-triplet (T1) energy transfers (ET; Scheme 1) [4–6]. Rate for S1 ET (S1 kET) turned out to be much slower than that found for LH I and LH II. Rates ranging between 1–35 ps were reported in LH systems [7, 8] whereas our fastest systems occur in the 70–100 ps time scale [5, 9]. While the Förster mechanism is invoked as the possible S1 ET in the LH systems and cofacial bismacrocycles, the key feature is that despite the much closer energy donor-acceptor separations (i.e. ~3.5 Å at best) in the cofacial model dyads in comparison to the various donor-acceptor

Through space singlet energy transfers in light-harvesting systems and cofacial bisporphyrin dyads

Pierre D. Harvey*a, Christine Sternb , Claude P. Grosb and Roger Guilard*b

a Département de Chimie, Université de Sherbrooke, Sherbrooke J1K 2R1, Québec, Canadab Institut de Chimie Moléculaire de l’Université de Bourgogne (ICMUB, UMR 5260), 9 Avenue Alain Savary, BP 47870, 21078 Dijon Cedex, France


ABSTRACT: Recent discoveries from our research groups on the photophysics of a few cofacial bisporphyrin dyads for through space singlet and triplet energy transfers raised several important investigations about the mechanism of energy transfers and energy migration in light-harvesting devices, notably LH II, in the heavily investigated purple photosynthetic bacteria. The key feature is that for face-to-face and slipped dyads with controlled structure using rigid spacers or spacers with limited fl exibilities, our fastest rates for singlet energy transfer are in the 10 × 109 s-1 (i.e. 100 ps time scale) for donor-acceptor distances of ~3.5–3.6 Å. The time scale for energy transfers between different bacteriochlorophylls, notably B800* B850, is in the ps despite the long Mg···Mg separation (~18 Å). This short rate drastically contrasts with the well-accepted Förster theory. This review focuses on the photophysical processes and dynamics in LH II and compares these parameters with our investigated model dyads build upon octa-etio-porphyrin chromophores and rigid and semi-rigid spacers. The recently discovered role of the rhodopin glucoside (carotenoid) will be analyzed as possible relay for energy transfers, including the possibility of uphill processes at room temperature. In this context the concept of energy migration may be complemented by parallel relays and uphill processes. It is also becoming more obvious that the irreversible electron transfer at the reaction center (electron transfer from the special pair to the phaeophytin) renders the rates for energy transfer and migration faster precluding all possibility of back transfers.

KEYWORDS: LH II, cofacial dyads, energy tranfer, bisporphyrin.


*Correspondence to: Pierre D. Harvey, email: [email protected], tel: +33 (0)3-80-39-68-39, fax: +33 (0)3-80-39-61-17 and Roger Guilard, email: [email protected], tel: +33 (0)3-80-39-61-11, fax: +33 (0)3-80-39-61-17


56 P.D. HARVEY ET AL.

frames in the living supramolecular structures (i.e. ~20 Å), the S1 ET data are strikingly slower (two orders of mag-nitude). So the question is what is missing?

This mini review stresses some structural relevant details recently investigated, notably with respect to the carotenoid residue, rhodopin glucoside, along with its role in the overall excited state dynamics in the bio-superstructures (LH I and LH II), all in the light of the recent singlet excited state dynamics obtained for simple but well-defi ned and -controlled model dyads.

Before moving on to the main body of this review, a comment about back energy transfer must be made. Back energy transfer is a recently investigated phenomenon in the area of porphyrins [10–12]. Conceptually, at room temperature, the surrounding to the chromophore, very often the solvent, gives its thermal energy to the latter. The amount of energy is obviously modest and S0 S1

transitions are simply impossible. However, for small energy gaps, for example S1(donor)-S1(acceptor), such a process is possible. In conclusion, if the energy gap is small, then the back energy transfer is more probable and the observed overall rate of transfer (energy transfer rate - back energy transfer rate = observed overall energy

transfer rate) will consequently be small as well. Based on the energy gaps observed in nature (such as the LH I and LH II sys-tems discussed below), this process should be dominant. However, this is not the case because nature structurally organized her chromophores in an extended energetically downhill cascade fashion, so back energy transfer becomes essentially improbable.

SUPRAMOLECULAR STRUC-TURES OF LH I AND LH II

Using X-ray methods [1–3] and parallel investigations such as AFM [13–15] and synchrotron small-angle X-ray scatterings [16], many circular and ellipsoid structures for LH I and LH II systems were elucidated. The photosyn-thetic membrane of the purple photosynthetic bacteria is composed of many phospholipid-fi lled ring systems (LH II) and several larger dissymmetric rings (LH I) stacked like a honey comb, as schematically shown in Fig. 1. Inside LH I is placed a protein called the reaction cen-ter (RC) where the primary photoinduced electron trans-fer arising from a special pair of bacteriochlorophylls a(BChl a) takes place (Scheme 2). These doughnut-shaped bio-architectures are set at a right angle with respect to the cytoplasmic membrane.

The bio-nanomaterial LH II has diameter of ~10 nm and is composed of a nonamer or an octema of αβ-apoprotein pairs placed in a circular shape. The α-apoprotein α-helices placed inside the ring are per-pendicular to the intra-cytoplasmic plane while the outer β-apoprotein α-helices make an 15° angle with the mem-brane plane. Each unit of αβ-apoproteins carry three BChl a. One of them, called B800, is placed parallel to the membrane plane and located at the polar lower end of the αβ-apoprotein. It is held by a carboxylate group of

N NNN

NH NHNN

M

SO N-H

t-Bu

t-Bu

O

DPB DPX DPA DPO DPS DPNH

O

DPOx

M = Zn, energy donor

free base, energy acceptor

spacer

Scheme 1.

Fig. 1. Left: drawing shows two LH II sitting next to one LH I unit in Rhodopseudomonas acidophila. The gray circles are polypep-tides and the bars are rings of interacting BChl a (called B850). In the middle of LH I is a protein called the reaction center (RC). Right: drawing of a LH II ring showing only the BChl a for the B850 network, the non-interacting B800 bacteriochlorophylls, and the glucoside rhodopins. Two of the B850 units were marked with turquoise arrows representing the transition moments of the Q bands stressing the quasi-parallel relationship between the two. This is also true for the rest of the B850 ring. Note that theRhodopseudomonas molischianum is composed of eight apoproteins. It is not shown for simplicity


THROUGH SPACE SINGLET ENERGY TRANSFERS 57

the N-terminal methionine residue via a Mg-O coordina-tion bond (2.04 Å). The nearest B800 BChl a is placed at 21 Å (Fig. 1, bottom) and is not interacting with the latter macrocycle. This arrangement gives rise to an absorption spectrum at 800 nm in the Q region (Qy band in Fig. 2).

These αβ-apoproteins also contain two other perpen-dicularly oriented BChl a located towards the middle of the apoprotein and are held in place by two histidines. The local geometry consists of a slipped dimer of B850 macrocycles (Fig. 3) where perfect cofacial arrangement is prevented by axial histidine ligands (Fig. 3, center) in a quasi-C2h local symmetry. The closest C···C contacts between the two B850 residues are 3.4 and 3.9 Å (Fig. 3,

right). When the cyclic nonamer is formed (Fig. 1), other close B850-B850 contacts occur (Fig. 3, left) with similar C···C separations (4.0 and 3.9 Å). These contacts promote inter-chromophoric interactions giving rise to a red-shifted Qy band (850 nm; Fig. 2). This quasi-C2h local symmetry also implies that the relative orientation of the transition moments of the Qy band in the pair is oriented in opposite directions (Fig. 1, right), and all around the B850 ring as well.

In the original X-ray report [2], one carotenoid was located per αβ-apoprotein pair (Fig. 1, right). This all-trans carotenoid exhibits a twisted geometry and stretches all across the apoprotein. Because the sugar

Scheme 2.

Fig. 2. Left: drawing of the bacteriochlorin macrocycle indicating the orientation of the transition moments associated with the Qx

and Qy bands. Right: absorption spectrum of the LH II supramolecular antenna of Rhodopseudomonas acidophila (left) stressing the various components belonging to the rhodopin glucoside, and the B800 and B850 chromophores (in the Qx and Qy regions). The band at 380 nm is the Soret band, also associated with B800 and B850 units



fragment is deeply anchored in one of the α-helice and the carotenoid chain runs through the neighbouring αβ-apoprotein, the rhodopin glucoside molecule acts as an attaching device holding the supramolecular structure of LH II in place. Beside this important property, the rhodopin glucoside also exhibits important close van der Waals contacts with all three BChl a included in the pairs (Fig. 4). For the B800 units, a close contact is also observed between the ketone of the bacteriochlorophyll and one of the C atoms of the carotenoid chain. This organization allows the carotenoid to also act as an effi -cient photo-protective accessory against singlet oxygen formation. Importantly, this long π-conjugated organic chain behaves as a molecular solar panel and can con-tribute to the antenna effect of LH I and LH II, notably because of the very close proximity with the neighbor-ing pigments.

EXCITED STATE DYNAMICS OF LH I AND LH II

When the LH II antenna absorbs light via one of the BChl a chromophores, a series of complex non-radiative photophysical processes take place to effi ciently chan-nel the excitation energy from the site of absorption to the special pair in the reaction center. These photophysi-cal events are of two types: energy migration (exciton)

and energy transfer. These fast events occur in the singlet state (S1) of the bacteriochlorophyll pigments. However, the contribution of the carotenoid chains was only recently elucidated. The carotenoids’ role turns out to be important particularly with respect to the non-radiative energy transport event. Fig. 5 (left) illustrates an example of a path for energy migration across the LH II-containing membrane. The “hoping” of excitation energy from one B850 chromophore to another inside and outside the LH II supramolecular edifi ce is in fact a series of reversible energy transfer processes, here in the S1 state. Ultimately, this excitation energy will very effi ciently end up in the special pair of the reaction cen-ter prior to triggering the primary photoinduced elec-tron transfer that sustains life. The time scales for these events are presented in Table 1. The difference between an energy transfer and energy migration is described in Fig. 5 (right).

Using primarily ultrafast fl ash photolysis techniques (transient absorption spectroscopy; UV-vis; lumines-cence; IR), the time scales for all inter-BChl a pigments in the LH I and LH II complexes were recently deter-mined (Table 1) [8]. These fs and ps events were the sub-jects of numerous theoretical investigations, notably with respect to the Förster theory, the amplitude of interchro-mophore couplings and molecular orbital properties (see for examples references 17–20).

Using X-ray data for LH II [1–3], the calculated elec-tronic coupling constant between two B850 pigments located within the same αβ-apoprotein is 320 cm-1. Simi-larly, the calculated coupling constant between two B850 pigments belonging to two different αβ-apotroteins is 255 cm-1. These two values are consistent with the different intermolecular separations (Mg···Mg distances = 9.2 and 9.5 Å, respectively; Fig. 1 and Fig. 3) [18]. These cou-pling constants are large and ab initio molecular orbital calculations on the B850 ring indicate a complete delo-calisation of the orbitals [21]. Such properties explain very well the ultrafast nanometer energy migration (fs time scale) across the circular 18 B850 chromophores. Concurrently, the intra LH II B800-B850 electronic coupling constant is one order of magnitude smaller (30 cm-1). This smaller value is consistent with the slower rate of S1 energy transfer between B880 and B850

Fig. 3. Left: local geometry of the two BChl a called B850 inside the αβ-apoprotein unit. The histidine residues are not shown and the polypeptides α-helices are shown as ribbons for the sake of clarity. Middle: B850-histidine pair in a quasi-C2h local symmetry. Right: local structure of one B850 unit with respect to the neighbouring B850 analogs stressing the closest C···C contacts

Fig. 4. Scheme showing the close C···C contacts between one of the B850 units (in gray) and one neighboring rhodopin gluco-side residues (in black; only a part of the chain is shown)



(~1 ps time scale) and is consistent with the longer Mg···Mg separation (18.3 and 17.6 Å) [18].

Curiously, the rate for energy migration across the B800 units in LH II appears particularly fast (time scale of ~0.5 ps) when considering the rather long Mg···Mg separation of 21.3 Å as compared with the other data summarized in Table 1. Although rates were calculated and agreed relatively well with this type of time scale [19, 21], some discrepancies were noted throughout the literature (including that for the B800 B800 energy transfer process in LH II) [22–25]. Perhaps the largest discrepancy comes from the B800 B800 energy migra-tion and a mixing with the B800-B850 exciton levels was proposed [24]. In search for another possible explana-tion, the role of the rhodopin glucoside (Scheme 2) in the antenna effect is now presented.

THE ROLE OF THE RHODOPIN GLUCO-SIDE

The photophysical behaviour of the carotenoid in LH I and LH II was the subject of recent intense investigations (see for examples references 26–32). The key feature is the presence of a close contact (C···O separation = 3.4 Å), between the B800 unit and a C-H group of the rhodophin glucoside (Fig. 6). Evidence for its role in the S1 energy transfer carotenoid B800 was demonstrated [33]. The binding energy was estimated to be about 2 kcal/mol. Time-dependent density functional theory calculations (TDDFT) of the B800-rhodopin glucoside dyad indicate a red-shift (of about 2 nm) of the B800 Qy transition, along with a substantial increase of its oscillator strength.

The through space S1 energy transfer mechanism from the carotenoid to B800 and B850 units is well estab-lished. The absorption spectrum of this chromophore spreads from 400 to 550 nm, between the Soret and the Qx bands of the BChl a (Fig. 2). This vibronically structured absorption is due to a S0 S2 radiative process (1Ag

− 1Bu+), whereas the symmetry-forbidden S0 S1

analogue (1Ag− 1Ag

−) is optically silent in the single-photon absorption spectra due to the C2h local symmetry. The exact location of this “dark” S1 state (12550 ± 150 cm-1; i.e. 797 ± 10 nm) was determined by fs transient and two photon absorption spectroscopy [34, 35]. There is only a 3 nm difference (excluding the uncertainties) between the B800 absorption and that evaluated for the

Table 1. Time scales for energy migration and transfer between B800 and B850 units of the LH I and LH II rings in the purple photosynthetic membrane [8, 19]

Non-radiative photophysical event Time scale, ps

B850 B850 (exciton in LH II) 0.1–0.2B800 B800 (exciton in LH II) ~0.5LH II LH II (exciton) ~7B850 B850 (exciton in LH I) 0.1–0.2B800 B850 (ET in LH II) 0.8–1.2LH II B875 (ET in LH I) 3–5B875 special pair 20–40

Fig. 5. Left: scheme showing the path of the ultrafast energy migration (exciton) from one LH II complex to another, via the B850 rings. Right: typical state diagram for a diamagnetic molecule designated as “Donor” stressing all radiative (A = absorption, F =fl uorescence, P = phosphorescence) and non-radiative processes (ic = internal conversion, isc = intersystem crossing, ip = intersys-tem crossing from the triplet state, ET = energy transfer from S1 or T1, if in the presence of an “Acceptor” molecule), and exciton phenomenon (reversible energy transfer) between identical coupled chromophores (here “Donor”). In this example, the exciton process is terminated by a downhill energy transfer rendering exciton reversibility less probable



“dark” S1 state of the carotenoid. If the absorption band for this symmetry-forbidden transition is strong, then the spectral overlap between the two bands would be large. The through space rates for S1 energy transfer from the rhodopin to B800 and B850, as well as the internal excited state dynamics (i.e. internal conversion) are compared in Fig. 7 [36, 37]. The rates for intermolecular through space S1 energy transfers are fast (fs and ps), notably from the 1Bu+ state for which more than 40% of the S2 population results in transfer to B800 and B850 Qx manifolds. Such rates are consistent with the relatively close proximity of the chlorophyll and carotenoid chromophores shown in Figs 4 and 6. Concurrently, the rates for energy transfer from the S1 state (1Ag−) to B800 and B850 are two orders of magnitude slower.

Ab initio calculations of the excited state properties of the 18 B850, nine B800 units, and nine rhodopin glu-cosides show a large mixing between the chlorophyll aand carotenoid states. The computed B800-B850 cou-lombic coupling, both with and without the rhodopin

glucoside, reveals that the carotenoid induces an increase in the B800-B850 coupling via a super exchange mechanism [36]. This carotenoid-mediated coupling explains, in part, the dis-crepancy between the measured rate for energy transfer between B800 and B850 chromophores and the calculated rates extracted from the tra-ditional Förster theory. It becomes evident that, in the light of these fi ndings, the ~1 ps time scale reported for the through space S1 energy transfer between B800 and B850 (Table 1) for a distance of about 21 Å (Mg···Mg separation) appears unreasonably too fast, unless mediation of the energy fl ow (through the carotenoid; Qx

B800 “hot S1” carotenoid Qx B850) occurs. Even if mediation occurs, the reported time scale of 30 ps (S1 carotenoid Qy B800; Fig. 7 [36]) appears on the slow side in comparison with the reported 1 ps time scale (Table 1). On the other

hand, the rates for the energy transfer originating from the S2 state are faster (0.2–0.3 ps time-scale; Table 1) than that of the reported rates for Qy B800 Qy B850 (1 ps time scale), which appear reasonable, but this channel is inaccessible on an energy stand point since S1 of B800 is placed lower than that for the carotenoid S2 manifold. The whole picture would perfectly be accounted for if the energy transfer rates originating from the recently discovered but still undefi ned “hot S1”state could be in the appropriate time scale and that back energy transfer and exciton processes would also occur with the B800 and B850 via the Qx levels (Figs. 6 and 7). This has never been described.

INVESTIGATIONS USING COFACIAL BISMACROCYCLE DYADS

The dyads are presented in Schemes 1 and 3, and Table 1. The dyads listed in Scheme 1 share a number of similarities with the B850 pair. First, the closest C···C

Fig. 6. Left: drawing showing the presence of a contact between the ketone of the BChl a B800 and one of the neighboring pro-tons of the rhodopin glucoside in LH II. Right: drawing of the local environment of the rhodopin glucoside residue (in black) withrespect to B800 and B850 units of LH II stressing the downhill S1 energy transfers (black arrows) and the possibility of slightly uphill transfers

Fig. 7. State diagram of the carotenoid along with the Qx, Qy and S0

levels of the B800 and B850 chromophores in the LH II complex. The “hot S1” state is yet to be determined. Selected intramolecular rela-xation and intermolecular S1 energy transfer processes (i.e. energy fl ow) are indicated in gray arrows along with their corresponding time scale (based on data extracted from references 36 and 37). The values in percent indicate the relative population that uses this relaxation path



separations in the pair shown (Fig. 3, center; 3.4 and 3.9 Å) resemble that of DPB and DPOx spacer-containing dyads. Secondly, X-ray studies reveal almost invariably a “slipped dimer” geometry, particularly for the DPB, DPX and DPA species. Finally, the parallel orientation of the S0 S1 transition moments in the dyads is also evident (Fig. 8) with the small exception that in BChl a, there is a signifi cant dipole moment, which is not the case for the symmetric porphyrin series of Table 2. All in all, these compounds are reasonable models in order to extract

through space S1 energy transfer rates that should ade-quately resemble that found in the LH I and II devices. Moreover, two other semi-fl exible models were stud-ied (Scheme 3, J. Weiss (Strasbourg) and P.D. Harvey) [40]. The open conformer (“calix open”) is induced by intramolecular H-bonds in the lower-rim between the hydroxyl groups and O-atoms of the crown ether, whereas the closed conformation (“calix closed”) is induced by the presence of ethoxy instead of hydroxyl functions pre-venting H-bonds. In the later case, it possibly induces the

Table 2. Rigid and flexible spacers used for the S1 energy transfers in cofacial dyads

Code Name Type d(Cmeso···Cmeso)/Åa

DPB 1,8-biphenylenyl Rigid 3.80DPX 4,5-(9,9-dimethylxanthenyl) Rigid 4.32DPA 1,8-anthracenyl Rigid 4.94DPO 4,6-dibenzofuranyl Rigid 5.53DPS 4,6-dibenzothiophenyl Rigid 6.30DPNH 1,1′-(3,3′-di-tert-butylcarbazoyl) Rigid 5.8b

DPOx 2,2′-diphenylether Flexible 4.2c

a These average distances are extracted from X-ray analyses from our research groups or from references included in references 9 and 38. b Calculated value from DFT (B3LYP; 3-21G) from reference 4. c Calculated value from DFT (B3LYP; 3-21G) from reference 5; other distances are N···N = 3.6 and C···Zn = 3.2 Å.

Fig. 8. HOMO and LUMO drawings for a model compound, Zn2(DPNH) [4]. Right: scheme showing the parallel orientation of the S0 S1 transition moment. The rigid spacer is DPNH as indicated in Scheme 1. Note that the electronic density does not shift from one side of the molecule to the other like that found for BChl a, but rather concentrate in the middle. Do note the absence of density in the HOMO at the N-atoms and central meso-carbons, and presence of electron density on these sites in the LUMO

NHN

HNN

NN

NN

Zn

O

O

O

O

H

H

OO

O ONH N

HNN

N NNN Zn

O

O

O

O

OO

O O

calix open calix closed

Scheme 3.



two porphyrins to get close to each other like a scissor effect and to promote fast transfer due to proximity.

The kET rates for through space S1 energy transfer can be evaluated by ns and ps fl uores-cence lifetime of quantum yield measurements of the zinc porphyrin donor chromophore. Because of the diffi culty in measuring the fl uorescence quantum yield of the donor (due to extremely low emission intensity and inac-curacy in extracting the absorbance of the zinc porphyrin component in the absorption spec-trum), the measurements of the fl uorescence lifetimes were performed instead. Using Equation 1, one can evaluate kET [39]:

kET = (1/τF) - (1/τF0) (1)

where τF is the fl uorescence lifetime of the donor in the dyad and τF

0 is the lifetime for a closely related bismacrocycle (when possible) where no energy transfer occurs. For example, the model compound for the dyad ZnH2(DPNH) pre-sented in Scheme 1 and Table 2, is Zn2(DPNH) (Fig. 8). Hence the internal radiative and non-radiative rate con-stants (here kF, kic and kisc) for the dyads and donor-donor models are equal and cancel out in Equation 1.

The fastest through space S1 energy transfer rates are indicated below (Scheme 4) [5, 9, 40]. These systems are ZnH2(DPB), where the rigid spacer exhibits a short Cmeso···Cmeso separation, ZnH2(DPOx), where the absence of rigidity promotes a collapse of the cofacial structure, and the “opened calix” of Scheme 3, where ethanol and low temperature promote intermolecular H-bonds and strong hydrophobic intermacrocycle interactions (changing for a “closed conformer”). In these cases, the time scale spreads in a window of 50–100 ps (i.e. for rates ranging from 10–20 × 109 s-1 or 10–20 (ns)-1) for systems where geometry is promoting the strongest cofacial intermacro-cycle interactions possible. Clearly, two to three orders of magnitude difference are noted for the ET rates measured for the cofacial dyads (Scheme 4) and those reported within the LH I and LH II complexes (Table 1 and Fig. 8), despite the similar interchromophore distances.

The answer can be found in the back energy transfer (or uphill energy transfer). On some occasions, back energy transfers were observed and the rates measured in the LH II systems [41], bisporphyrins [42] and poly-porphyrin assemblies [43]. In the energy transfer dyads (Schemes 1 and 3), the overall observed rates include both the forward and back transfers. In the LH I and LH II complexes, the exciton process dissipates the excitation energy very rapidly. In this way, the rate for back energy transfer from the acceptor chromophore to the donor is greatly reduced. Ultimately, the energy fl ow migrates from the LH II device to the LH I sys-tem, and then to the reaction center inside the photosyn-thetic membrane. Inside the reaction center, the primary

electron transfer takes place, followed by a series of thermody namically downhill subsequent electron trans-fers. Overall, the energy migration becomes irreversible and the phenomenon discussed earlier in the introduction becomes highly improbable. In the designed models for the cofacial systems, such a process (downhill energy cas-cade) is not present, and so back energy transfer is pos-sible. The presence of back energy transfer contributes to the slowing down of the overall observed rate. Improve-ment in the future should include fragments for energy delocalization away from the donor in order to increase the rate of transfer.

CONCLUSION

A close examination of the literature on LH II and the carotenoid excited dynamics, in parallel with well-defi ned and structurally addressable dyads allowed one to question some aspects of the apparent “long range” through space S1 energy transfer, notably between B800 and B850, and secondly to stress on the outmost impor-tance of the role of the exciton process (excitation energy migration) on the excited state dynamics of these supra-molecular bio-organisations. It became clear that the downhill cascade energy migration along with the exci-tonic processes contribute greatly to the effi ciency of transfer. Also, based on the comparison of the molecular structures of the BChl a and porphyrin, it is also obvious that the larger transition moment in the former also con-tributes to the faster rate of energy migration in the LH II system. Finally, nature made sure that there was no atom nor group of at least a minimal amount to ensure that the rates for excitation energy dissipation through the light-harvesting complexes were maximum. Again, simple cofacial dyads clearly demonstrated that.

OO

OH

OH

N

NN

NZn

NHHN

N

N

O

O

OO

N NNN

O

Zn

NH NHNN

N NNN Zn

NH NHNN

EtOH

EtOH

kET = 20.8 (ns)-1, 298 K, 2MeTHFkET = 15.4 (ns)-1, 77 K, 2MeTHF

kET > 9.4 (ns)-1, 298 K, 2MeTHFkET = 6.7 (ns)-1, 77 K, 2MeTHF

kET = 14 (ns)-1, 77 K, EtOH

Scheme 4.



Acknowledgements

The Natural Sciences and Engineering Research Council of Canada (NSERC), the Centre National de Recherche Scientifi que (CNRS UMR 5260) and the Pro-gramme International de Coopération Scientifi que (PICS no. 3183) are acknowledged for funding.

REFERENCES

1. Karrasch S, Bullough PA and Ghosh R. EMBO J.1995; 14: 631–638.

2. McDermott G, Prince SM, Freer AA, Hawthorn-thwaite-Lawless AM, Papiz MZ, Cogdell RJ and Isaacs NW. Nature (London, UK) 1995; 374:517–521.

3. Koepke J, Hu X, Muenke C, Schulten K and Michel H. Structure (Cambridge, MA, USA) 1996; 4: 581–597.

4. Gros CP, Aly SM, El Ojaimi M, Barbe J-M, Bri-sach F, Abd-El-Aziz AS, Guilard R and Harvey PD. J. Porphyrins Phthalocyanines 2007; 11: 244–257.

5. Gros CP, Brisach F, Meristoudi A, Espinosa E, Guilard R and Harvey P. Inorg. Chem. 2007; 46:125–135.

6. Harvey P, Stern C, Gros CP and Guilard R. Coord. Chem. Rev. 2007; 251: 401–428.

7. Cogdell RJ, Gardiner AT, Roszak AW, Law CJ, Southall J and Isaacs NW. Photosynth. Res. 2004; 81: 207–214.

8. Light-Harvesting Antennas in Photosynthesis, Vol. 13, Green BR and Parson WW. (Eds.) Kluwer: 2003.

9. Faure S, Stern C, Guilard R and Harvey PD. J. Am. Chem. Soc. 2004; 126: 1253–1261.

10. Morandeira A, Vauthey E, Schuwey A and Gossauer A. J. Phys. Chem. A 2004; 108: 5741–5751.

11. Yeow EKL, Ziolek M, Karolczak J, Shevyakov SV, Asato AE, Maciejewski A and Steer RP. J. Phys. Chem. A 2004; 108: 10980–10988.

12. Leray I, Valeur B, Paul D, Regnier E, Koepf M, Wytko JA, Boudon C and Weiss J. Photochem. Pho-tobiol. Sci. 2005; 4: 280–286.

13. Scheuring S, Reiss-Husson F, Engel A, Rigaud J-L and Ranck J-L. EMBO J. 2001; 20: 3029–3035.

14. Scheuring S, Seguin J, Marco S, Levy D, Breyton C, Robert B and Rigaud J-L. J. Mol. Biol. 2003; 325: 569–580.

15. Scheuring S, Rigaud J-L and Sturgis JN. EMBO J.2004; 23: 4127–4133.

16. Iida K, Inagaki J-I, Shinohara K, Suemori Y, Ogawa M, Dewa T and Nango M. Langmuir 2005; 21:3069–3075.

17. Mukai K, Abe S and Sumi H. J. Phys. Chem. B1999; 103: 6096–6102.

18. Scholes GD and Fleming GR. J. Phys. Chem. B2000; 104: 1854–1868.

19. Ritz T, Park S and Schulten K. J. Phys. Chem. B2001: 8259–8267.

20. Kimura A and Kakitani T. J. Phys. Chem. B 2003; 107: 7932–7939.

21. Fleming GR and Scholes GD. Nature (London, UK)2004; 431: 256–257.

22. Timpmann K, Woodbury NW and Freiberg A. J. Phys. Chem. B 2000; 104: 9769–9771.

23. Tretiak S, Middleton C, Chernyak V and Mukamel S. J. Phys. Chem. B 2000; 104: 9540–9553.

24. Zazubovich V, Jankowiak R and Small GJ. J. Lumin.2002; 98: 123–129.

25. Borisov AY and Nikulova AN. Biofi zika 2004; 49:653–658.

26. Kyndt JA, Meyer TE and Cusanovich MA. Photo-chem. Photobiol. Sci. 2004; 3: 519–530.

27. Papagiannakis E, Van Stokkum IHM, Vengris M, Cogdell RJ, van Grondelle R and Larsen DS. J. Phys. Chem. B 2006; 110: 5727–5736.

28. Papagiannakis E, Vengris M, Valkunas L, Cogdell RJ, van Grondelle R and Larsen DS. J. Phys. Chem. B 2006; 110: 5737–5746.

29. Polli D, Cerullo G, Lanzani G, De Silvestri S, Hashimoto H and Cogdell RJ. Biophys. J. 2006; 90:2486–2497.

30. Cerullo G, Manzoni C, Lüer L and Polli D. Photo-chem. Photobiol. Sci. 2007; 6: 135–144.

31. Kakitani Y, Akahane J, Ishii H, Sogabe H, Nagae H and Koyama Y. Biochemistry 2007; 46: 2181–2197.

32. Niedzwiedzki D, Koscielecki JF, Cong H, Sullivan JO, Gibson GN, Birge RR and Frank HA. J. Phys. Chem. B 2007; 111: 5984–5998.

33. He Z, Sundström V and Pullerits T. FEBS Lett.2001; 496: 36–39.

34. Polivka T, Herek JL, Zigmantas D, Akerlund HE and Sundström V. Proc. Natl. Acad. Sci. USA. 1999; 96: 4914–4917.

35. Polivka T, Zigmantas D, Herek JL, He Z, Pascher T, Pullerits T, Cogdell RJ, Frank HA and Sundström V. J. Phys. Chem. B 2002; 106: 11016–11025.

36. Krueger BP, Scholes GD, Gould IR and Fleming GR. Phys. Chem. Comm. 1999; 2: 34–40.

37. Wohlleben W, Buckup T, Herek JL, Cogdell RJ and Motzkus M. Biophys. J. 2003; 85: 442–450.

38. Faure S, Stern C, Guilard R and Harvey PD. Inorg. Chem. 2005; 44: 9232–9241.

39. Harvey PD. In The Porphyrin Handbook, Vol. 18, Kadish KM, Smith KM and Guilard R. (Eds.) Aca-demic Press: San Diego, 2003; pp 63–250.

40. Pognon G, Wytko JA, Harvey PD and Weiss J. Chem. Eur. J. 2009; 15: 524–535.

41. Trissl H-W, Law CJ and Cogdell RJ. Biochim. Bio-phys. Acta 1999; 1412: 149–172.

42. Leray I, Valeur B, Paul D, Regnier E, Koepf M, Wytko JA, Boudon C and Weiss J. Photochem. Pho-tobiol. Sci. 2005; 4: 280–286.

43. Morandeira A, Vauthey E, Schuwey A and Gossauer A. J. Phys. Chem. A 2004; 108: 5741–5751.


DOI: 10.1142/S108842461000174X



INTRODUCTION

Palladium complexes with weakly coordinating anions have been known to activate substrates in organic transformations [1]. As an example, μ-(dihydroxo) and μ-(dialkoxo)dipalladium(II) complexes have been draw-ing much interest and their usage as catalyst for substitu-tion, carbonylation and Mannich-type reactions is known [2]. We have reported the synthesis and the reactions of unique μ-(dihydroxo) dipalladium(II) bisporphyrin (2)that was obtained by the reaction of dicationic bis(aqua)Pd(II) complexes of N21,N22-etheno bridged porphyrin (1) with NaOH or triethyamine (Scheme 1) [3]. Since palladium(II) porphyrin is a coordinatively saturated metal complex having a tetradentate ligand, it is not useful for metal-catalyzed reactions that require coordination sites for substrates. By using N21,N22-etheno bridged porphyrin

as a bidentate ligand, we have been investigating the coordination chemistry and organometallic chemistry of palladium porphyrin complexes [4], and showed that the N21,N22-bridged porphyrin ligand strongly coordinated to Pd and infl uenced the dynamic behavior of the coexisting ligand [5, 6]. It is of interest to make use of N21,N22-bridgedporphyrins considering a role of palladium catalysts with bidentate ligands in modern organic synthesis, though there is concern about palladium porphyrins with regard to their photophysical and electrochemical properties [7].

In this paper we report the application of the μ-dihydroxo complex of dimeric palladium(II) porphy-rin for photooxidation of phenol derivatives. Though the oxidation of several substrates with porphyrin ana-logs as photosensitizers have been developed [8, 9], the oxidation of phenol derivatives is of special inter-est because the oxidized products of phenol derivatives such as quinone derivatives are known to be important chemicals in the pharmaceutical and dye industries [10], and because some phenol derivatives such as chlorinated or p-alkylated phenols are known to be pollutants that accu-mulate in the environment and thus sustainable methods

Photooxidation of phenol derivatives usingμ-(dihydroxo)dipalladium(II) bisporphyrin complex

Yuko Takao*a , Toshinobu Ohnoa, Kazuyuki Moriwakia, Fukashi Matsumotoa and Jun-ichiro Setsune*b

a Osaka Municipal Technical Research Institute, Joto-ku, Osaka 536-8553, Japanb Department of Chemistry, Graduate School of Science, Kobe University, Nada-ku, Kobe 657-8501, Japan

Received 24 October 2008Accepted 10 February 2009

ABSTRACT: μ-(dihydroxo)dipalladium(II) complex with N21,N22-etheno bridged tetraphenylporphyrin ligand was employed in the catalytic photooxidation of phenol derivative in aerated homogeneous solution with visible light irradiation. The Pd complex promoted the degradation of p-tert-butylphenol as well as Cu phthalocyanine under basic conditions and it worked as a photosensitizer even in neutral conditions in contrast with Cu phthalocyanine. Some phenol derivatives afforded corresponding quinones selectively in this photooxidation. The present Pd complex coordinated by the bidentate porphyrin ligand showed higher photosensitizing ability than tetraphenylporphyrin free-base and ordinary tetradentate tetraphenylporphyrin palladium(II) complex for photooxidation of 2,6-di-tert-butylphenol. This bidentate Pd complex showed higher light durability even under the conditions where photodegradation of porphyrin free-base and tetradentate Pd porphyrin were observed by extensive irradiation.

KEYWORDS: bridged porphyrin, μ-(dihydroxo)dipalladium complex, photooxidation, phenol derivative.


*Correspondence to: Yuko Takao, email: [email protected], fax: +81 06-6963-8049 and Jun-ichiro Setsune, email: [email protected], fax: +81 078-803-5770


PHOTOOXIDATION OF PHENOL DERIVATIVES USING μ-(DIHYDROXO) DIPALLADIUM(II) BISPORPHYRIN COMPLEX 65

for their decomposition are needed [11–13]. So oxi-dation of several phenol derivatives has been studied using porphyrins or phthalocyanines in the presence of various oxidants, for example, oxygen [8b, 14], hydrogen peroxide [11, 12, 15] or potassium mono-peroxysulfate [11, 15]. Furthermore, photooxidations using a singlet oxygen photosensitizer are regarded as a friendly process for environment by considering minimum usage of toxic chemicals, use of oxygen from air and visible light which has the largest energy den-sity in solar radiation as unlimited natural resources. A number of examples have been reported for the use of classic dyes such as Methylene Blue and Rose Bengal as photosensitizers in these reactions. However, there is a problem of photodegradation of sensitizer upon exten-sive irradiation. Thus it is important to develop effi cient and stable photosensitizers [10, 16]. With this purpose in mind, effective photooxidations using new phthalo-cyanine and porphyrin derivatives have been developed [8, 13, 17].

Here, the photooxidations of some phenol derivatives using μ-dihydroxo complex of dimeric palladium(II) porphyrin are described, and the oxidation activity and photostability of the bidentate porphyrin complex are compared with those of a tetradentate normal metal por-phyrin and a free-base porphyrin.

EXPERIMENTAL

Materials and synthesis

μ-(dihydroxo)dipalladium(II) complex with N21,N22-(diphenyletheno) bridged TPP (tetraphenylporphyrin) dimer (2) was prepared by the reaction of dicationic bis(aqua)Pd(II) complexes of N21,N22-etheno bridged porphyrin with NaOH or triethyamine according to the pre-viously reported procedures [3]. Bis(perchlorato)bis(aqua)palladium(II) complex of N21,N22-(diphenyletheno)

bridged TPP (1) was also prepared accor-ding to a previously reported procedure [4]. Tetradentate (tetraphenylporphyrinato)Pd(II) [(TPP)Pd] was prepared as follows: A solution of tetraphenylporphyrin [TPPH2] (0.1 mmol) in CHCl3 (14 mL) was added to a solution of Pd(OAc)2 (0.2 mmol) in propionic acid (10 mL). The mixture was refl uxed for 6 h, then evaporated, followed by column chroma-tography on silica gel with dichloromethane. The product was recrystallized from dichlo-romethane and n-hexane to afford (TPP)Pd as a red solid (yield: 71%). Solvents were puri-fi ed prior to use by conventional methods. Cupper(II) phthalocyanine-3,4′,4″,4′″-tetra-sulfonic acid, tetrasodium salt [(Pc)Cu] and

other chemicals were of reagent grade. CDCl3 was passed through basic Al2O3 before use.

Measurement1H NMR spectra were recorded on JEOL AL-300 spec-

trometer. Chemical shifts were referenced with respect to (CH3)4Si (0 ppm) as an internal standard. Gas chromatography was carried out using a HiCap-CBP1 (25 m × 0.53 mm × 1 μm) Shimadzu capillary column on a Shimadzu-14B chromatography. GC analysis was run at 140°C; detector temperature 250°C, injector tem-perature 250°C. Absorption spectra were measured on Shimadzu UV-3100B instrument.

Photoreactivity study

The solution containing a substrate and a sensitizer was placed into a 1 cm optical cell with external water cooling and fi lled with oxygen gas, and then irradiated by a 500 W xenon lamp through cut-fi lters (350 nm < λ< 800 nm). In the case of measurement by GC, 3 mL of acetonitrile solution containing substrate (0.5 mM) and a sensitizer (5 × 10-5 M) was irradiated. A 0.05 mL amount of the solution was diluted 3.5 times and passed through a little silica gel, and then analyzed. In the case of obser-vation by 1H NMR spectrometry, a mixture of a substrate (3 mM) and a sensitizer (0.03–0.3 mM) in a deuterated solvent was irradiated. A 0.5 mL amount of sample solu-tion was analyzed and content of product was calculated from integral values of signals.

Photostability study

To study the photostability of sensitizers, 0.3 mL of the CH2Cl2 solution containing a sample (2.5 × 10-5 M) was placed into a 1 mm optical cell with external cool-ing, and irradiated by a xenon lamp. The intact optical cell was set in the spectrophotometer, and the absorbance at the Soret band was measured to monitor the change of concentration of samples.

Scheme 1. Preparation of μ-(dihydroxo)dipalladium(II) complex with N21,N22-(diphenyletheno) bridged porphyrin dimer


66 Y. TAKAO ET AL.


The photocatalytic activities of the compound 1, 2,(TPP)Pd, TPPH2 and copper phthalocyanine ((Pc)Cu) were examined in the photooxidation of some phenol derivatives such as 3 and 5 as shown in Scheme 2. At fi rst, the solution of p-tert-butylphenol (3a; 0.5 mM) in 3:5 (v/v) water-acetonitrile binary mixture containing photosensitizers (0.05 mM) was irradiated by a xenon lamp (350 nm < λ < 800 nm). The concentration of 3awas measured by GC analysis to monitor the photodegra-dation of 3a (Fig. 1).

Addition of complex 2 without photoirradiation resulted in little change in the concentration of 3a after3 h (fi lled square, Fig. 1). But under photoirradiation the concentration of 3a decreased rapidly, especially in the presence of NaOH (open circle, Fig. 1). The effi ciency of complex 2 was compared with that of (Pc)Cu and they similarly promoted photodegradation of 3a under basic conditions (open triangle, Fig. 1). On the other hand, it is noteworthy that complex 2 showed 94% of photodegradation of 3a (fi lled circle, Fig. 1) though (Pc)Cu showed only 20% of photodegradation (fi lled triangle, Fig. 1) after 3 h under neutral conditions. This result means that 2 works as a photosensitizer without adding some bases and is suitable for water treatment; e.g. photodegradation of phenols in normal water.

In the case of photooxidation of 3a oxidized products were of complex mixtures, while in the photooxidation of 2,6-di-tert-butylphenol (3b) the conversion of substrate to products was confi rmed by monitoring with 1H NMR spectroscopy. Figure 2 shows the 1H NMR spectrum on the photooxidation process of 3b in acetonitrile after 15 min irradiation in the presence of 2. Then it was proved that almost single product was formed selectively as an

initial product under such conditions. Although 2,6-di-tert-butyl-1,4-benzoquinone (4b) and 3,3′,5,5′-tetra-tert-butyl-4,4′-diphenoquinone were formed as oxidized products in thermal oxidation of 3b with porphyrin com-plexes [13], only one product was formed in this photo-oxidation and identifi ed as 4b by comparing the chemical shifts of the product with those of the standard 2,6-di-tert-butyl-1,4-benzoquinone. The infl uence of the amount of complex 2 as a photosensitizer on the conversion of 3bto 4b was also examined after 60 min of irradiation in acetonitrile. When 2 was added to the solution of 3b in the molar ratio of 0.1, 0.01 and 0.001 to 3b, the formation of 4b was 94%, 73% and 25%, respectively.

The photooxidation effi ciency of sensitizers was evalu-ated by the measurement of formation of 4b in the photo-oxidation of 3b. Table 1 shows the yield of 4b determined Scheme 2. Photooxidation of some phenol derivatives

Fig. 1. GC analysis of photosensitized degradation of 3a (5 ×10-4 M) in MeCN-H2O with irradiation with a 500 W xenon lamp at 80 cm distance. [photosensitizer] = 5 × 10-5 M; [NaOH] = 5 × 10-3 M

Fig. 2. 1H NMR spectrum in CD3CN of the reaction mixture after irradiating 3b (3 × 10-3 M) with a 500 W xenon lamp for 15 min in acetonitrile in the presence of 2 (0.1 molar equiv.)


PHOTOOXIDATION OF PHENOL DERIVATIVES USING μ-(DIHYDROXO) DIPALLADIUM(II) BISPORPHYRIN COMPLEX 67

by 1H NMR spectrum after 40 min irradiation. The best result was obtained (90%) under the conditions in which 10 mol.% of Pd atom in the form of 2 was added into acetonitrile solution. For the purpose of comparison of the photooxidation effi ciency with other related porphy-rin derivatives, the molar ratio of the sensitizers to 3b was decreased to 0.01 and chloroform was added to the solu-tion (7%) by considering the solubility of other sensitiz-ers such as TPPH2. The conditions resulted in decrease of yield of 4b from 56% by using 2 to 39, 24, and 44% yield by using (TPP)Pd, TPPH2 and 1, respectively, while 3bdid not react in the absence of sensitizers under otherwise the same conditions. When aqueous NaN3 or β-carotenewas added to the acetonitrile solution as a singlet oxy-gen scavenger, the photooxidation of 3b was inhibited. These experiments might suggest that singlet oxygen is involved in the present oxidation reactions.

The photooxidation of some other phenol derivatives such as 1,2,4-trimethylphenol (3c) and naphthols (5) were also carried out because the corresponding quinones 4c and 6 are regarded as intermediates for the syntheses of bio-chemicals, physiologically active substances or starting materials for preparing antiviral and antitumor agents [10a]. The results of photooxidation in the presence of 5 mol.% of 2 under some conditions are shown in Table 2. The oxida-tion products, 4c, 6a and 6b, were identifi ed on the basis of the reported spectroscopic data [10]. These photooxidation yields were dependent on the structures of substrates and reaction solvents. It was found that reactions in MeCN were much faster than those in CHCl3 and DMSO by monitoring with 1H NMR spectroscopy, while 3b did not react in the absence of sensitizers in both MeCN and CHCl3 under the same irradiation conditions. 3b and 5a could be converted to the corresponding quinones in excellent yields (96%) though the conversion of 3c and 5b was low.

The photostability of porphyrin sensitizers was evaluated by measuring the relative absorbance at the Soret band with irradiation time. The decreasing of the relative absorbance of 2 by irradiation was also compared with

those of TPPH2 and (TPP)Pd under the same conditions. As shown in Fig. 3, complex 2 was the most photostable and remained unchanged after 100 min irradiation, while TPPH2 was almost completely photodecomposed under the same conditions. This remarkable difference in the photostability among 2, TPPH2 and (TPP)Pd was not refl ected in the photoactivity examined in Table 1, because the yield data were taken while photosensitizers are alive.

Thus, the photooxidation of phenol derivatives pro-ceeded under mild conditions in neutral solution at room temperature using complex 2. It is noteworthy that it works without a base as a photosensitizer probably because the OH ligand of the Pd2(OH)2 core of 2 acted as a base, which was confi rmed from previous experiments that 2 reacted with an acidic substrate such as methyl 3-butenoate to give (p-methoxycarbonylallyl)Pd complexes by deprotonation under neutral conditions [3].

Table 1. Dependence of the photooxidation efficiency on the sensitizer and solventa

Por. Molar ratio Solvent Yield of(Por./3b) 4b, %b

2 0.05 CH3CN 90

2 0.005 CH3CN 68

2 0.005 CH3CN(7%CHCl3) 56

(TPP)Pd 0.01 CH3CN(7%CHCl3) 39

TPPH2 0.01 CH3CN(7%CHCl3) 24

1 0.01 CH3CN(7%CHCl3) 44

2 c 0.05 CH3CN(20%H2O) 0

2 d 0.05 CH3CN(33%CHCl3) 0

a [3b] = 3 × 10-3 M, irradiation with a 500-W xenon lamp at 80 cm distance, b after 40 min, c NaN3 was added, d [3b] = 1 × 10-3 M, β-carotene was added.

Table 2. Photooxidation of some phenol derivativesa

Substrate Solvent Time, min Product yield, %

3b MeCN 40b 3

3b MeCN 40c 0

3b MeCN 60 96

3b CHCl3 120 48

3b DMSO 40 6

3c MeCN 40 15

3c CHCl3 120 13

3c DMSO 120 0

5a MeCN 5 96

5b DMSO 120 25

a [Substrate] = 3 × 10-3 M, [2] = 1.5 × 10-4 M, irradiation with a 500 W xenon lamp at 80 cm distance, b not irradiated c no sensitizer.

Fig. 3. The change in the absorbance at the Soret bands of porphyrins (2.5 × 10-5 M) with time through irradiation with a 500 W xenon lamp at 80 cm distance in CH2Cl2


68 Y. TAKAO ET AL.

Photooxidation of phenol derivatives was promoted by the Pd complex of the bidentate porphyrin using visible light energy. Moreover, complex 2 showed a slightly more superior effi ciency and much higher photodurability as a photosensitizer than tetradentate (TPP)Pd and free-base TPPH2. The UV-vis absorption spectrum of 2 in Fig. 4 shows split Soret bands and absorption bands extending to 700 nm in the Q-band region. This spectral pattern is very different from those of TPPH2 and (TPP)Pd and the photophysical properties should be of interest in relation to the present photooxidation reactions. We are planning to study the photophysical properties of 2.

REFERENCES

1. Pisano C, Consiglio G, Sironi A and Moret M. J. Chem. Soc., Chem. Commun. 1991; 421–423.

2. a) Mann G, Incarvito C, Rheingold AL and Hartwig JF. J. Am. Chem. Soc. 1999; 121: 3224–3225. b) Grushin VV and Alper H. Organometallics 1993; 12: 1890–1901. c) Hagiwara E, Fuji A and Sode-oka M. J. Am. Chem. Soc. 1998; 120: 2474–2475. d) Fujii A, Hagiwara E and Sodeoka M. J. Am. Chem. Soc. 1999; 121: 5450–5458.

3. Takao Y, Takeda T, Tanikawa S and Setsune J. J. Porphyrins Phthalocyanines 2003; 7: 521–525.

4. a) Takao Y, Takeda T, Watanabe J and Setsune J. Organometallics 1999; 18: 2936–2938. b) Takao Y, Takeda T, Watanabe J and Setsune J. Organometal-lics 2003; 22: 233–241.

5. a) Takao Y, Takeda T, Miyashita K and Setsune J. Chem. Lett. 1996: 761–762. b) Takao Y, Takeda T, Miyashita K and Setsune J. Bull. Chem. Soc. Jpn.1998; 71: 1327–1335.

6. Takao Y, Takeda T and Setsune J. Bull. Chem. Soc. Jpn. 2003; 76: 1549–1553.

7. a) Richeter S, Jeandon C, Gisselbrecht JP, Graff R, Ruppert R and Callot HJ. Inorg. Chem. 2004;

43: 251–263. b) Setsune J. J. Porphyrins Phthalo-cyanines 2004; 8: 93–102. c) Pereira MM, Ruano F, Azenha MEDG, Burrows HD, Miguel MGM, Doug-las P and Eaton K. J. Porphyrins Phthalocyanines2006; 10: 87–95.

8. a) Gerdes R, Bartels O, Schneider G, Wöhrle D and S-Ékloff G. Polym. Adv. Technol. 2001; 12: 152–160. b) Wöhrle D, Suvorova O, Gerdes R, Bartels R, Lapok L, Baziakina N, Makarov S and Slodek A. J. Porphyrins Phthalocyanines 2004; 8: 1020–1041. c) Ribeiro SM, Serra AC and Gonsalves AMD’AR. Tetrahedron 2007; 63: 7885–7891.

9. a) Tsaryova O, Semioshkin A, Wöhrle D and Bregadze VI. J. Porphyrins Phthalocyanines 2005; 9: 268–274. b) Kasuga K, Imai M, Irie H, Tanaka H, Ikeue T, Handa M, Wada S and Sugimori T. J. Porphyrins Phthalocyanines 2006; 10: 1212–1218.

10. a) Murtinho D, Pineiro M, Pereira MM, Gonsalves AMD’AR, Arnaut LG, Miguel MG and Burrows HD. J. Chem. Soc., Perkin Trans. 2 2000; 2441–2447. b) Suchard O, Kane R, Roe BJ, Zimmermann E, Jung C, Waske PA, Mattay J and Oelgemöller M. Tetrahedron 2006; 62: 1467–1473.

11. Kasuga K, Mori K, Sugimori T and Handa M. Bull.Chem. Soc. Jpn. 2000; 73: 939–940.

12. a) Labat G, Seris J-L and Meunier B. Angew. Chem. Int. Ed. Engl. 1990; 29: 1471–1473. b) Sorokin A and Meunier B. J. Chem. Soc., Chem., Commun.1994; 1799–1800.

13. a) Kasuga K, Miyazako T, Sugimori T and Handa M. Inorg. Chem. Commun. 2003; 6: 807–809. b) Kluson P, Drobek M, Krejcikova S, Krysa J, Kalaji A, Cajthaml T and Rakusan J. Appl. Catal., B2008; 80: 321–326.

14. a) F-Rio M, Pujol D, B-Charreton C, P-Fauvet M and Gaudemer A. J. Chem. Soc. Perkin. Trans. 1 1984; 1971–1979. b) Hassanein MT, Gerges SS, Abdo MA and Elkhalafy SH. J. Porphyrins Phthalocyanines 2005; 9: 621–625.

15. Sanchez M, Hadash A, Fell RT and Meunier B. J. Catal. 2001; 202: 177–186.

16. a) Wiehe A, Stollberg H, Runge S, Paul A, Senge MO and Röder B. J. Porphyrins Phthalocyanines2001; 5: 853–860. b) Mosinger J, Kliment Jr. V, Sejbal J, Kubát P and Lang K. J. Porphyrins Phtha-locyanines 2002; 6: 514–526. c) Scalise I and Durantini EN. J. Photochem. Photobiol. A 2004; 162: 105–113. d) Engelmann FM, Mayer I, Araki K, Toma HE, Baptista MS, Maeda H, Osuka A and Furuta H. J. Photochem. Photobiol. A 2004; 163:403–411.

17. a) Gerdes R, Wöhrle D, Spiller W, Schneider G, Schnurpfeil G and Schulz-Ekloff G. J. Photochem. Photobiol. A 1997; 111: 65–74. b) Kasuga K, Fujita A, Miyazako T, Handa M and Sugimori T. Inorg. Chem. Commun. 2000; 3: 634–636. c) Hu M, Xu Y and Xiong Z. Chem. Lett. 2004; 1092–1093.

Fig. 4. UV-vis absorption spectrum of 2 in MeCN


DOI: 10.1142/S1088424610001751



INTRODUCTION

Past and present interest in the metallated and metal- free phthalocyanines has been motivated by their exten-sive applications to numerous fields of chemistry [1]. The use of phthalocyanines bonded to organic polymers as catalysts of reactions in homogeneous and hetero-geneous phase figures among these fields. Works in the literature on polymers derivatized with transition metal complexes have followed two general strategies for the preparation of these polymers [1t, 2–4]. One is

based on the polymerization of pendent groups present in a ligand of the complex [3]. In the other, the phtha-locyanine is covalently bonded to a preformed polymer [2, 4]. In the latter, pendent phthalocyanine groups can change the morphology of the preformed polymer. For example, pendent -CO2RhIII(pc) groups in the derivatized poly(acrylate) stabilize the strand’s hypercoiled structure [4]. Moreover, the incorporation of the phthalocyanines in a preformed polymer has a marked effect on the ther-mal and photochemical reactivity of the phthalocyanine pendants. In the previously studied polyacrylate deriva-tized with -CO2RhIII(pc) [4], redox reactions with radi-cals and transition metal complexes mostly changed the oxidation state of the phthalocyanine ligand and formed

Morphology and radical reactions of Cu(II) and Co(II)sulfonated phthalocyanines covalently linked to poly(ethyleneamide)

Gustavo T. Ruiza, Alexander G. Lappinb and Guillermo Ferraudi*c

a Radiation Research Building, University of Notre Dame, Notre Dame, IN 46556, USAb Chemistry Department, University of Notre Dame, Notre Dame, IN 46556, USAc Chemistry Department, Radiation Research Building, University of Notre Dame, Notre Dame, IN 46556, USA


ABSTRACT: The tetrasulfonated CuII(tspc)4-, tspc = 4,4′,4″,4″′-phthalocyaninetetrasulfonate, and the trisulfonated CoII(trspc)3-, trspc = n,n′,n″-phthalocyanine-trisulfonate and n, n′ and n″ indicate the sulfonated positions of the various isomers, were covalently linked to a polyethyleneimine, Mn ~ 10000 and Mw ~ 25000, backbone. A fraction of the amine groups in a strand was converted to CuIIpc(-SO3)3

(-SO2N<)\3- or CoIIpc(-SO3)2(-SO2N<)\2- pendants and the remaining were acetylated. Strands of the polymers, poly(K3CuIItspc) and poly(K2CoIItrspc), formed spherical bundles with diameters ~1000 nm for poly(K3CuIItspc) and ~100 nm for poly(K2CoIItrspc). The stability of the bundles is metal-dependent. Poly(K2CoIItrspc) forms the most stable bundles with respect to hydrolytic and redox reactions. ESR and optical spectroscopies as well as reactions of the pendants with pulse radiolytically generated radicals, namely e-

sol, C•H2OH, (CH3)2COHC•H2, CO2

•-, N3• and SO4

•-, revealed a distribution of pendants between isolated and aggregated forms in a bundle. The reduction of the Cu(II) to Cu(I) in aqueous solutions of the poly(K3CuIItspc) was associated with pendants present in a mostly hydrophobic environment. Unstable phthalocyanine pendants with CoIII-carbon bonds are formed when CoIIpc(-SO3)2(-SO2N<)\2-

reacted with C-centered radicals, (CH3)2COHC•H2 and C•H2OH. The species with CoIII-carbon bonds were precursors to the formation of CoIpc(-SO3)2(-SO2N<)\3- pendants. The redox reactions of the pendants in these polymers are compared with those of K4CuIItspc and K3CoIItrspc.

KEYWORDS: phthalocyanine-derivatized poly(ethylenimine), copper, sulfonated phthalocyanine, cobalt, phthalocyanine radicals, pulse radiolysis, reaction kinetics, reaction intermediates.

*Correspondence to: Guillermo Ferraudi, email: [email protected], tel: +1 (574)-631-7676


70 G.T. RUIZ ET AL.

pendent phthalocyanine radicals. The radical attack on the phtalocyanine ligand instead of the metal center is a consequence of the unfavorable redox potentials of the Rh(II)/Rh(III) and Rh(III)/Rh(IV) couples with respect to the ligand couples.

In some applications, the metal-centered redox reac-tions of the phthalocyanine pendants can be more advan-tageous than the phthalocyanine-centered reactions [5, 6]. In contrast to the Rh(III) in -CO2RhIII(pc) pendants of the derivatized poly(acrylate), the more positive redox potentials of the Cu(II) and Co(II) metal centers in the corresponding phthalocyanines favor the metal-centered redox reactions [1s, 7]. A potential shortcoming of a change in the oxidation state of the metal center is that it can make it ligand-substitution labile and the ensuing solvolysis may detach the pendants from the polymeric backbone. One way to address this possible cause of polymer instability is to have the phthalocyanine ligand covalently linked to the polymeric backbone. To explore some of these points, polymers containing pendent Cu(II) and Co(II) phthalocyanines, covalently bonded through a sulfonamide group to a polyamide backbone, were pre-pared and their chemical reactions were investigated. The chemistry of their nanometric aggregates in aqueous solution and the species formed when they reacted with radicals were investigated using steady state and time-resolved spectroscopies.

EXPERIMENTAL

Unless explicitly stated, the spectroscopy and redox reactions of the polymeric compounds described below were investigated with freshly prepared aqueous solu-tions where KOH was added to make the pH = 10.

Pulse-radiolytic procedures

Pulse radiolysis experiments were carried out with a model TB-8/16-1S electron linear accelerator following previously described methods [4, 8]. The instrument and computerized data collection for time-resolved UV-vis spectroscopy and reaction kinetics have been described elsewhere in the literature [9]. Thiocyanate dosimetry was carried out at the beginning of each experimental session. The details of the dosimetry have been reported elsewhere [9, 10]. The procedure is based on the con-centration of (SCN)2

•- radicals generated by the electron pulse in a N2O saturated 10-2 M SCN- solution. In the pro-cedure, the calculations were made with G = 6.13 and an extinction coefficient, ε = 7.58 × 103 M-1.cm-1 at 472 nm, for the (SCN)2

•- radicals [10]. In general, the experiments were carried out with doses that in N2 saturated aqueous solutions resulted in (2.0 ± 0.1) × 10-6 M to (6.0 ± 0.3) × 10-6 M concentrations of e-

sol. In these experiments, solu-tions were deaerated with streams of the O2-free gas, N2

or N2O that was required for the experiment. Reactions of

the radiolytically generated OH• radicals with N3-, SO4

2-,methanol or 2-propanol were used for the preparation of reactive radicals, e.g. N3

•, CO2•-, SO4

•-, C•H2OH and (CH3)2COHC•H2 (Scheme 1). Other experiments, where the simultaneous generation of (CH3)2COHC•H2 and e-

sol

was desired, were conducted with N2-deaerated solutions containing 0.1 M (CH3)3COH as a scavenger of the radio-lytically generated OH• radicals.

In order to radiolyze a fresh sample with each radi-olytic pulse, an appropriate flow of the solution through the reaction cell was maintained during the experi-ment. Because the radiolyzed solutions had a pH = 10, the radicals C•H2O

- and C•H2OH radicals were pro-duced in a molar relationship 1:5. This mixture of the C•H2O

- and C•H2OH radicals produced at pH = 10 will be generically described as C•H2OH radicals in the fol-lowing sections. Other conditions used for the time-resolved spectroscopy of the reaction intermediates or in the investigation of the reaction kinetics are given in the Results section.

Electrophoresis

A commercial Owl separation system, apparatus model A1, was used for the electrophoretic analysis. The beds for the separations, a gel of 3.0% agarose in Owl’s 1 × TBE running buffer, pH = 8.3, had 80 cm3 slab-shaped indentations for the placing of the samples. Separations were made with the beds immersed in the 1 × TBE run-ning buffer under those current and voltages specified by the manufacturer for the apparatus. Comparisons were made between solutions of the pure or impure polymers, and blank solutions made with pure Cu(II) or Co(II) tetra- and tris-sulfonated phthalocyanines. Nearly the same concentrations of phthalocyanine were used in the sample and blank solutions. The slab-shaped indentation made in the bed with the Teflon combs provided with the apparatus accommodated volumes of sample too small for a preparative scale. A Teflon comb to make 10 cm3

slab-shaped indentations in the bed was house-made and used in preparative electrophoresis.

OH-

X•

OH•

+ N2O, + H+

radiolytic pulse

X-

- N2

X- = HCO2-, SO4

2-, N3-, CH3OH, (CH3)3COH;

X• = CO2•-, SO4

•-, N3•, C•H2OH, (CH3)2COHC•H2

Scheme 1. Flow chart descibing the radiolytic preparation of radical reactants


MORPHOLOGY AND RADICAL REACTIONS OF CU(II) AND CO(II) SULFO NATED PHTHALOCYANINES 71

TEM microscopy

A drop of the sample solution was placed on a 400 mesh cooper grid covered with collodion. It was kept on the grid 5 min and the excess liquid was then removed by placing the grid on filter paper. The preparations were investigated at 100 K in a transmission electronic micro-scope JEM 1200 EX II (JEOL) at the Servicio Central de Microscopía Electrónica of the Facultad de Ciencias Vet-erinarias, UNLP. Stock solutions of the polymers were prepared by dissolving the polymers in basic medium and diluting them as necessary for the preparation of the sample solutions.

Cyclic voltammetry

A computerized BAS CV-1B cyclic voltammograph apparatus with a three electrode cell was used for the electrochemical measurements. Platinum discs, pol-ished to a mirror finish, were used as working electrodes. Potentials were measured against a Ag/AgCl/KClsat ref-erence electrode. A 0.1 M concentration of NaClO4 was used as supporting electrolyte and oxygen was removed by bubbling purified N2 for at least 30 min prior to each experiment.

Spectroscopic measurements

ESR spectra were recorded with a Bruker ER 080 spectrometer fitted with a TE102 wavebridge and inter-faced to a Gateway 2000 computer. UV-vis-NIR spectra were recorded with a Shimadzu UV-3101PC spectropho-tometer. IR spectra were recorded with a Bio-Rad FTS 175 Spectrometer.

Materials

The Cu(II) tetrasulfonated phthalocyanine Na4[CuII(tspc)], tspc = 4,4′,4″,4″′-phthalocyaninetetrasulfonate, and CoI-Ipc, both from Eastman, were available from previous works. A procedure described in literature was followed for the sulfonating of the Co(II) phthalocyanines with chlorosulfonic acid [11–13]. The elemental analysis of sulfur and cobalt showed that the method produced tris-sulfonated Co(II) phthalocyanine, Na3[CoII(trspc)], trspc = n,n′,n″-phthalocyanine-trisulfonate where n, n′ and n″stand for the sulfonated positions of the isomers con-tained in the product. Calcd. for Na3CoC32N8H13O9S3, S: 10.96, Co: 6.72. Found, S: 11.20, Co: 6.35.

Chlorosulfonated phthalocyanines. Chlorosul-fonated derivatives of Na4[CuII(tspc)] and Na3[CoII(trspc)], namely CuII(chltspc) and CoII(chltrspc), were prepared by the thionylchloride-DMF method in the literature and used as obtained for the preparation of the polymers [13, 15]. Water free polyethyleneimine, Mn ~ 10000 and Mw ~ 25000, and acetylbromide from Aldrich were used as received.

Polyethyleneimine-amide, PAAMr where r = 0.07 or 0.2. Polyethyleneimine, 5 g (~0.12 mol of amine

groups) was dissolved in 500 cm3 of THF which already contained 27.6 g of K2CO3 (0.2 mol). The solution was stirred a couple of hours and chilled in an ice bath. Acetylbromide, 8 cm3 (13 g, 0.11 mol) or 4 cm3 (6.5 g, 0.055 mol) were dissolved in 80 cm3 of THF and added dropwise to the polyethyleneimine solution with a vig-orous stirring under a N2 atmosphere. The mixture was warmed up to room temperature, refluxed overnight and brought back to room temperature before exposing it to air. The solid was removed by filtration and the fil-trate, mainly THF, was discarded. In order to remove the polymer from the solid, it was thoroughly washed with EtOH. Rotovaporation of the ethanolic solution yield the solid products. The polymers were dried under vacuum overnight. Approximately ~95% yields of polymers, PAAMr, containing both amine and acetamide groups were obtained with this procedure. Elemental analyses were used for the evaluations of the number of amine groups relative to the amide groups in each preparation. The PAAMr polymers respectively obtained with the two masses of acetyl bromide had an average number r of amine groups relative to the number of amine plus amide groups, r = 0.07 (1 amine in 15 groups) and r = 0.2 (1 amine in 5 groups). In a larger scale preparation of PAAM0.07, 7.8 g of polyethleneimine and 45.5 g K2CO3 in 700 cm3 of THF and 12.5 cm3 of acetylbromide in 100 of THF were used instead of the quantities described above. The same steps of the previous preparation were fol-lowed until the solid was obtained by rotovaporation of the ethanolic solution. To further remove impurities, the solid was thoroughly washed with 15% (v:v) EtOH in CH2Cl2 and dried under vacuum.

Phthalocyanine containing polymers, poly-(K3CuIItspc). In the preparation of the polymer, CuII(chltspc), 10 g, was dissolved in 500 cm3 of CH2Cl2

containing 1.5 g of K2CO3. A solution containing 1.6 g of PAAM0.2 in 100 cm3 of 10% (v:v) EtOH in CH2Cl2 was added dropwise and with a vigorous stirring to the former solution. The mixture was first stirred at room tempera-ture for 10 h and then refluxed for 1 h. Rotovaporation of CH2Cl2 to dryness left ~13 g of a solid that was washed with 3 L of H2O until the filtrate became colorless. The insoluble solid, ~9 g, was suspended in 500 cm3 of 0.08 M K2CO3 and kept at 60 °C for 40 min. Electrophoresis revealed that a solid obtained after centrifugation of the suspension, ~1.9 g, was a mixture of monomeric phthalo-cyanine and the desired polymer. Preparative electropho-resis or GPC on column grade Sephadex G50 was used for the separation of the polymer from the mixture. A 95 cm3 volume of a saturated solution was loaded on a column, 64 cm in length and 3 cm in diameter, and eluted with H2O. The fraction containing the polymer eluted ahead of the monomeric phthalocyanines. Rotovapora-tion of the elutant to dryness and drying the solid under vacuum for 72 h yield 0.6 g of the pure polymer. The nature of the product was confirmed by UV-vis, IR and ESR spectroscopies and by electrophoresis. Anal. calcd.


72 G.T. RUIZ ET AL.

for the repeating unit, (I) and bottom of the Fig. 1, H44C50N13O15S4K3Cu, Cu, 4.62; S, 9.32. Found Cu, 4.31; S, 8.93.

Poly(K2CoIItrspc). This was prepared by a procedure similar to the one described above for the poly(K3CuIItspc)synthesis. CoII(chltrspc), 6.3 g, was dissolved in 500 cm3

of CH2Cl2 containing 0.9 g of K2CO3. A solution

containing 1.0 g of PAAM0.07 in 450 cm3 of 10% (v:v) EtOH in CH2Cl2 was added dropwise and with a vigorous stirring to the former solution. CH2Cl2 was added to the mixture until the total volume of the mixture was 1.5 L. The mixture was first stirred at room temperature for 10 h under a N2 atmosphere and then refluxed for 2 h. Rotovaporation of CH2Cl2 to dryness left a solid that was dried under vacuum for three days. The solid, ~7.5 g, was washed with 5.7 L of water at room temperature until the filtrate became colorless. To purify the polymer by GPC, a fraction of the remaining solid was dissolved in ~160 cm3 of a solution containing 6.1 mmol of KOH. The phthalocyanine solution, pH = 7, was divided in two 80 cm3 fractions. Each fraction was separately loaded on a column, 64 cm in length and 3 cm in diameter, of column grade Sephadex G50 and eluted with H2O. The fraction containing the polymer eluted ahead of the monomeric phthalocyanines. Rotovaporation of the elutant to dryness and drying the solid under vacuum for 72 h yield 120 mg of the pure polymer. The polymer obtained in this manner contained carbonate ions that were removed by recrystal-lization from warm 0.1 M CH3CO2H. The solid was kept at 50 °C under vacuum for 72 h. Acetate and elemental analyses showed about four acetate ions and one phthalo-cyanine pendant per 15 -CH2CH2N< groups in the repeat-ing unit, (II) in Fig. 1 without the acetates anions. Anal. calcd. for the repeating unit H138C102N24O31S3K2Co, H, 5.72; C, 50.42; N, 13.84; S, 3.96; Co, 2.43. Found H, 5.64; C, 49.97; N, 12.49; S, 3.46; Co, 2.55. The nature of the product was confirmed by UV-vis, IR and ESR spectroscopies and by electrophoresis.

Other materials and solvents were reagent grade and used without further purification. Solutions for the experiments described herein were prepared with triple distilled water.

RESULTS

The redox reactions of poly(K2CoIItrspc) and poly(K3CuIItspc) (Fig. 1) will be better understood after some consideration is given to the physical properties and morphologies of these materials. Therefore, the structural observations are communicated ahead of a study of the polymers redox reactions in the following sections.

Strand morphology

In order to ascertain the morphologies of poly(K2CoIItrspc) and poly(K3CuIItspc), TEM observa-tions were made on various preparations of the poly-mers. Rosettes with ~1000 nm diameter were observed in the preparations of poly(K3CuIItspc) (Fig. 2a). Smaller but similarly shaped structures with an average diam-eter of ~120 nm were observed in the preparations of poly(K2CoIItrspc). Because of the large size of these structures, they must be bundles of polymer strands which could be reminiscent of micelles in aqueous solution.

(II): M = Co(II), x + y = 14 and only one representative isom

(I): M = Cu(II), x + y = 4 and only one representative isomer

COCH3

COCH3

NH

CH2

CH2

--[(--NCH2CH2--)x--(--NCH2CH2 --)y]n--(--NCH2CH2--)--

N

N

N

NN

N

N

N

SO2

M

KSO3

KSO3

KO3S

COCH3

COCH3

NH

CH2

CH2

--[(--NCH2CH2--)x--(--NCH2CH2 --)y]n--(--NCH2CH2--)--

N

N

N

NN

N

N

N

SO2

M

KSO3

KO3S

Fig. 1. Average composition of the repeating unit of poly(K3CuIItspc), (I), and poly(K2CoIItrspc), (II), without the acetate anions. A balls and cylinders model of the repeating unit in poly(K3CuIItspc) is shown at the bottom of the fi gure



Preparations made from solutions of poly(K3CuIItspc)that were aged for 24 h showed elongated structures with ~1000 nm length nm and ~70 nm width (Fig. 2b). Cir-cular shapes, ~400 nm diameter, appear superimposed with the elongated structures in preparations made from solutions aged several days (Fig. 2c). The TEM observa-tions show, therefore, that the rosette-shaped structures of poly(K3CuIItspc) are unstable with respect to the elon-gated shapes over a 24 h period. Changes experienced by

the elongated structures over a period of days are most likely due to chemical reactions, such as the hydrolysis of the sulfonamide group in the poly(K3CuIItspc) poly-mer. In agreement with this interpretation, the chemical reactions were evidenced by changes in the optical spec-trum of the chromophores in the aged solutions and by electrophoresis experiments. Both showed the formation of free CuIItspc4- in concentrations that increased with the age of the solutions. The poly(K2CoIItrspc) proved to be more stable not showing such structural changes over several days. Due to the changing morphologies of the polymers some precautions were taken. In order to work with solutions of poly(K2CoIItrspc) and poly(K3CuIItspc)of known composition and morphologies, all the experi-ments described herein were carried out with freshly prepared solutions of the polymers and were completed when these solutions had aged less than several hours.

Pendants oligomers

ESR and UV-vis spectroscopies were used in a study of the state of the pendants in the bundle. The ESR spectra of powder samples of Na4[CuII(tspc)] and poly(K3CuIItspc)were recorded at 298 and 77 K. The respective spec-tra recorded at room temperature are shown in Fig. 3. Regardless of the temperature, the presence of weak hyperfine resonances was observed in the ESR spec-trum of poly(K3CuIItspc). Also the signal of the tetrago-nal Cu(II) metal center in the pc macrocycle is slightly broader in the polymer than in the powder spectrum of Na4[CuII(tspc)] [14]. Modeling of the polymer’s ESR spectrum suggested that the weak hyperfine signals and the broadening of the spectrum are due to the presence of a significant number of magnetically diluted CuIIpc(-SO3)3(-SO2N<)\3- pendants in the bundles of polymer. A large fraction of the pendants must be forming, however, π-π stacks similar to those of CuII(tspc)4- in homogeneous solution [7]. The ESR spectrum of the CoIIpc(-SO3)2

(-SO2N<)\2- pendants was recorded with powder samples

Fig. 2. (a) Rosettes with ~1000 nm diameter were observed in preparations of poly(K3CuIItspc). (b) Elongated structures with ~1000 nm length and ~70 nm width were observed in prepara-tions made from solutions of poly(K3CuIItspc) aged for 24 h. (c) Circular shapes, ~400 nm diameter, appear superimposed with the elongated structures in preparations made from solu-tions aged several days

Fig. 3. ESR powder spectra of (a) poly(K2CoIItrspc) at 77 K, (b) poly(K3CuIItspc), and (c) Na4[CuII(tspc)] at room tempera-ture. Markers indicate the calculated hyperfi ne positions of magnetically diluted CuIIpc(-SO3)3(-SO2N<)\3- pendants in the bundles of poly(K3CuIItspc)


74 G.T. RUIZ ET AL.

of poly(K2CoIItrspc) at 77 K and was in agreement with the spectra of axially symmetric low spin Co(II) com-plexes [14, 15]. By comparison to the ESR spectrum of CoIIpc, some broadening of the hyperfine lines and in the general shape of the spectrum can be attributed, like in the ESR spectrum of poly(K3CuIItspc), to similar stack-ing of pendants within the spherical bundles.

The presence of the stacks of pendants is evidenced also by the shape of the Q band in the UV-vis absorp-tion spectrum of poly(K3CuIItspc) (Fig. 4). In aqueous solution, the Q band in the spectrum of the pendants is broader than that of the monomeric Cu(tspc)4- [7, 16]. Superimposed with the Q band, a much sharper absorp-tion band at ~666 nm must be assigned to the fraction of CuIIpc(-SO3)3(-SO2N<)\3- pendants not forming the π-π stacks. In order to investigate the existence of a mobile association equilibrium between the monomeric and oligomeric CuIIpc(-SO3)3(-SO2N<)\3- pendants, the dependence of the absorbance of poly(K3CuIItspc) on the concentration of pendants was investigated with solutions of the polymer where the ionic strength was adjusted with 10-3 M NaClO4. A linear dependence of the absorbance on the pendant’s concentration was observed at λob = 217, 345, 666 and 753 nm when the concentration was equal to or less than 10-4 M. Such a linear behavior shows that the Beer’s law is obeyed over a 10-4 to 0 M concentra-tion range and that the distribution of pendants between monomeric and oligomeric forms is fixed by the bundles morphology and not by mobile equilibria.

The staking of pendants was also revealed by the UV-vis absorption spectrum of poly(K2CoIItrspc) in aqueous solu-tion (Fig. 5). In the UV-vis spectrum of poly(K2CoIItrspc)as in the spectrum of poly(K3CuIItspc), the shape of the Q band is accounted for a large fraction of CoIIpc(-SO3)2

(-SO2N<)\2- pendants associated in stacks.

Cyclic voltammograms of the poly(K2CoIItrspc) and poly(K3CuIItspc) were respectively recorded in aqueous solutions, saturated in either polymer and buffered at pH = 10. The corresponding voltammograms exhibited broad anodic waves at ~1.26 and 1.33 V vs. NHE near the potentials where the one-electron oxidation of the phthalocyanine ligand is expected [1s, 7]. No discernible catodic waves were observed with sweep rates between 90 and 200 mV.s-1. The decay of the phthalocyanine radi-cals formed in the anodic process account for the absence of a cathodic wave. Waves observed at ~1.0 V vs. NHE were attributed to the reduction of the product formed by the decay of the phthalocyanine radicals. The results of the cyclic voltammograms with poly(K2CoIItrspc) and poly(K3CuIItspc), contrasted against those of CuII(tspc)4-,support the existence of aggregates of the phthalocyanine pendants. Indeed features in the voltammograms of the polymers are similar those in the cyclic voltammogram of Cu(tspc)4- saturated solutions buffered at pH = 10 where oligomers of CuII(tspc)4- are in equilibrium with mono-mer. The cyclic voltammogram of the Cu(tspc)4- solution showed anodic and cathodic waves, ΔEpeak = 0.052 V, with a half wave potential, E1/2 = 1.237 V vs. NHE. However the observed waves were considerably wider than those recorded under experimental conditions where only the monomeric species exists in solution.

Ligand vs. metal-centered redox reactions of poly(K3CuIItspc)

Fast reactions of the pulse radiolytically generated radicals e-

sol, CO2•- and C•H2OH reduced poly(K3CuIItspc)

(Table 1). The solutions of the polymer used for the kinetic and spectroscopic studies contained a 10-4 M concentration of CuIIpc(-SO3)3(-SO2N<)\3- pendants. To study the reactions of CO2

•- and C•H2OH radicals, N2O-saturated solutions containing either 0.1 M in NaHCO2 or 0.1 M CH3OH in 0.1 M NaClO4 were used in pulse radi-olysis. N2-saturated solutions containing the polymer and 0.1 M NaClO4 were used in the study of the e-

sol reactions.

Fig. 4. Electronic spectra of (a) poly(K3CuIItspc) and (b) [CuII(tspc)]4- in aqueous solution. The inset shows the depen-dence of 666 nm absorbance on the pendant concentration. It is consistent with an intra-bundle and/or intra-strand stacking of the pendants

Fig. 5. Electronic spectra of (a) poly(K2CoIItrspc) and (b) [CoII(trspc)]2- in aqueous solution



In contrast to e-sol and C•H2OH which only

reduced the phthalocyanine ligand, the CO2•-

radical reduces both the metal center and the ligand of the pendants with an overall rate constant, k = 8.6 × 108 M-1.s-1 (Scheme 2). The different positions of their absorption bands in the visible region of the UV-vis spectrum of CuIpc(-SO3)3(-SO2N<)\4- and the ligand-radical, CuIIpc•(-SO3)3(-SO2N<)\4- (Fig. 6a), made possible the observation of these spe-cies transformations [7, 16, 20]. In a longer time-scale ranging from ms to s, the decay of the CuIpc(-SO3)3(-SO2N<)\4- pendants further bleached the Q-band and increased the absor-bance at ~500 nm (Fig. 6b). This change in the solution’s spectrum is consistent with a transformation of the CuIpc(-SO3)3(-SO2N<)\4-

pendants into the ligand-radical CuIIpc•

(-SO3)3(-SO2N<)\4- pendants (Scheme 2). The transformation of the Cu(I) product to the ligand radical exhibited a first order kinetics with a rate constant, k = 1.4 × 10 s-1.

N2O-saturated solutions containing 2.4 ×10-5 M CuIIpc(-SO3)3(-SO2N<)\3- pendants were used for a study of the reactions of poly(K3CuIItspc) with pulse radiolytically generated N3

•, SO4•- and (CH3)2COHC•H2 radi-

cals were investigated with. The ionic strength of the solutions was adjusted to 0.1 M with NaN3, K2SO4 or NaClO4. By comparison to changes in the UV-vis absorption spectrum

Table 1. Reduction potentials, E0, and rate constants, k, of the radicals reacting with CuIIpc(-SO3)3(-SO2N<)\3-

and CoIIpc(-SO3)2(-SO2N<)\2- pendants

Radical E0, Volt k, M-1.s-1

CuIIpc(-SO3)3(-SO2N<)\3-

Reaction product

e-sol -2.75a CuIIpc•(-SO3)3(-SO2N<)\4- 3.5 × 1010

CO2• - -1.20a CuIIpc•(-SO3)3(-SO2N<)\4- 8.6 × 108

+

CuIpc(-SO3)3(-SO2N<)\4-

C•H2OH -0.92a CuIIpc•(-SO3)3(-SO2N<)\4- - 1.0 × 108

N3• 1.7b, CuIIpc•(-SO3)3(-SO2N<)\2- 2.7 × 109

SO4•- 2.43c CuIIpc•(-SO3)3(-SO2N<)\2- 4.0 × 109

CoIIpc(-SO3)2(-SO2N<)\2-

Reaction product

(CH3)2COHC•H2 0.6a ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2- 1.0 × 1010

C•H2OH 1.29a (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- 8.0 × 109

N3• 1.7b XCoIIIpc(-SO3)2(-SO2N<)\2- 3.7 × 109

a Reduction potentials for A/Radical couples vs. NHE. Values from reference 17. b Reduction potential for Radical/A couples vs. NHE. Value from reference 18. c Reduction potential for Radical/A couples vs. NHE. Value from reference 19.

Scheme 2. Metal-centered and ligand-centered reduction of CuIIpc(-SO3)3

(-SO2N<)\3- pendants by CO2•- radicals and the interconversion of the Cu(I)

product into the more stable phthalocyanine radical. The polyamide back-bone and bonds of the pendants are represented by doted lines


76 G.T. RUIZ ET AL.

of solutions containing oligomers of CuII(tspc)4-,[7, 16, 20] changes in the spectrum of polymer’s solution revealed the formation of CuIIpc•(-SO3)3(-SO2N<)\2- pen-dants as the product of the reactions with N3

•, SO4•- and

(CH3)2COHC•H2 radicals (Equation 1).

R• + CuIIpc(SO3)3(SO2N<)\3-

→ R- + CuIIpc•(SO3)3(SO2N<)\2- (1)R• = N3

•, SO4•-, (CH3)2COHCH2

•

No reactions between these radicals and groups in the PAAMr polymers were apparent in blanks made by replacing poly(K3CuIItspc) with an equivalent concentra-tion of PAAMr. It was concluded that the reactions of the radicals with PAAMr were slower than the radicals’ dis-proportionation and/or dimerization reactions and those of the radicals with CuIIpc(-SO3)3(-SO2N<)\3- pendants. The kinetics of the reactions of the N3

• and SO4•- radicals

with CuIIpc(-SO3)3(-SO2N<)\3- pendants were investigated at the maximum, λmax = 500 nm, of the CuIIpc•(-SO3)3

(-SO2N<)\2- spectrum under a pseudo-first order regime. A nearly diffusion-controlled rate was exhibited by both reactions (Table 1). The calculated rate constants for each reaction are k = 2.7 × 109 M-1.s-1 for N3

• and k = 4.0 × 109 M-1.s-1 for SO4

•-. Further spectral changes resulted from the decay of the radical CuIIpc•(-SO3)3

(-SO2N<)\2- pendants at times longer than a millisecond. A decrease of the solution’s absorbance was observed between 500 and 550 nm where the absorption band of the radical CuIIpc•(-SO3)3(-SO2N<)\2- is placed. In addi-tion to this change in the spectrum, a simultaneous recov-ery of the absorbance took place at ~700 nm, i.e. where the Q band of the CuIIpc(-SO3)3(-SO2N<)\3- pendant is placed. These changes in the absorption spectrum of the solution are consistent with the transformation of a frac-tion of the ~1.5 × 10-5 M CuIIpc•(-SO3)3(-SO2N<)\2- gen-erated in the experiment to ~0.75 × 10-5 M CuIIpc(-SO3)3

(-SO2N<)\3- pendants. Such a stoichiometry is expected if the CuIIpc•(-SO3)3(-SO2N<)\2- pendants disproportionate to reform CuIIpc(-SO3)3(-SO2N<)\3- (Equation 2).

2 CuIIpc•(-SO3)3(-SO2N<)\2-

→ CuIIpc(-SO3)3(-SO2N<)\3- + CuIIL(-SO2N<) (2)

The CuIIL(-SO2N<) product represents a complex of the open-cycle macrocyclic ligand L(-SO2N<) with weak absorption bands, i.e. extinction coefficients ε < 102 M-1.cm-1, at any of the probing wavelengths, λob > 400 nm [7]. The formation of CuIIL(-SO2N<) in the radiolyzed solutions was confirmed when Cu2+

aq tested positive after it was liberated from its L(-SO2N<) complex with 0.1 M HClO4. In accordance with the disproportionation of the CuIIpc•(-SO3)3(-SO2N<)\2- pendants (Equation 2), the reaction rate, investigated at λob = 500 and 700 nm, exhib-ited a second-order dependence on the concentration of CuIIpc•(-SO3)3(-SO2N<)\2-. The rate constant, 2k = 8.1 ×107 M-1.s-1, of Equation 2 was calculated from the slope of plots of the reciprocal of the concentration vs. time and the initial concentration of CuIIpc•(-SO3)3(-SO2N<)\2-.Initial concentrations of CuIIpc•(-SO3)3(-SO2N<)\2- were estimated on the basis of the dosimetry. A comparative study was made on the disproportionation of the oligo-meric radicals (CuII(tspc)\4-)n-1CuII(tspc•)\3- (Equation 3), generated when (CuII(tspc)\4-)n reacted with N3

• radicals under the same medium conditions and CuII concentra-tions given above for the polymer.

2 CuIIpc•(-SO3)3(-SO2N<)\2-

→ CuIIpc(-SO3)3(-SO2N<)\3- + CuIIL(-SO2N<) (2)

2 (CuII(tspc)\4-)n-1CuII(tspc•)\3-

→ (CuII(tspc)\4-)n + (CuII(tspc)\4-)n-1CuIIL’ (3)

The second order rate constant , 2k = 1.3 × 106 M-1.s-1,was calculated for the (CuII(tspc)\4-)n-1CuII(tspc•)\3- dispro-portionation. It is almost two orders of magnitude smaller than the rate constant, 2k = 8.1 × 107 M-1.s-1, of the CuIIpc•(-SO3)3(-SO2N<)\2- pendants disproportionation (Equation 2), suggesting that Equations 2 and 3 proceed through different mechanisms.

Reactions of radicals with poly(K2CoIItrspc) forming CoIII-hydroxyalkyl and CoI products

The reactions of poly(K2CoIItrspc) with radiolyti-cally generated C-centered radicals and N3

• radicals have similar spectroscopic features (Fig. 7). Minor differences

Fig. 6. Reduction of poly(CuIItspc)\3- by pulse radiolytically generated CO2

•- radicals. The time-resolved formation of the metal-reduced, CuIpc(-SO3)3(-SO2N<)\4- and the ligand-radical, CuIIpc•(-SO3)3(-SO2N<)\4- is shown in (a). In (b), further changes in the spectrum correspond to the transformation of CuIpc(-SO3)3

(-SO2N<)\4- into ligand-radical CuIIpc•(-SO3)3(-SO2N<)\4- pen-dants. Delays from the radiolytic pulse are indicated in the fi gure



between the spectrum of the Co(III) phthalocyanine pendants produced with N3

• radicals or C-centered radi-cals were attributed to the coordination of the reduced oxidants to the metal center. Indeed, specific ligand-to-Co(III) charge transfer transitions superimposed with the intense phthalocyanine-centered electronic transitions account for the differences in the spectrum. These spec-tra also show similarities with those of the CoIIItrspc2- and CoIIItspc3- [20, 21].

The reaction of N3• radicals with poly(K2CoIItrspc)

(Equation 4), was investigated in N2O-saturated solutions containing 2 × 10-5 M CoIIpc(-SO3)2(-SO2N<)\2- pendants and 0.1 M NaN3.

N3• + CoIIpc(SO3)2(SO2N<)\2-

→ XCoIIIpc(SO3)2(SO2N<)\2- (4) X = N3, OH

The concentration of N3• radicals generated in these

experiments produced ~10% oxidized pendants of the total pendant concentration. Therefore, the kinetics of the XCoIIIpc(-SO3)2(-SO2N<)\2- pendants formation was investigated under a pseudo-first order regime. A sec-ond-order rate constant, k = 3.7 × 109 M-1.s-1, was cal-culated using oscillographic traces recorded at various wavelengths, λob = 295, 425 and 665 nm. The rate con-stants of the reactions of the radicals with CoIIpc(-SO3)2

(-SO2N<)\2- have been collected in Table 1.The reactions of poly(K2CoIItrspc) with C-centered

radiolytically generated radicals exhibited some differ-ences with those of [CoII(tspc)]2

\8- [20, 21]. In a 1 to 100 ms time domain, the reaction of 2-methyl-2-hydroxy-1-propyl radicals, (CH3)2COHC•H2, with poly(K2CoIItrspc)

solutions containing 2 × 10-5 M CoIIpc(-SO3)2(-SO2N<)\2-

pendants in 0.1 M (CH3)3COH and 0.1 M NaClO4 forms unstable species (Fig. 8). The difference spectrum exhib-ited absorption bands with λmax = 290 and 460 nm. No bleach of the Q band, λmax = 650 nm, indicative of a phthalocyanine-centered radical was observed. Because of these spectroscopic features and the resemblance to the spectrum of a Co(III) product generated when poly(K2CoIItrspc) reacted with N3

• radicals (Equation 4), the unstable species was assigned as a pendant with CoIII-C bonds, namely the Co(III)-hydroxyalkyl moiety ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2- instead of CoIIpc•(-SO3)2(-SO2N<)\3- (Equation 5).

(CH3)2COHC•H2 + CoIIpc(SO3)2(SO2N<)\2-

→ ((CH3)2COHCH2)CoIIIpc(SO3)2(SO2N<)\2- (5)

A second order rate constant, k = 1.0 × 1010 M-1.s-1, for the reaction of poly(K2CoIItrspc) with (CH3)2COHC•H2

radicals was calculated under a pseudo-first order regime from oscillographic traces recorded at λob = 290 and 460 nm. CoIpc(-SO3)2(-SO2N<)\3- pendants (Equation 6), were produced when the Co(III) species formed in Equa-tion 5, decayed in a time scale t > 1 ms.

((CH3)2COHCH2)CoIIIpc(SO3)2(SO2N<)\2-

—→ (CH3)2COHCH2OH + CoIpc(SO3)2(SO2N<)\3- (6)

The CoIpc(-SO3)2(-SO2N<)\3- pendants formed in this reaction were identified by means of the appearing spectroscopic features between 350 and 450 nm (Fig. 8), which are characteristic of Co(I) phthalocyanine [20, 21]. Analysis of the traces collected in these range of wavelengths showed that the rate law of the reaction was first order on the Co(III) concentration with a rate

Fig. 7. Changes in the absorption spectrum when poly(K2CoIItrspc) reacts with N3

• and (CH3)2COHCH2• radicals

forming respectively the XCoIIIpc(-SO3)2(-SO2N<)\2- (X = N3 or OH) and ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2-. pendants. A key to the reactions is in the fi gure and experimental condi-tions are communicated elsewhere in the text

Fig. 8. Changes in the absorption spectrum due to the trans-formation of ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2- pen-dants into CoIpc(-SO3)2(-SO2N<)\3- pendants. Experimental conditions are indicated elsewhere in the text

OH-


78 G.T. RUIZ ET AL.

constant, k = 5.3 × 10 s-1. An additional experiment was carried out in order to verify that Co(III)-alkyl pendants were converted to CoIpc(-SO3)2(-SO2N<)\3- (Equation 6). In the experiment, both e-

sol and (CH3)2COHC•H2 radi-cals were simultaneously generated by the radiolysis of poly(K3CoIItrspc) in a N2-saturated solution containing 2 × 10-5 M CoIIpc(-SO3)2(-SO2N<)\2- pendants, 0.1 M (CH3)3COH and 0.1 M NaClO4. Changes in the absorp-tion spectrum of the solution showed the formation of the ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2- and CoIpc(-SO3)2(-SO2N<)\3- pendants in a time scale equal to or shorter than 40 μs. The two products are produced by the respective reactions of the (CH3)2COHC•H2 and e-

sol

radicals with the Co(II) pendants. The decay of the ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2- pendants was investigated in the 1 to 102 ms time domain and the spec-tra recorded with 40 and 80 ms delays revealed only the spectroscopic features of the CoIpc(-SO3)2(-SO2N<)\3-

spectrum. Based on the absorbance changes caused by the disappearance of ((CH3)2COHCH2)CoIIIpc(-SO3)2

(-SO2N<)\2-, it was estimated that the conversion of the latter to CoIpc(-SO3)2(-SO2N<)\3- pendants was equal to or larger than 95%.

Spectroscopic changes observed when C•H2OH radi-cals reacted with poly(K2CoIItrspc) (Fig. 9), were simi-lar to those observed when (CH3)2COHC•H2 radicals reacted with the polymer (Equations 5 and 6). The reac-tion was investigated with N2O-saturated solutions of poly(K2CoIItrspc) containing 2 × 10-5 or 1.33 × 10-5 M CoIIpc(-SO3)2(-SO2N<)\2- pendants in 0.1 M CH3OH and 0.1 M NaClO4. In the reaction with C•H2OH, a transient species also mediates the formation of the CoIpc(-SO3)2

(-SO2N<)\3- pendants. Similarities between the spectrum of the precursor of the Co(I) pendants and the Co(III)- hydroxyalkyl pendants in Equation 4 suggest that the pre-cursor are (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- pendants (Equation 7).

C•H2OH + CoIIpc(SO3)2(SO2(SO2N<))\2-

→ (OHCH2)CoIIIpc(SO3)2(SO2N<)\2- (7)

The assignment of transient spectra recorded in a 1 to 100 μs time domain to the formation of (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- pendants is consistent with their slow transformation into Co(I) pendants and CH2O.The transformation was observed in a time scale t > 2 ms by means of changes in the spectrum, from that of (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- recorded early in the reaction to the CoIpc(-SO3)2(-SO2N<)\3- pendants. In agreement with this proposition, the formation of CoIpc(-SO3)2(-SO2N<)\3- is kinetically of a first order in (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- concentration and with a rate constant, k = 3.7 × 10 s-1, slighthly smaller than the rate constant of Equation 4.

DISCUSSION

Based on the known tendency of the phthalocyanines to form π-π stacks, the different morphologies of the poly(K3CuIItspc) and poly(K2CoIItrspc) in solid state must be attributed to the different interactions of the CuIIpc(-SO3)3(-SO2N<)\3- and CoIIpc(-SO3)2(-SO2N<)\2-

pendants in the respective polymers (Fig. 2). In terms of the sizes of the structures visualized in Fig. 2, they must represent aggregates of strands in bundles of different shapes. The rosette shapes (Fig. 2a) can be visualized as an aggregation of strands which are randomly oriented and/or coiled inside the bundle. A preminent side-on association of strands may lead to a more elongated bundle (Fig. 2b).

Since a spherical aggregation of strands maximizes the volume of material enclosed by the sphere’s surface, it can be expected that a spherical aggregation will be favored when the ionic charge provided by the phthalocyanine pendants cannot compensate for the hydrophobic charac-ter of the polymer’s strand. Differences in the stability of the π-π stacks and the hydrophilicity of the CuIIpc(-SO3K)3

(-SO2N<)\3- and CoIIpc(-SO3)2(-SO2N<)\2- must control the conversion of the rosette shaped aggregates (Fig. 2a) to the elongated bundles (Fig. 2b).

Regardless of the morphologies in solid or solution phase, the UV-vis and ESR spectroscopies show that the CuIIpc(-SO3)3(-SO2N<)\3- and CoIIpc(-SO3)2(-SO2N<)\-

pendants in the respective bundles of poly(K3CuIItspc)and poly(K2CoIItrspc) are isolated as well as form π-πstacks. A distribution of monomeric and oligomeric pen-dants exposed to the protic medium of the solution and to a less protic inner space of the bundle accounts for the chemical reactions discussed below and the cyclic volta-mmograms of the polymers. Indeed, similarities between

Fig. 9. Changes in the absorption spectrum observed when C•H2OH radicals reacted with poly(K2CoIItrspc). The change at 50 μs, was assigned to the formation of (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- pendants and those observed at 70 ms were assigned to the formation of Co(I) pendants, CoIpc(-SO3)2

(-SO2N<)\3-



the oxidation waves in the cyclic voltamograms of the polymers and the oligomers [MII(tspc)]n

\(4•n)-, M = Cu, Co, suggest that the redox processes of the polymers involve pendants in the surface of the bundle rather than in the interior.

The presence of pendants in different media helps to rationalize the formation of different products when CuIIpc(-SO3)3(-SO2N<)\3- pendants react with radi-olytically generated CO2

•- radicals. The simultaneous formation of CuIpc(-SO3)3(-SO2N<)\4- and CuIIpc•(-SO3)3

(-SO2N<)\4- pendants was observed only when CO2•- reacted

with poly(K3CuIItspc) (Scheme 2). Cu(I) phthal ocyanines were previously generated when [CuII(tetrakis(N-ocata-decylsulfamoyl) phthalocyanine)]2, (CuII(tsapc))2, was photolyzed in non-aqueous media (Equation 8) [21].

(CuII(tsapc))2 + hν(solvent CH2Cl2)

CuItsapc-

+ CuIIItsapc+ (8)

In contrast to the reaction in non-aqueous medium, the electrochemical reduction of CuII(tspc)\4- in DMSO and the photoredox dissociation of (CuII(tspc)\4-)2 in aqueous solution produced Cu(II)-ligand radical species [7, 16, 20]. It is possible, therefore, that a large contribut-ing factor to the formation of the CuIpc(-SO3)3(-SO2N<)\4-

is the presence of CuIIpc(-SO3)3(-SO2N<)\3- pendants with peculiar solvation inside the bundle. Although the solvation of the pendants must be one factor directing the reduction of the CuIIpc(-SO3)3(-SO2N<)\3- pendants, it cannot be the determining. If it were the sole factor, e-

sol, and C•H2OH will also reduce the Cu(II) center in an appreciable proportion. Kinetic effects, illustrated by the dependence of the reaction rate constant on the redox potential of the radical in Table 1, and the coordination of CO2

•- to the Cu(II) center, i.e. in a process similar to those observed with poly(K2CoIItrspc) (Equations 5 and 7), must be regarded as additional factors that deter-mine the reaction pathway.

The reactions of the C-centered radicals, C•H2OH and (CH3)2COHC•H2, with poly(K2CoIItrspc) have in com-mon the formation of CoIII-alkyl pendants, (HOCH2)CoIIIpc(-SO3)2(-SO2N<)\2- and ((CH3)2COHCH2)CoIIIpc(-SO3)2(-SO2N<)\2-, as intermediates in the formation of the CoIpc(-SO3)2(-SO2N<)\3- product. In contrast to these reactions, those of [CoII(tspc)]2

\8- with radicals in aqueous solution mainly reduced the phthalocyanine to ligand-radical species in a first step. The transfer of the elec-tron from the ligand-radical to the CoII center occurs in a slower second step [7, 21]. Conditions, such as the media where the pendants of poly(K2CoIItrspc) are placed in the bundle, must favor the coordination of the C-centered radicals to the Co(II) over the reduction of the phthalo-cyanine ligand. Indeed, the rate constant for the forma-tion of the Co(III)-hydroxyalkyl intermediates (Table 1) is larger than the rate constants for the reduction of the phthalocyanine ligand [7] and has values expected for a diffusion-controlled process. The formation of the CoI

products through paths involving the coordination of the radical to the Co(II) center or the reduction of the phthalocyanine ligand is as expected on thermochemi-cal grounds. Both paths yield the stable Co(I) product as electrochemical experiments have previously demon-strated [4].

FINAL REMARKS

Marked differences have been observed between the redox reactions of the phthalocyanine pendants in poly(K3CuIItspc) and poly(K2CoIItrspc) and those of the [MII(tspc)]2

\8-, M = Cu, Co. These differences are caused by various factors which include stacking of the phthalocyanine pendants, the medium inside the poly-mer bundle and other kinetic and thermochemical fac-tors. Although the final products of the redox reactions are those expected from a thermochemical standpoint, these combined effects lead to new reaction pathways. In spite of CoIIpc(-SO3)2(-SO2N<)\2- and CuIIpc(-SO3)3

(-SO2N<)\3- pendants being both covalently linked to the polymer backbone, the latter is prone to redox degrada-tions which must be taken into consideration in applica-tions of the polymer to the catalysis of redox processes. The medium inside the bundle and the coordination of reactants such as radicals to the metal center is a function of the organic backbone’s nature. Therefore, groups other than the acetamide ones must be used to induce major changes in the thermal and photochemical reactivity of the phthalocyanine pendants. For example, the polyeth-yleneimine can be derivatized with reactants producing functional groups much more hydrophilic or hydropho-bic than the acetamide groups or functional groups with a particular affinity for the metal center of the pendants. The experimental observations show that phthalocyanine polymers tailored to specific applications, such as the catalyzed reduction of CO2 in aqueous media, [5, 6] can be prepared using a judicious selection of the functional groups in the polymer backbone and the transition metal ion in the phthalocyanine pendant.

Acknowledgements

The work described herein was supported by the Office of Basic Energy Sciences of the US Department of Energy. (Contribution No. NDRL- 4746 from the Notre Dame Radiation Laboratory.) GTR acknowledges sup-port from CONICET and ACS. The authors are thankful to Dr. E. Carrasco Flores for his assistance with the FTIR measurements.


A discussion of the polymers preparation and a flow chart (Scheme S1) of the preparative steps are given in the supplementary material. This material is available free of charge via the Internet at http://www.worldscinet.com/jpp/jpp.shtml.


80 G.T. RUIZ ET AL.

REFERENCES

1. The following references are limited in number and only exemplify some of the recent phtha-locyanines applications; a) Haywood-Small SL, Vernon DI, Griffiths J, Schofield J and Brown SB. Biochem. Biophys. Research Comm. 2006; 339:569–576 (and Erratum ibid 2006; 341: 673–673). b) Tzekov R, Lin T, Zhang KM, Jackson B, Oye-jide A, Orilla W, Kulkarni AD, Kuppermann BD, Wheeler L and Burke J. Invest. Ophthalmol. Visual Sci. 2006; 47: 377–385. c) Gurek AG, Basova T,Luneau D, Lebrun C, Kol’tsov E, Hassan AKand Ahsen V. Inorg. Chem. 2006; 45: 1667–1676. d) Hasegawa H, Ueda R, Kubota T and Mashiko S.Thin Solid Films 2006; 499: 289–292. e) Yang LF,Guan M, Bian ZQ, Xie JQ, Chen TP and HuangCH. Thin Solid Films 2006; 500: 224–230. f) Gen-erosi A, Paci B, Albertini VR, Perfetti P, PaolettiAM, Pennesi G, Rossi G and Caminiti R. J. Appl. Phys. 2006; 99: 044901. g) Mayer I, Nunes GS,Toma HE and Araki K. Eur. J. Inorg. Chem. 2006; 850–856. h) Kitamura Y, Mifune M, Hino D, Yoko-tani S, Saito M, Tsukamoto I, Iwado A and SaitoY. Talanta 2006; 69: 43–47. i) Kostka M, ZimcikP, Miletin M, Klemera P, Kopecky K and MusilZ. J. Photochem. Photobiol. A 2006; 178: 16–25. j) Rajendiran N and Santhanalakshmi J. J. Mol. Catal. A 2006; 245: 185–191. k) Ballesteros B,de la Torre G, Torres T, Hug GL, Rahman GMAand Guldi DM. Tetrahedron 2006; 62: 2097–2101. l) Wrobel D and Dudkowiak A. Mol. Cryst. Liq. Cryst. 2006; 448: 15–38. m) Christendat D, DavidMA, Morin S, Lever ABP, Kadish KM and ShaoJG. J. Porphyrins Phthalocyanines 2005; 9: 626–636. n) Chen J and Lever ABP. J. Porphyrins Phtha-locyanines 2007; 11: 151–159. o) Magaraggia M,Visona A, Furlan A, Pagnan A, Miotto G, TognonG and Jori G. J. Photochem. Photobiol. B 2006; 82:53–58. p) Karandikar P, Chandwadkar AJ, AgasheM, Ramgir NS and Sivasanker S. App. Catal. A2006; 297: 220–230. q) Zhang Y, Sun X, Niu Y, XuR, Wang G and Jiang Z. Polymer 2006; 47: 1569–1574. r) Lever ABP. J. Porphyrins Phthalocyanines2004; 8: 1327–1342. s) Phthalocyanines. Properties and Applications, Vols. 1–4, Leznoff CC and LeverABP. (Eds.) VCH: New York, 1989, 1993, 1996. t) Manners I. Science 2001; 294: 1664–1666.

2. De Wit DG. Coord. Chem. Rev. 1996; 147:209–246.

3. Wang X-Y, Prabhu RN, Schmehl RH and Weck M.Macromol. 2006; 39: 3140–3146.

4. Thomas S, Ruiz G and Ferraudi G. Macromol.2006; 39: 6615–6621.

5. Costamagna J, Ferraudi G, Canales J and Vargas J.Coord. Chem. Rev. 1996; 148: 221–248.

6. Costamagna J, Ferraudi G, Matsuhiro B, Campos-Vallette M, Canales J, Villagrán M, Vargas Jand Aguirre MJ. Coord. Chem. Rev. 2000; 196:125–164.

7. Ferraudi G. In Phthalocyanines. Properties and Applications, Vol. 1, Leznoff CC and Lever ABP. (Eds.) VCH: New York, 1989; Chapter 4, pp 302–307, 326 and references therein.

8. Dutta SK and Ferraudi G. J. Phys. Chem. A 2001; 105: 4241–4247.

9. Hug GL, Wang Y, Schöneich C, Jiang P-Y and Fessenden RW. Radiat. Phys. Chem. 1999; 54:559–566.

10. Buxton GV, Greenstock CL, Hellman WP and RossAB. J. Phys. Chem. Ref. Data 1988; 17: 513–886.

11. Zickendraht C and Koller EJ. New Phthalocyanine Dyestuffs, U.S. Patent 2,897,2007.

12. Elementary Practical Organic Chemistry, VogelAI. (Ed.) Wiley: New York, 1958; pp 307, 400, 404, 420, 421.

13. Fieser LF and Fieser M. Advanced Organic Chem-istry, Reinhold Publishing Corporation: New York, 1961; pp 382–384, 698.

14. Goodman BA and Raynor JB. Adv. Inorg. Chem. Radiochem. 1970; 13: 135–362 and references therein.

15. Ukrainczyk L, Chibwe M, Pinnavaia TJ and BoydSA. J. Phys. Chem. 1994; 98: 2668–2676.

16. Ferraudi G and Srisankar EV. Inorg. Chem. 1978; 17: 3164–3168.

17. Wardman P. J. Phys. Chem. Ref. Data 1989; 18:1637–1755.

18. Ruiz GT, Juliarena MP, Wolcan E and Ferraudi G.Inorg. Chim. Acta 2007; 360: 3681–3687.

19. Stanbury DM. Adv. Inorg. Chem. 1989; 33:69–138.

20. Ferraudi G. Inorg. Chem. 1979; 18: 1005–1013. 21. Ferraudi G and Patterson LK. J. Chem. Soc., Dalton

Trans. 1980; 3: 476–480. 22. Prasad DR and Ferraudi G. Inorg. Chem. 1982; 21:

2967–2971. 23. Rollmann LD and Iwamoto RT. J. Am. Chem. Soc.

1968; 90: 1455–1463.


DOI: 10.1142/S1088424610001696



INTRODUCTION

The anchoring of molecules onto the surface of a solid substrate to form self-assembled monolayers (SAMs) is now an established and well documented technology [1]. Molecules designed for particular applications contain a “functional core”, i.e. a moiety that provides a desired effect or response to external stimuli, and a means for attaching it to a solid surface [2]. The latter frequently, but not invariably, exploits a tether chain that terminates with a functional group that reacts or interacts with the surface. A common combination for anchoring molecules to sur-faces uses a thiol- or disulfi de-terminated tether to form a thiolate linkage to gold; a second approach employs a silyl function to react with a silica/silicon surface [1, 2]. In previous work, one of our laboratories exploited both of these approaches to develop examples of phtha-locyanine SAMs [3, 4]. SAMs deposited onto the surface of gold, supported on a glass slide, were investigated for optically based gas-sensing devices [5]. More recently

these studies were extended to the surface coating of gold nanoparticles with a zinc phthalocyanine derivative in the presence of tetraoctylammonium bromide [6]. This con-struct has been exploited as a drug delivery system within a photodynamic therapy (PDT) protocol. Irradiation of these nanoparticle conjugates within HeLa cells induced substantial cell mortality through the photodynamic pro-duction of singlet oxygen [7].

We now report some parallel work using single- component porphyrin SAMs on the surface of gold and thus contribute to the burgeoning body of research into the characterization and applications of this type of SAM [8–11]. An interesting development has been the appli-cation of porphyrin SAMs for heterogeneous oxidation catalysis and this has exploited, for example, the redox behavior of Mn [9] and Co [10] contained within the macrocycle core. To our knowledge there has been no focus on exploiting porphyrin SAM surfaces for photody-namic oxidations. The present paper addresses this area, describing the synthesis of purpose-designed porphyrin derivatives and their deposition onto a gold surface sup-ported on a glass slide. The role of these constructs in providing a ‘‘photodynamic surface’’ is demonstrated by the photooxidation of the tryptophyl residue in human

Porphyrin self-assembled monolayers and photodynamic oxidation of tryptophan

Isabelle Chambriera, David A. Russella, Derek E. Brundishb, William G. Loveb,Giulio Jori*c , M. Magaraggiac and Michael J. Cook*a

a Wolfson Materials and Catalysis Center, School of Chemistry, University of East Anglia, Norwich NR4 7TJ, UKb Destiny Pharma Ltd., Sussex Innovation Center, Science Park Square, Falmer, Brighton BN1 9SB, UKc Department of Biology, University of Padova, via Trieste 75, 35121 Padova, Italy


ABSTRACT: The zinc and magnesium metalated derivatives of 5,5′-[12,12′-di(thiododecyloxy)-4,4′-phenyl)]-10,10′,15,15′,20,20′-hexakis(3,3′,4,4′,5,5′-hexakisdecyloxyphenyl)diporphyrin, 1b and 1c,have been synthesised and deposited to form self-assembled monolayer (SAM) fi lms on the surface of gold-coated glass substrates. The SAM fi lms have been characterized by RAIR spectroscopy and fl uorescence spectroscopy. The potential for the porphyrin fi lms to catalyze the oxidation of tryptophan within human serum albumin upon irradiation with white light has been demonstrated and attributed to the porphyrins acting as photosensitizers of oxygen to form oxidizing species.

KEYWORDS: porphyrin self-assembled monolayers, SAM fi lms on gold, photodynamic surface.


*Correspondence to: Michael J. Cook, email: [email protected], fax: +44 (0)1603-592015


82 I. CHAMBRIER ET AL.

serum albumin (HSA) when the fi lms are illuminated with white light.


Design of compounds

The non-symmetrically tetra-aryl substituted porphyrin derivatives investigated for deposition as SAM fi lms, 1band 1c (Scheme 1), contain two porphyrin moieties linked through one of their aryl rings via a tether terminated with a disulfi de unit. Three long-chain alkoxy groups on each of the remaining three aryl rings per porphyrin unit were incorporated to minimize self- association and the detri-mental quenching effects this has on the electronically excited states of the porphyrin. Zinc and magnesium were chosen as central metals because of their limited adverse toxic properties so long as these ions are incorporated into the tetrapyrrolic macrocycle [12]. The route to pre-pare the compounds 1b and 1c proceeds via the synthesis of a non-symmetrically substitued metal-free porphyrin derivative with a tether terminated with a hydroxyl group 2b. This functionality was deemed suitable for subsequent introduction of the disulfi de moiety via the mesylate 2c.Metalation of the ring with zinc and magnesium provided the fi nal step of the synthetic work.

Synthesis

Aldehydes. The required aromatic aldehyde derivatives for the synthesis of the desired porphyrins were prepared from 4-hydroxybenzaldehyde and 3,4,5-trihydroxybenzal-dehyde. Yields were improved compared to those reported in the literature [13] by using a combination of both a phase-transfer catalyst, tetra-n-butylammonium iodide, to facilitate the substitution and a high boiling solvent.

Condensation. The mixed cyclotetramerization reac-tion (Step iv in Scheme 1) was fi rst attempted using BF3-etherate as the acid and p-chloranil as the oxidant. No porphyrins were obtained by this method. The reaction was therefore carried out using propionic acid under aer-obic conditions. The effi ciency of the reaction was further improved by using an oxidizing co-solvent, nitrobenzene [14]. Under these conditions there was no evidence of chlorin formation. However, on using 4-(12-hydroxy-dodecyloxy)benzaldehyde it was found that the product was partially esterifi ed to form the propionate leading to mixtures and non-reproducible preparations. The acetate derivative of 4-(12-hydroxydodecyloxy)benzaldehyde (i.e. X = OCOCH3 in Step iv) was therefore preferred as a precursor. Higher yields for the cyclotetrameriza-tion were obtained and purifi cation of the product 2a was straightforward. This compensated for the inclusion of the extra step required to deprotect 2a to afford 2b.

CHO

OH

CHO

O(CH2)12X

X= OH

ROOR

OR

CHO

HOOH

OH

CHO

N N

NNM

RO

RO

RO

RO

OROR

OR

OR

OR

O(CH2)12Y

N N

NNM

RO

RO

RO

RO

OROR

OR

OR

OR

O(CH2)12S2

M= H2

M= Zn

i

ii

iii

iv

v

vi

vii

viii

M= Mg

ix

1a

1c

1b

X= OCOCH3

X= OCOCH3

R= n-C10H21

M= H2, Y= OCOCH3

M= H2, Y= OH

M= H2, Y= OSO2CH3

2a

2c

2b

Scheme 1. Synthesis of compounds 1b and 1c. In all cases R = C10H21. Reagents: i. Br(CH2)12OH, MEK, K2CO3, KI, tetra-n-butylammonium iodide; ii. Ac2O; iii. n-bromodecane, MEK, KI, K2CO3, tetra-n-butylammonium iodide; iv. pyrrole, propionic acid, nitrobenzene; v. NaOH; vi. Methanesulfonylchloride, Et3N; vii. 1) Thiourea, 2) aq. NaOH; viii. Zn(OAc)2; ix. Mg(ClO4)2, dry pyridine


PORPHYRIN SELF-ASSEMBLED MONOLAYERS AND PHOTODYNAMIC OXIDATION OF TRYPTOPHAN 83

Metal-free porphyrin disulfi de 1a. The free hydroxyl group of 2b was converted into a mesylate, to afford 2c,which was reacted with thiourea under aerobic condi-tions. The porphyrin disulfi de derivative 1a was obtained after hydrolysis using aqueous sodium hydroxide. The use of an aqueous solution of NaOH proved to be a satis-factory alternative to an NaOH solution in ethanol which led to a product whose 1H NMR spectrum contained an unexplained extra set of signals.

Metalated porphyrin disulfi des 1b and 1c. Finally, the metalated derivatives 1b and 1c were obtained by refl uxing the metal-free compound 1a in refl uxing THF and dry pyridine containing zinc acetate and magnesium perchlorate respectively. Expected changes in the UV-vis spectra were observed upon the introduction of the metal into the macrocycle. The four weak absorption bands at 516.0, 552.5, 592.5 and 650.5 nm present in the metal-free porphyrin 1a are replaced by two weak bands at 557.0 and 598.0 nm, and 569.0 and 615.0 nm for the Zn and Mg derivatives, 1b and 1c, respectively. Similarly, the Soret band was shifted from 423.0 nm in the metal-free compound to 427.5 nm for the zinc derivative and 432.0 nm for the magnesium derivative.

Both products were characterized by 1H NMR spec-troscopy which showed the signal corresponding to the protons α to the sulfur atom as a triplet at 2.71 and 2.6 ppm for 1b and 1c, respectively. MALDI-TOF mass spectra showed isotopic clusters at 4601 [M+] for 1b and4518 [M+] for 1c. Satisfactory C, H, N elemental analysis data confi rmed the structures and attested to the purity of the compounds.

Formation and spectral characterization of self- assembled monolayer fi lms of 1b and 1c

Conditions for the preparation of the gold-coated glass substrates are described in the Experimental section. The deposition of 1b and 1c to form SAM fi lms on the

substrates was undertaken by immersing substrates into a solution of each compound in cyclohexane at concentra-tions of ca. 2.5-5 × 10-7 M for 20 hours. The constructs were then rinsed with cyclohexane and allowed to dry in air. Reference (blank) substrates, i.e. without a SAM coating, were similarly immersed in neat cyclohexane for 20 hours and dried as above. Refl ection-absorption infra-red spectroscopy (RAIRS) was used to detect formation of SAM fi lms on the gold surface.

Figure 1 shows the RAIR spectra (2600–3200 cm-1

region) of an example of a SAM fi lm of 1b. Comparable spectra were obtained for SAMs of 1c. The absorption bands shown in Fig. 1 are assigned to the aliphatic C-H stretches associated with the peripheral alkyloxy chains viz. CH3 (νAS) at 2959 cm-1, CH2 (νAS) at 2926 cm-1, and CH2 (νS) at 2857 cm-1. Assignments of other bands in the spectra were not attempted. Intensities of the absorption bands showed good reproducibility over different regions of the fi lm and from one fi lm to another prepared in the same batch. The nature of the central metal ion does not seem to have any effect on SAM formation.

The thickness of the gold coating precluded acquisi-tion of UV-vis absorption spectra. However, fl uorescence emission spectra of the fi lms and fl uorescence excitation spectra were recorded. Examples of these are shown in Figs 2 and 3, respectively. The emission spectrum (Fig. 2) shows the fl uorescence spectrum obtained upon irradia-tion at 415 nm. The emission is weak and gives rise to a broad band spectrum, not unexpected and comparable to that observed for phthalocyanine SAMs [5]. There is a narrower peak centered at ca. 620 nm which we tenta-tively assign to a Raman band. Evidence in support of this is that the band position varies with the irradiating wavelength. Its intensity can be ascribed to the effect of the gold surface, as exploited in surface enhanced reso-nance Raman scattering (SERRS) measurements. Esti-mation of the fl uorescence quantum yields was attempted by comparing band intensities assigned to fl uorescence

Fig. 1. RAIR spectrum of a SAM fi lm of 1b showing the 3100–2700 cm-1 region



emission and band intensities in the excitation spectra with spectral data from solutions of haematoporphy-rin IX used as a fl uorescence standard (quantum yield 0.11 in a non-associated state in EtOH). This indirect approach, using data for seven fi lms of 1b and six fi lms of 1c, gave consistent data which indicated that for both types of SAM the quantum effi ciency was ca. 0.02 with

errors of the order of +/- 0.01. The low values may refl ect a degree of quenching of the fi rst excited singlet state of the mol-ecules within the fi lms through a variety of possible pathways, arising for example from their proximity to the gold surface or self-quenching through close association of the molecules despite the presence of the long alkoxy chains substituents [11, 15]. The occurrence of ground state inter-actions between porphyrin molecules in the SAMs is suggested by the broadness of the Soret band in the excitation spec-trum and the presence of at least two maxima for such bands (Fig. 3); moreover, the poorly resolved emission bands above 700 nm (Fig. 2) is suggestive of the for-mation of weakly fl uorescent side-by-side porphyrin dimers [12].

Photodynamic oxidation studies

The photosensitizing effi ciency of the two types of porphyrin SAMs was tested against a tryptophan model, because the indole side chain of this amino acid is susceptible to photooxidative attack by both type I (radical-involving) and type II (singlet oxygen-involving) pathways [12, 16–18]. Thus, a kinetic study of its pho-tosensitized modifi cation provides reliable information on the photoeffi ciency of the porphyrin derivatives that is independent of the specifi c mechanism of the photo-process. In particular, we selected a widely diffused protein, albumin, as a substrate since the three-dimensional structure of the polypeptide chain provides a “bio-logical” microenvironment for the trypto-phan (human serum albumin contains one tryptophyl residue per protein molecule). Most conveniently, the time-course of the photosensitized modifi cation of tryptophyl residues can be readily followed by spec-trophotofl uorimetric analysis, namely by following the decrease in the intensity of the 290 nm-excited fl uorescence emission in the 360 nm wavelength interval. Trypto-phan is known to be largely converted into photooxidation products that are kynure-

nine or hydroxyindole based [18]. These do not interfere with the determination of tryptophan fl uorescence, since their absorption and emission are shifted to the 400–500 nm spectral region [19]. For both SAMs, the photooxida-tion (white light irradiation) of the albumin-bound tryp-tophan sensitized by the porphyrin occurred according to fi rst-order kinetics, as represented by semilogarithmic

Fig. 2. Fluorescence emission spectrum from a SAM of 1b upon irradiation at 415 nm

Fig. 3. Excitation spectrum of a fi lm of 1b (emission at 620 nm)



plots (see Fig. 4). The slope of the plots allows the calcu-lation of the rate constant k for the photoprocess; hence, k can be taken as a parameter defi ning the photosensitizing effi ciency of the porphyrin SAMs. Under our experimen-tal conditions, the rate constant for the photoprocesses is of the order of 2 × 105 s-l; this value is about 20-fold smaller than those found for the photosensitized modi-fi cation of tryptophan by free porphyrins [12, 20]. The decrease in the effi ciency of a photosensitizer molecule, once it is immobilized on a solid matrix, is to be expected [17]. However, it is worth emphasizing that the photo-effi ciency of the immobilized porphyrins 1b and 1c is unusually large when compared with that observed for other matrix-bound photosensitizing agents [17].

EXPERIMENTAL

Equipment1H NMR spectra were measured in CDCl3 at 270 MHz

on a Jeol EX 270 or at 300 MHz on a Varian Gemini-300 using TMS as the internal reference. Mass spectra were obtained using the MALDI technique in Dithranol matrix or the FAB technique in Noba matrix. UV-vis spectra of solutions were measured in THF (unless otherwise speci-fi ed) using a Hitachi U-3000 spectrophotometer. RAIRS experiments were performed using a Bio-Rad FTS40 Fou-rier transform infrared spectrometer. Fluorescence emis-sion and fl uorescence excitation spectra were obtained using a Perkin Elmer MPF4 spectrophotofl uorimeter.

Materials

12-bromododecanol. 1,12-dodecanediol (50 g, 0.25 mol) in hydrogen bromide 48% (220 ml) was continu-ously extracted with petroleum ether (bp 80–100°C) (300 ml) for 18 h. The solvent was evaporated and the crude oil was fi ltered through silica gel (eluent: petroleum ether (bp 40–60°C)). A colorless fraction was obtained containing 1,12-dibromododecane (10.2 g, 13%). The

silica gel was then eluted with acetone. A pale yellow oil was obtained after evaporation of the solvent which crystallized upon cooling. The title product was recrys-tallized from petroleum ether (bp 40–60°C). Yield 43.1 g (65%). 1H NMR (60 MHz): δ, ppm 3.62 (t, 2H), 3.42 (t, 2H), 1.4 (brs, 21H).

4-[12-hydroxydodecyloxy]benzaldehyde. A mixture of 12-bromododecanol (15 g, 57 mmol), 4-hydroxy-benzaldehyde (6.9 g, 57 mmol), potassium carbonate (excess), catalytic amounts of potassium iodide and tetra-n- butylammonium iodide was refl uxed in methyl ethyl ketone (50 ml) for 16 h. After cooling, the solid was fi l-tered off and washed with acetone. The combined organ-ics were concentrated under reduced pressure. Diethyl ether was added to the oil and the fl ask placed in the fridge. The precipitate was fi ltered. Yield 12.5 g (72%). 1H NMR (60 MHz): δ, ppm 9.9 (s, 1H), 7.9 (d, 2H), 7.0 (d, 2H), 4.1 (t, 2H), 3.62 (t, 2H), 1.3 (brs, 21H).

4-[12-acetyloxydodecyloxy]benzaldehyde. 4-[12-hyd-roxydodecyloxy]benzaldehyde (1.1 g, 3.6 mmol) was refl uxed in acetic anhydride (15 ml) for 90 min. After cooling, the solution was poured into water with stirring. A precipitate formed which was fi ltered, washed with water and dried. Yield 1.1 g (88%). 1H NMR (60 MHz): δ, ppm 9.82 (s, 1H), 7.8 (d, 2H), 7.0 (d, 2H), 4.0 (t, 4H), 2.02 (s, 3H), 1.3 (brs, 20H).

3,4,5-trisdecyloxybenzaldehyde. A mixture of 3,4,5-trihydroxybenzaldehyde (2 g, 13 mmol), 10-bromod-ecane (15 ml, excess), potassium carbonate (excess), catalytic amounts of potassium iodide and tetra-n-bu-tylammonium iodide was refl uxed in methyl ethyl ketone (25 ml) for 16 h. After cooling, the solution was fi ltered and the solid washed with acetone. The solvent was evaporated leaving an oil. This was fi ltered through silica gel (eluent: petroleum ether (bp 40–60°C)). Excess bro-modecane was obtained after evaporation of the solvent. The silica gel was then eluted with acetone. The solvent was evaporated leaving a pale yellow oil which was used without further purifi cation. Yield 7 g (93%). 1H NMR (60 MHz): δ, ppm 9.8 (s, 1H), 7.1 (s, 2H), 4.1 (brt, 6H), 1.3 (brs, 48H), 0.9 (t, 9H).

5-(12-acetyloxydodecyloxyphenyl)-10,15,20-tri-(3,4,5-trisdecyloxyphenyl)porphyrin (2a). 3,4,5-tris -decy loxybenzaldehyde (3.51 g, 6 mmol) and 4-(12-acety-loxydodecyloxy)benzaldehyde (0.71 g, 2 mmol) were heated in propionic acid (40 ml) containing nitrobenzene (10 ml) to 130°C. Pyrrole (0.55 g, 8 mmol) was added at once and the temperature maintained for 3 h. The solution was then cooled and excess methanol added. The fl ask was placed in the fridge overnight. The dark oil at the bottom of the fl ask was separated and washed with methanol. The product was separated by column chromatography over silica gel (eluent: petroleum ether (bp 40–60°C)/THF 10:1) twice. The sample was used in the next stage without further purifi cation. 1H NMR (270 MHz): δ, ppm 8.86–8.94 (m, 8H), 8.1 (d, J = 8 Hz, 2H), 7.42 (s, 6H), 7.27 (d, J = 8 Hz, 2H), 4.29 (brt, 8H),

Fig. 4. Typical example of photokinetic plot describing the photo-sensitized oxidation of the tryptophyl residue in albumin for 1c



4.08 (t, 14H), 2.26 (s, 3H), 1.2–2.0 (m, 164H), 0.91 (t, 9H), 0.83 (t, 18H), -2.79 (s, 2H).

5-(12-hydroxydodecyloxyphenyl)-10,15,20-tri (3,4,5-trisdecyloxyphenyl)porphyrin (2b). 5-(12-acetyloxydo-decyl-oxyphenyl)-10,15,20-tri(3,4,5-trisdecyloxyphenyl)porphyrin obtained above was refl uxed in THF (10 ml). Ethanolic NaOH (excess) was added. The reaction was followed by tlc (eluent: petroleum ether (bp 40–60°C)/THF 10:1) until complete (rf starting mate-rial = 0.8, rf product = 0.2). The solvent was evaporated and the residue was chromatographed over silica gel (eluent: petroleum ether (bp 40–60°C)/THF 4:1). Yield 0.49 g (11% from benzaldehydes). 1H NMR (270 MHz): δ, ppm 8.85–9.0 (m, 8H), 8.1 (d, J = 8 Hz, 2H), 7.43 (s, 6H), 7.3 (d, J = 8 Hz, 2H), 4.3 (t, 8H), 4.09 (t, 12H), 3.65 (brt, 2H), 1.2–2.05 (m, 164H), 0.91 (t, 9H), 0.84 (t, 18H), -2.77 (s, 2H).

5-(12-methanesulfonyloxydodecyloxyphenyl)-10,15,20-tri(3,4,5-trisdecyloxyphenyl)porphyrin (2c). 5-(12-hydroxydodecyloxyphenyl)-10,15,20-tri(3,4,5-trisdecyloxyphenyl)porphyrin (0.193 g, 86 μmol) was dissolved in dichloromethane (8 ml) and the solution cooled with a water bath. Et3N (1 ml) was added, fol-lowed by methanesulfonylchloride (20 drops). The reac-tion was stirred at rt for 1 h. The organics were washed with a dilute HCl solution, brine, dried (MgSO4), fi ltered and the solvent removed under reduced pressure. The residue was purifi ed by column chromatography over silica gel (eluent: petroleum ether (bp 40–60°C)/THF 4:1). Yield 0.159 g (80%). 1H NMR (270 MHz): δ, ppm 8.85–9.0 (m, 8H), 8.1 (d, J = 8 Hz, 2H), 7.43 (s, 6H), 7.28 (d, J = 8 Hz, 2H), 4.2–4.35 (m, 10H), 4.09 (t, 12H), 2.97 (s, 3H), 1.1–2.0 (m, 192H), 0.91 (t,9H), 0.84 (t, 18H), -2.77 (s, 2H).

5,5′-(12,12′-dithiodidodecyloxyphenyl)-10,10′,15,15′,20,20′-hexa(3,3′,4,4′,5,5′-hexa kis decyloxy-phenyl) diporphyrin (1a). 5-(12-methanesulfonyloxydo-decyloxyphenyl)-10,15,20-tri(3,4,5-trisde cylox-yphenyl)-porphyrin (159 mg, 0.07 mmol) and thiourea (excess) were refl uxed in 1-pentanol (5 ml) until tlc (elu-ent: petroleum ether (bp 40–60°C)/THF 4:1) indicated no starting material was left (approx. 1 h) (rf product=0). Eth-anol was added (2 ml) followed by aqueous NaOH (10%) (2 ml) and refl ux continued for 5 min. The solution was left to cool and then added to dil. HCl (10%) and the organics were extracted with dichloromethane, washed with brine, dried (MgSO4), fi ltered and the solvent evapo-rated. The residue was chromatographed over silica gel (eluent: petroleum ether (bp 40–60°C)/THF 4:1). Yield 114 mg (74%). 1H NMR (270 MHz): δ, ppm 8.87–9.0 (m, 8H), 8.11 (d, J = 8 Hz, 2H), 7.44 (s, 6H), 7.3 (d, J =8 Hz, 2H), 4.31 (t, 8H), 4.27 (t, 2H), 4.1 (t, 12H), 2.71 (t, 2H), 1.2–2.05 (m, 164H), 0.92 (t, 9H), 0.85 (t, 18H), -2.76 (s, 2H). UV-vis (THF): λ, nm 650.5, 592.5, 552.5, 516.0, 423.0.

5,5′-(12,12′-dithiodidodecyloxyphenyl)-10,10′,15,15′,20,20′-hexa(3,3′,4,4′,5,5′-hexakisdecyloxyphenyl)-

diporphyrinatozinc (1b). 5,5′-(12,12′-dithiodidode-cyloxyphenyl)-10,10′,15,15′,20,20′-hexa(3,3′,4,4′,5,5′-hexakisdecyloxyphenyl)diporphyrin (114 mg, 25 μmol)was refl uxed in THF (5 ml). Zinc acetate dihydrate (20 mg, 90 μmol) was added and refl ux continued for 30 min. The solvent was evaporated and the residue chromatographed through a short column over silica gel (eluent: petroleum ether (bp 40–60°C)/THF 4:1). Yield 116 mg (99%). 1H NMR (270 MHz): δ, ppm 8.9–9.1 (m, 8H), 8.1 (d, J = 8 Hz, 2H), 7.42 (s, 6H), 7.27 (d, J =8 Hz, 2H), 4.29 (t, 8H), 4.24 (t, 2H), 4.09 (t, 12H), 2.67 (t, 2H), 1.2–2.05 (m, 164H), 0.91 (t, 9H), 0.83 (t, 18H). UV-vis (THF): λ, nm (ε × 105 M-1.cm-1) 598.0 (0.12), 557.0 (0.38), 427.5 (12.6), 406.5 (0.86). MALDI-MS: isotopic clusters at m/z 4601 [M]+ and 2300 ([M]2+,100%). Anal. calcd. for C292H462N8O20S2Zn2: C, 76.25; H, 10.12; N, 2.44%. Found: C, 76.39; H, 10.39; N, 2.12.

5,5′-(12,12′-dithiodidodecyloxyphenyl)-10,10′,15,15′,20,20′-hexa(3,3′,4,4′,5,5′-hexakisdecylo xy-phenyl)diporphyrinatomagnesium (1c). 5,5′-(12,12′-dithiodidodecyloxyphenyl)-10,10′,15,15′,20,20′-hexa(3,3′,4,4′,5,5′-hexakisdecyloxyphenyl)diporphyrin (100 mg, 22 μmol) was refl uxed in pyridine (10 ml, dried over KOH) under N2 atmosphere. Magnesium perchlorate (10 mg, 80 μmol) was added and refl ux continued for 4 h. After cooling, the mixture was poured onto water and extracted with diethyl ether. The organics were washed twice with dil. HCl (20%), brine, dried (MgSO4), fi ltered and the solvent evaporated. The product was purifi ed by column chromatography over silica gel (eluent: petro-leum ether (bp 40–60°C)/THF 10:1). Yield 92 mg (93%). 1H NMR (270 MHz): δ, ppm 8.8–9.1 (m, 8H), 8.1 (d, 2H), 7.25–7.5 (m, 8H), 3.9–4.4 (m, 22H), 2.7 (t, 2H), 1.1–2.1 (m, 164H), 0.8–1.0 (m, 27H). UV-vis (THF): λmax, nm (ε × 105 M-1.cm-1) 615.0 (0.23), 569.0 (0.26), 432.0 (10.77). MALDI-MS: isotopic clusters at m/z 4518 ([M]+, 100%) and 2260 [M]2+. Anal. calcd. for C292H462N8O20S2Mg2: C, 77.63; H, 10.31; N, 2.48%. Found: C, 77.39; H, 10.28; N, 2.42.

Deposition of SAM fi lms

Glass microscope slides (BDH Ltd.) were used as the substrates for the SAMs. The glass surface was wiped clean with a soft, detergent-free tissue and then washed with a stream of methanol in order to remove any bulk surface contamination. The glass substrates were then immersed into a solution of potassium hydroxide in aqueous methanol (100 g of potassium hydroxide was dissolved in 100 ml of Millipore water, then diluted to 250 ml with methanol) for 12 hours. The substrates were rinsed with fresh Millipore water and then dried in a stream of refl uxing propan-2-ol. The resultant clean sub-strates were stored in sample jars, with air-tight lids.

The cleaned, dried glass substrates were then coated with a layer of chromium (99.999% purity, Johnson Mat-they Ltd., 1 nm) followed by a layer of gold (99.999%



purity, Johnson Matthey Ltd., 45 nm). The chromium layer was deposited to ensure a good adhesion of the gold onto the glass surface. Both the chromium and gold lay-ers were deposited by thermal evaporation under vacuum using an Edwards Auto 306 vacuum evaporator.

The SAMs were formed by immersing the freshly prepared gold-coated substrates into a solution of known concentration of porphyrin, typically between 2.5 × 10-7

and 5 × 10-7 M in cyclohexane at room temperature for 20 h.

Reference substrates for the SAMs were fabricated by immersing freshly prepared gold-coated substrates into cyclohexane for 20 h. The SAMs and the references were then washed in fresh cyclohexane, dried in air and subse-quently stored in clean amber jars with air-tight lids.

Spectroscopic characterization of SAM fi lms

The BioRad RAIR spectrometer housed a liquid nitrogen cooled MCT detector and was coupled with a Spectra-Tech FT85 refl ectance unit. RAIR spectra of the fi lms were acquired using p-polarized light at an inci-dence angle of 85°. For further details see reference 3c. The spectra were obtained from the co-addition of 1024 scans at a resolution of 4 cm-1. Measurement of the fl uo-rescence emission and fl uorescence excitation spectra of the fi lms was achieved using a Perkin Elmer MPF4 spec-trophotofl uorimeter equipped with accessories for front-surface fl uorescence detection.

Photooxidation studies

Typically, a glass slide decorated with the SAM was placed in a Petri dish and covered by 2 ml of a phosphate-buffered (pH 7.4) aqueous solution of HSA; the initial protein concentration was 0.1 mM. The system thus obtained was irradiated, during gentle magnetic stirring, by full spectrum visible light emitted from a quartz/halo-gen lamp (Teclas, Lugano, Switzerland), which was oper-ated at a fl uence rate of 50 mW/cm2. The light source was equipped with a heat-refl ecting fi lter and a cut-off fi lter at 390 nm to eliminate any UV radiation. The light beam was piloted to the irradiation site by means of a bundle of optical fi bers (external diameter = 0.8 cm). During irra-diation, the temperature of the HSA solution was kept at 25°C by circulating thermostated water. At predeter-mined time intervals after the beginning of the irradia-tion, 0.1 ml aliquots of the HSA solution were taken and the tryptophan fl uorescence emission was measured in a Perkin Elmer MPF4 spectrophotofl uorimeter using 290 nm excitation and 360 nm emission; the excitation and emission slits were fi xed at 7 nm.

CONCLUSION

The research has demonstrated that SAM fi lms of derivatives of Zn and Mg metalated tetraphenylporphy-rins (TPPs) substituted with three long alkoxy chains on

each of three of the benzenoid rings, with a tether chain on the fourth, can serve as photodynamic surfaces. The choice of highly substituted TPP derivatives was made to disrupt or at least minimize close porphyrin-porphyrin ring interactions and this, together with a long tether to maintain a signifi cant distance of the core from the gold surface, has ensured that the photoreactive electronically excited states of the molecules are not fully quenched. The latter has been demonstrated by the observation of fl uorescence from the fi lm and further by the photooxi-dation of the tryptophan moiety within human serum albumin. The effi ciency of the porphyrin SAMs as het-erogeneous catalysts for photooxidations has potential applications within chemical synthesis and also within medical therapies. Thus other porphyrin derivatives, deposited onto appropriate surfaces, could in principle be incorporated within medical devices or implants where a photodynamic effect could be advantageous.

Acknowledgements

We thank the EPSRC for access to the mass spectrom-etry service at Swansea.

REFERENCES

1. a) Ulman A. An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Self-Assembly (1st edn.) Academic Press: California, 1991. b) Smith RK, Lewis PA and Weiss PS. Prog. Surface Sci.2004; 75: 1–68. c) Tao F and Bernasek SL. Chem.Rev. 2007; 107: 1408–1453.

2. a) Kondo T and Uosaki K. J. Photochem. Photobiol. C 2007; 8: 1–17. b) Bedioui F, Griveau S, Nyokong T, Appleby AJ, Caro CA, Gulppi M, Ochoa G and Zagal JH. Phys. Chem. Chem. Phys. 2007; 9:3383–3396. c) Love JC, Estroff LA, Kriebel JK, Nuzzo RG and Whitesides GM. Chem. Rev. 2005; 105: 1103–1169. d) Schreiber F. J. Phys-Condensed Matter 2004; 16: R881–R900. e) Witt D, Klajn R, Barski P and Grzybowski BA. Curr. Org. Chem.2004; 8: 1763–1797.

3. a) Chambrier I, Cook MJ and Russell DA. Syn-thesis 1996; 1283–1386. b) Simpson TRE, Revell DJ, Cook MJ and Russell DA. Langmuir 1997; 13:460–464. c) Revell DJ, Chambrier I, Cook MJ and Russell DA. J. Mater. Chem. 2000; 10: 31–37.

4. Cook MJ, Hersans R, McMurdo J and Russell DA. J. Mater. Chem. 1996; 6: 149–154.

5. a) Simpson TRE, Russell DA, Chambrier I, Cook MJ, Horn AB and Thorpe SC. Sensors Actuators B 1995; 29: 353–357. b) Simpson TRE, Cook MJ, Petty MC, Thorpe SC and Russell DA. Analyst1996; 121: 1501–1505.

6. Hone DC, Walker PI, Evans-Gowing R, FitzGerald S, Beeby A, Chambrier I, Cook MJ and Russell DA. Langmuir 2002; 18: 2985–2987.



7. Wieder ME, Hone DC, Cook MJ, Handsley MM, Gavrilovic J and Russell DA. Photochem. Photo-biol. Sci. 2006; 5: 727–734.

8. a) Watcharinyanon S, Nilsson D, Moons E, Shaporenko A, Zharnikov M, Albinsson B, Maartensson J and Johansson LSO. Phys. Chem. Chem. Phys. 2008; 10: 5264–5275. b) Zuo G, Liu X, Yang J, Li X and Lu X. J. Electroanal. Chem.2007; 605: 81–88. c) Lu X, Zhang L, Li M, Wang X, Zhang Yan, Liu X and Zuo G. Chem. Phys. Chem.2006; 7: 854–862. d) Yasseri AA, Syomin D, Mal-inovskii LV, Loewe RS, Lindsey JS, Zaera F and Bocian DF. J. Am. Chem. Soc. 2004; 126: 11944–11953. e) Rawlett AM, Hopson TJ, Amlani I, Zhang R, Tresek J, Nagahara LA, Tsui RK and Goronkin H. Nanotechnology 2003; 14: 377–384. f ) Bramblett AL, Boeckl MS, Hauch KD, Ratner BD, Sasaki T and Roger JW. Surf. Interface Anal. 2002; 33: 506–515. g) Roth KM, Gryko DT, Clausen C, Li J, Lind-say JS, Kuhr WG and Bocian D. J. Phys. Chem. B2002; 106: 8639–8648. h) Imahori H, Hasobe T, Yamada H, Nishimaru Y, Yamazaki I and Fukuzumi S. Langmuir 2001; 17: 4925–4931. i) Boeckl MS, Bramblett AL, Hauch KD, Sasaki T, Ratner BD and Roger JW. Langmuir 2000; 16: 5644–5653. j) Imahori H, Norieda H, Nishimaru Y, Yanmazaki I, Higuchi K, Kato N, Motohiro T, Yamada H, Tamaki K, Arimura M and Sakata Y. J. Phys. Chem. B 2000; 104: 1253–1260. k) Shimazu K, Takechi M, Fujii H, Suzuki M, Saiki H, Yoshimura T and Uosaki K. ThinSolid Films 1996; 273: 250–253. l) Hutchison JE, Postlethwaite TA and Murray RW. Langmuir 1993; 9: 3277–3283. m) Zak J, Yuan H, Ho M, Woo LK and Porter MD. Langmuir 1993; 9: 2772–2774.

9. Benitez IO, Bujoli B, Camus LJ, Lee CM, Odobel F and Talham DR. J. Am. Chem. Soc. 2002; 124:4363–4370.

10. Berner S, Biela S, Ledung G, Gogoll A, Baeckvall J-E, Puglia C and Oscarsson S. J. Catalysis 2006;244: 86–91.

11. Zhang X-L, Chen L-G, Lv P, Gao H-Y, Wei S-J, Dong Z-C and Hou JG. Appl. Phys. Lett. 2008; 92:223118/1–223118/3.

12. Jori G and Spikes JD. In Topics in Photomedicine,Smith KC. (Ed.) Plenum Press: New York, 1983; pp 183–319.

13. Meier H, Prass E, Zerban G and Kosteyn K. Z.Naturforsch. B 1988; 43: 889–896.

14. Johnstone RAW, Nunes MLPG, Pereira MM, Gon-salves AMDR and Serra AC. Heterocycles 1996; 43: 1423–1437.

15. Imahori H, Kashiwagi Y, Endo Y, Hanada T, Nishimura Y, Yamazaki I, Araki Y, Ito O and Fuku-zumi S. Langmuir 2004; 20: 73–81.

16. Silva E, Ugarte R, Andrade A and Edwards AM. J. Photochem. Photobiol. B 1994; 23: 43–48.

17. Jori G and Spikes JD. In Oxygen and Oxy-radicals in Chemistry and Biology, Rodgers MAJ and Pow-ers EL. (Eds.) Academic Press: New York, 1981; pp 441–459.

18. Foote CS. In Free Radicals in Biology, Pryor WA. (Ed.) Academic Press: New York, 1976; pp 85–133.

19. Benassi CA, Scoffone E, Galiazzo G and Jori G. Photochem. Photobiol. 1987; 6: 857–866.

20. Reddi E, Valduga G, Rodgers MAJ and Jori G. Photochem. Photobiol. 1991; 54: 633–637.


DOI: 10.1142/S1088424610001763



INTRODUCTION

Tolyporphin A was isolated in 1992 by Moore and co-workers from the lipophilic extract of the blue-green microalga Tolypothrix nodosa [1]. This structurally unique porphyrin, which may strengthen the cytotoxicity of adriamycin or vinblastine in human ovarian adenocar-cinoma cells at doses as low as 1 μg.mL-1, is character-ized as a multidrug resistance (MDR) reversing agent [2]. Subsequently, ten additional tolyporphins (B-K) which have chromophoric skeleton similar to that of tolypor-phin A were isolated and were found to possess varying degrees of anti-MDR ability [3, 4]. On the basis of exten-sive spectroscopic studies, the structure of tolyporphin A, the representative member of the tolyporphin class of natural products, was concluded to be 1, which belongs to bacteriochlorins [1]. In 1997 Minehan and Kishi pro-posed a concept and a synthesis of a chromophore model of tolyporphin 2 [5], and accordingly reported a total synthesis of tolyporphin A O,O-diacetate with a “2+2” method [6, 7]. According to the synthesis, the bulk tet-rahydropyrane-ring side chains in the cycle A and C of 1which were introduced by the building blocks of pyrrole derivatives, respectively, participate in the whole process

of the formation of the tetrapyrrolic macrocycle and seri-ously affect the transformation and yield of the reaction [8]. Here, we propose a new structural model of tolypor-phin 3, which refl ects better the steric confi guration of the natural tolyporphin compounds than 2. Tolyporphin 3can be used to synthesize various tolyporphins by trans-forming the methyl acetate side chains in the cycle A and C into other groups, which is very important to study the biological activity of tolyporphins. This synthetic route has the advantage of dispelling the steric hindrance effect of the tetrahydropyrane side chain on the reaction and promoting the process greatly. In this paper, we report our work on the synthesis of A, B and C rings and their combination A-B-C-tricycle fragment, unambiguously establishing the absolute confi gurations of the key build-ing blocks for 3.


Strategy and retrosynthetic analysis. The A-B-C-tricycle strategy

It is known from the work of Montforts et al. [9] that the “3+1” method has clear superiority over other methods in syntheses of cyclic tetrapyrroles, such as porphyrins, chlorins and corrins. The method, namely,

Effi cient synthesis of an A-B-C-tricycle fragmentfor a structural model of tolyporphin

Bing C. Hu*, Wei Y. Zhou, Zu L. Liu, Chao J. Cai and Shi C. Xu

School of Chemical Engineering, Nanjing University of Science & Technology, 200 Xiaolingwei Street, Nanjing 210094, China

Received 9 January 2009Accepted 13 March 2009

ABSTRACT: An effi cient stereocontrolled synthesis of an A-B-C-tricycle fragment 7 for a structural model of tolyporphin 3 is described. All the rings were prepared from readily available starting materials. One of the two key steps is a selective ring-opening reaction of the lactone cycle in bicyclolactam-lactone 17 to cyanopyrrolidone 18, which introduces the chirality into synthetic compounds. The other key step is the combination of A ring with B-C-bicycle via a two-time Eschenmoser sulfi de contraction. A-B-C-tricycle fragment 7 allows a new approach toward tolypophin compounds and other uroporphinoids.

KEYWORDS: tolyporphin, structural model, selective ring-opening reaction, sulfi de contraction.

*Correspondence to: Hu B. Cheng, email: [email protected]


90 B.C. HU ET AL.

synthesizing the linear tripyrrole intermediate A-B-C-tricycle fragment fi rstly, then combining it with the monomer D ring to afford the target compound (B-C-D-tricycle and A ring can also be used as the intermediates), adopts the way of connecting the pyrrole monomers one by one, which is convenient for adjusting their positions and joining sequence [8]. Consequently, we decided to use a convergent and fl exible A-B-C + D strategy in the synthesis of 3 and utilize four pyyrole derivatives, 9a, 10, 11 and 8, as the elementary building blocks (Scheme 2).

The four pyrrole derivatives, the rings A, B, C, and D, could be synthesized individually from readily avail-able starting materials, and then connected together in a defi nite sequence to form 6. Here, the conversion from 6to 5 could be realized via acetyl substitution with acetyl-halide followed by stereospecifi c reduction of the acetyl group. Amide 4 could be formed via a stereospecifi c

amide acetal Claisen rearrangement with N,N-dimethyl-acetamide-dimethylacetal from alcohol 5. Hydrolysis of amide 4 followed by oxidation could provide diester 3.

Retrosynthetic analysis of 7

In our synthesis plan for A-B-C-tricycle fragment 7,2-pyrrolinone 10 is in the central position. According to the protocol developed by Eschenmoser et al. in the syn-thesis of corrins [10], the electrophilic carbonyl group of pyrrolinone 10 should be connected with thione 9a via a selectively splitable, nucleophilic ester 12. Coupling of the nucleophilic 5-position of pyrrolinone 10 with the aldehyde group of pyrrole 11 could be easily realized viaa base-catalyzed condensation [11].

Synthesis of A ring (9a)

For the synthesis of thionone 9a, we used a fi ve-step route from 2,3-butanedione 13 and malononitrile 14 (Scheme 3). Condensation of 2,3-butandione with malononitrile gave bicyclolactam 15. Under acidic condi-tion 15 was hydrolyzed to bicyclolactone 16, which con-verted to bicyclolactam-lactone 17 upon treatment with concentrated aqueous ammonia. In the molecule of 17C-1 atom is very reactive as it connects simultaneously with electronegative oxygen and nitrogen atoms. It is lia-ble for the C-O bond on C-1 atom to split by the action of nucleophilic reagents [12–14]. By the action of strongly nucleophile hydrogen cyanide 17 was easily converted

Scheme 1

Scheme 2


EFFICIENT SYNTHESIS OF AN A-B-C-TRICYCLE FRAGMENT 91

to cyanopyrrolidone 18 via substitution. We have shown that P4S10/NaHCO3, rather than Lawesson’s reagent, is the effective promoter in the sulfurization of 18 [15]. Lit-tle target product was observed with Lawesson’s reagent as the sulfurizing agent, which might be caused by partial hydrolysis of the reagent in the reaction. By the action of the reagent P4S10/NaHCO3 in THF cyanopyrrolidone 18 was smoothly converted to thionone 9 in 78% yield. Without sodium hydrogen carbonate very low yield of thionone 9 was observed. Here, cyanopyrrolidone 18and thionone 9 are both diastereoisomeric mixtures. The mono-isomeric 9a and 9b, whose absolute confi gura-tions were determined via differential 1H NOE (nuclear Overhauser effect) spectroscopy, were obtained via semi-preparative HPLC separation from the diastereoisomeric mixture 9 in 64.9% and 34.3% yield, respectively [16].

Synthesis of B ring (10)

In our fi rst approach to the synthesis of pyrrolinone 10, we have successfully prepared its important precursor lactam 23 in 18% overall yield (Scheme 4). Starting from commercially available γ-butanolide 19, bromide 20 was synthesized via bromination with bromine and phosphor.

Homologation to lactam 23 was accomplished by sub-stitution with triphenyl phosphine followed by addition with polyformaldehyde (Wittig-reactions) and fi nally by ammoniation. Unfortunately, though we had tried vari-ous ways to carry out the transformation from lactam 23into pyrrolinone 10, we failed to acquire the target product. The result shows that the exocyclic double bond in 23 is more stable than the endocyclic double bond in 10 [17].

This caused us to investigate an alternate method of preparing pyrrolinone 10 (Scheme 5). Reaction of oxal-acetic ester 24 with ethyl acetate followed by amino-acetone condensation provided pyrrole 26, which was converted to pyrrole 27 via morpholine. After oxida-tion with hydrogen peroxide to pyrrolinone 28, the C-4 acetate side chain was removed via decarboxylation reaction to afford 10.

The synthesis of C ring (11)

The synthesis of C ring (11) is shown in Schemes 6 and 7. For the synthesis of dimethylpyrrole 35, we used a novel one-pot three-component reaction which simpli-fi es the preparation process remarkably. Thus, treatment

Scheme 3

Scheme 4


92 B.C. HU ET AL.

of the nitro sation reaction mixture of acetacetic ether 29with the Claisen condensation reaction mixture of ethyl formate 32 and butanone 33 gave dimethylpyrrole 35 in 60% yield, in the process of which the nitrosation prod-uct hydroximino-ester 30 was reduced by the action of zinc powder to amino ester 31 fi rstly.

We have shown that bromo-formylpyrrole 11 can be formed from dimethylpyrrole 35 preferably via bromiza-tion before oxidation. Thus, dimethylpyrrole 35 was fi rstly converted to the bromide 36 on treatment with liq-uid bromine. We have found that the solvent has notable effect on the bromization. With carbon tetrachloride as the solvent the yield of bromide 36 reached as high as 87%, while the yield was not more than 53% with other solvents, such as dichloromethane, chloroform, ethanol and acetic acid. The conversion from bromide 36 to bro-mo-formylpyrrole 11 involves a selective oxidation for α-alkyl substituents of pyrroles, in which the oxidation depth of the α-methyl substituent in 36 should also be controlled as changing it into formyl rather than carboxyl substituent. Consequently, the crux of this reaction lies in selecting an appropriate oxidizer [18]. We have investi-gated various oxidizers for their reactive behavior in this oxidation and found that ammonium ceric nitrate (ACN) is the most effective reagent with 90% yield of bromo-formylpyrrole 11, while other selected oxidizers such as sulfuryl chloride, lead tetraacetate, sodium periodate and hydrogen peroxide afforded bromo-formylpyrrole 11 in yields ranging from 6 to 73%.

We have also investigated an alternative way, namely via oxidation before bromization, to synthesize bromo-formylpyrrole 11 from dimethylpyrrole 35, and found that the yield of bromization drops greatly. From the

result it is concluded that the α-electron-withdrawing group of pyrroles can restrain their β-bromization (β-electrophilic substitution) while the α-electron-donating group can promote the β-bromization (β-electrophilic substitution).

Synthesis of B-C-bicycle fragment (39)

Many dipyrrylmethenes may be synthesized via the base-catalyzed condensation between pyrrolinones or N-acyl pyrrolinones and α-pyrryl-aldehydes. For the condensation of pyrrolinone 10 and bromo-formylpyr-role 11, the existence of base can also cause the ester group of the condensation product 38 to hydrolyze and decrease the yield of product (Scheme 8). In view of the above, we have investigated various bases and found that 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) is the most effective reagent with 65% yield of bicycle 38, while other organic bases such as piperidine, dimethylam-inopyridine, tetrabutylammonium bromide and trieth-ylamine afforded bicycle 38 in yields from 16 to 50%, inorganic alkalis such as sodium hydroxide, sodium eth-ylate and potassium tert-butoxide provided only a little of product.

Scheme 5

Scheme 6

Scheme 7



According to our synthesis strategy, the next work was to incorporate the saturated A ring 9a into the B-C-bicycle 38 via sulfi de contraction. Therefore, bicycle 38must be converted to its thioanalog 39. By the action of P4S10/NaHCO3 in THF the sulphurization went on very smoothly in 82% yield.

Synthesis of an A-ring precursor enamine 43 via sul-fi de contraction

The sulfi de contraction was originally developed by Eschenmoser and Fischli during the synthesis of vitamin B12 [19], which in principle represents an intramolecular aza-analogous Claisen condensation. According to the protocol developed by Eschenmoser et al. [10], treatment of a mixture of thionone 9a and the selective splitable malonic ester 12 with dibenzoyl peroxide furnished the stable thioiminoester 41 (Scheme 9). This reaction is pre-sumably initiated by the formation of disulfi de 40, which is in situ cleaved by a nucleophilic attack from malonic ester 12. Thereby thionone 9a is regenerated and can be oxidized with dibenzoyl peroxide again. On being heated in triethyl phosphite (80 °C, 8 h), 41 was desulfurized to form enamine 42, as a mixture of E, Z-isomers. By the action of piperidine/Pd(PPh3)4 in THF, enamine 42was converted via elimination of the allyl ester and decar-boxylation into the single Z-isomer enamine 43, which is thermodynamically stabile owing to one intramolecular hydrogen bridge. Isomerically pure 43, whose purity was confi rmed via HPLC, was obtained in 58% yield from thionone 9a after column chromatography separation.

Synthesis of A-B-C-tricycle fragment (7)

The next synthetic target was the linkage of the B-C-bicycle 39 with the vinyl carbon atom of A-ring precur-sor enamine 43 via sulfi de contraction (Scheme 10). We have gained excellent connection results by means of “brominated connection” [20]. Thus, enamine 43 was brominated at its vinyl position on treatment with N-bro-mo-succinimide (NBS) to form a purple brown bromide 44, which was immediately and without purifi cation connected with bicycle 39 by the action of DBU. The latter operation is essential to the preparation of tricy-clic sulfi de 46, because bromide 44 is extremely sensi-tive to light and oxygen. We once attempted to separate bromide 44 from the reaction mixture, as a result fi nding that most of the separated product decomposed. Presum-ably, bromoenamine 44 wss converted to bromoimine 45before reacting with bicycle 39, in the course of which an equilibrium existed between 44 and 45. As a mild base, DBU can incline this equilibrium to the creation of bro-moimine 45 as well as enhance the reactivity of bicycle 39 via deprotonation.

Although two E, Z-isomers of the tricyclic sulfi de can be formed, only the thermodynamically favored E-confi gurated product 46 was observed because of the intramolecular hydrogen bridge between the tert-butoxy-carbonyl group and the enamine-hydrogen. According to an improved sulfi de contraction method developed by Rasetti et al. [21], the C-C-connected tricycle 7 was suc-cessfully generated from the sulfur-bridged tricycle 46upon treatment with CF3COOH and P(CH2CH2CN)3 under

Scheme 8

Scheme 9


94 B.C. HU ET AL.

refl ux for 3 h. Presumably, this new sulfi de contraction proceeds via a tandem reaction mechanism as tert-butoxy cleavage-decarboxylation-desulfuration. In the presence of the strongly thiophilic reagent P(CH2CH2CN)3, which played a decisive role in this acid-catalyzed reaction, sulfi de 46 was desulfurated after the loss of the tert-bu-toxycarbonyl group. After isolation by chromatography on silica gel/aluminum oxide with dichloromethane/pet-ro-ether 2:1 as eluents, the sterically identical tricycle 7,which conjecturally possesses via intramolecular hydro-gen bridges stabilized Z,Z-confi guration, was obtained as fi ne orange-red needle-like crystals in about 48% yield from sulfi de 46. Spectroscopic data, 1H NMR, UV-vis, IR, MS and elemental analysis, confi rmed the structure shown in Scheme 2.

EXPERIMENTAL

Melting points determined on a Kofl er hot-stage- microscope were uncorrected. IR spectra were recorded on a Perkin-Elmer Paragon 500 FT-IR in KBr. 1H NMR spectra were run using a Brucker DPX-200 (or DPX-360) spectrometer, with tetramethylsilane and CDCl3 as the standard and the stated solvent, respectively; chemical shift values are quoted on a δ scale relative to tetrameth-ylsilane as δ = 0. MS spectra were measured on a Finni-gan MAT-8200 with ion-bombardment-energy 70 eV (or ThermoFinnigan TSQ Quantum Ultra AM with ESI ion source). Elemental analysis was completed on a Perkin-Elmer PE-2400. Analytical thin-layer chroma-tography was carried out on commercial Fluka plates coated with silica gel 60 F254 (0.25 mm). The diam-eter and aperture of the silica gel (ICN Biomedicals) for column chromatography are 32–63 μm and 6 nm, respectively.

1,5-dimethyl-2,8-diaza-3,7-dioxo-bicyclo[3.3.0]octane-4,6-dinitrile (15). To a solution of 3.45 g (150 mmo1) of natrium in 300 mL of dry ethanol were succes-sively added 12.9 g (150 mmo1) of butanedione 13 and 19.8 g (300 mmo1) of malononitrile 14 in 150 mL of dry ethanol at 0 °C over 1 h. After 4 h under argon at 5 °C, the mixture was acidifi ed with HCl (37%) until the pH was 1.0 and laid aside for 1 h. The precipitate was fi ltered, washed with water and dried to give 10.36 g (48 mmo1,

32%) of bicyclolactam 15 as colorless crystals: mp > 220 °C (decomposed). 1H NMR (200 MHz, DMSO-d6):δH, ppm 1.29 (3H, s, 5-CH3), 1.35 (3H, s, 1-CH3), 4.62 (1H, s, CHCN), 4.84 (1H, s, CHCN), 9.22 (1H, s, NH), 9.38 (1H, s, NH). IR (KBr): ν, cm-1 3313 (s, N–H), 2989 (w, C–H), 2902 (s, C–H), 2255 (s, C≡N), 1750 (s, C=O, lactam), 1710 (s, C=O, lactam). MS (EI, 70 ev): m/z (%) 218 ([M]+, 15), 217 ([M+ - H], 14), 203 ([M+ - CH3], 20), 175 (63), 160 (23), 147 (18), 135 (100). Anal. calcd. for C10H10N4O2: C, 55.04%; H, 4.62%; N, 25.68%. Found: C, 55.16%; H, 4.69%; N, 25.88%.

1,5-dimethyl-2,8-dioxo-bicyclo[3.3.0]octane-3,7-dione (16). Bicyclolactam 15 (8.80 g, 40 mmol) was dis-solved in 400 mL of boiling HBr (48%). After 1 h under refl ux, the solvent was removed by rotary evaporation and the residue was dissolved in 100 mL of CH2Cl2. The solution was fi ltered and concentrated to give a sticky red residue. Purifi cation by silica gel chromatography (95:5 CH2Cl2/CH3OH, Rf = 0.29) afforded 3.41 g (20 mmo1, 50%) of bicyclolactone 16 as colorless crystals: mp 128–129 °C. 1H NMR (200 MHz, CD3Cl): δH, ppm 1.40 (3H, s, 5-CH3), 1.71 (3H, s, 1-CH3), 2.72 (4H, q, AB-system, 2J = 6.8 Hz, 2 × CH2). IR (KBr): ν, cm-1 2993 (s, C–H), 2954 (s, C–H), 1794 (s, C=O, lactone), 1778 (s, C=O, lactone). MS (EI, 70 eV): m/z (%) 171 ([M+ + H], 24), 126 ([M+ - CO2], 14), 113 (16), 100 (26), 82 (34), 55 (73), 43 (100). Anal. calcd. for C8H10O4: C, 56.47%; H, 5.92%. Found: C, 56.24%; H, 6.07%.

1,5-dimethyl-2-aza-8-oxa-bicyclo[3.3.0]octane-3,7-dione (17). Bicyclolactone 16 (3.06 g, 18 mmo1) was suspen ded in 90 mL of concentrated aqueous ammonia at room temperature (rt), from which a limpid solution emerged in a short time. After 28 h under argon at rt, the solvent was removed by rotary evaporation. The yel-low residue was purifi ed by silica gel chromatography (95:5 CH2Cl2/CH3OH, Rf = 0.52) to produce 2.63 g (15.5 mmo1, 86%) of bicyclolactam-lactone 17 as colorless crystals: mp 192–193 °C. 1H NMR (200 MHz, CD3Cl):δH, ppm 1.34 (3H, s, 5-CH3), 1.62 (3H, s, l-CH3), 2.47 (2H, q, AB-system, 2J = 7.2 Hz, 6-CH2), 2.73 (2H, q, AB- system, 2J = 7.6 Hz, 4-CH2), 6.77 (1H, s, NH). IR (KBr): ν, cm-1 3210 (s, N–H), 2975 (s, C–H), 2888 (s, C–H), 1779 (s, C=O, lactone), 1698 (s, C=O, lactam). MS (EI, 70 eV): m/z (%) 170 ([M+ + H], 5), 125 ([M+ - CO2], 42),

Scheme 10



110 ([M+ - C2H3O2], 100), 82 (40). Anal. calcd. for C8H11NO3: C, 56.78%; H, 6.56%; N, 8.28%. Found: C, 56.65%; H, 6.63%; N, 8.36%.

5-cyano-2,3,4,5-tetrahydro-4,5-dimethyl-4-meth-oxycarbonylmethyl-1H-pyrrol-2-one (18). A solution of 2.37 g (14 mmol) of bicyclolactam-lactone 17 in 160 mL of dry methanol was treated under argon with 1.82 g (28 mmol) of KCN. After 20 h at rt, about 3/4 of the sol-vent were removed by rotary evaporation and the remain-der was transferred to a solution of 120 mL of NaH2PO4

(2 M). At 0 °C, H3PO4 (85%) was added to the mixture until the pH was 2.0. After being saturated with NaCl, the mixture was extracted with acetic ether (3 × 100 mL). The combined organic layers were dried (Na2SO4), fi l-tered and concentrated. The obtained colorless oily liquid was dried under vacuum and then dissolved in 80 mL of dry methanol. A solution of 50 mL of diazomethane in diethyl ether (0.5 M) was added at 0 °C to the mixture in 45 min. After 1 h at rt, the superfl uous diazomethane and the solvents were removed by rotary evaporation. Column chromatography on silica gel (95:5 CH2Cl2./CH3OH)afforded 2.59 g (12.3 mmol, 88%) of colorless crystalline cyanopyrrolidone 18 as a diastereoisomeric mixture: mp 127–128 °C. UV-vis (MeOH): λmax, nm (ε) 207 (12400). IR (KBr): ν, cm-1 3180 (s, N–H), 3115 (s, N–H), 2975 (s, C–H), 2230 (m, C≡N), 1735 (s, C=O, ester), 1684 (s, C=O, lactam). MS (EI, 70 eV): m/z (%) 210 ([M]+, 18), 179 ([M+ - OCH3], 47), 142 (28), 114 (100), 86 (29).

For the purpose of analysis, 0.181 g of the diastereoi-someric mixture 18 was separated via semi-preparative HPLC (polygosil 60–10, 10:14 n-hexane/acetic ther, 3 mL.min-1, test wavelength 207 nm, column temperature 40 °C) to give 0.118 g (65.2%, tR = 5.4 min) of trans-iso-mer 18a and 0.062 g (34.2%, tR = 7.2 min) of cis-isomer18b. 18a. TLC (silica gel, 10:13 n-hexane/methylac-etate): Rf = 0.54. 1H NMR (200 MHz, CD3Cl): δH, ppm 1.20 (3H, s, 4-CH3), 1.70 (3H, s, 5-CH3), 2.78 (2H, q, AB-system, 2J = 6.4 Hz, 3-CH2), 3.06 (2H, q, AB-system, 2J = 6.6 Hz, 4-CH2), 3.74 (3H, s, O-CH3), 8.11 (1H, s, br, NH). Anal. calcd. for C10H14N2O3: C, 57.13%; H, 6.71%; N, 13.32%. Found: C, 57.24%; H, 6.55%; N 13.45%. 18b. TLC (silica gel, 10:14 n-hexane/methylacetate): Rf = 0.50. 1H NMR (200 MHz, CD3Cl): δH, ppm 1.48 (3H, s, 4-CH3), 1.68 (3H, s, 5-CH3), 2.72 (2H, q, AB-system, 2J = 6.6 Hz, 3-CH2), 3.01 (2H, q, AB-system, 2J = 6.8 Hz, 4-CH2), 3.75 (3H, s, O-CH3), 8.36 (1H, s, br, NH).

5-cyano-2,3,4,5-tetrahydro-4,5-dimethyl-4-meth-oxycarbonylmethyl-1H-pyrrol-2-thione (9). A suspen-sion of 2.67 g (6 mmol) of P4S10 and 6.06 g (72 mmol) of NaHCO3 in 80 mL of dry THF was treated under argon with 2.52 g (12 mmol) of cyanopyrrolidone 18 in 100 mL of dry THF. After being refl uxed under argon for 2 h, the mixture was cooled, fi ltered and washed with CH2Cl2.The organic layer was dried (Na2SO4), fi ltered and con-centrated. Column chromatography on silica gel (4:1 CH2Cl2/ethyl acetate) afforded 2.15 g (9.5 mmol, 79%) of colorless crystalline thionone 9 as a diastereoisomeric

mixture. Through semi-preparative HPLC (polygosil 60-10, 65:35 n-hexane/acetic ther, 2 mL.min-1, test wave-length 265 nm, column temperature 40 °C), 2.020 g of the diastereoisomeric mixture 9 were sepa rated to afford 1.311 g (64.9%, tR = 4.1 min) of trans-isomer 9a and 0.692 g (34.3%, tR = 4.9 min) of cis-isomer 9b. thion-one 9: mp 118–119 °C. UV-vis (CHCl3): λmax, nm (ε) 270 (15200). IR (KBr): ν, cm-1 3449 (s, N–H), 3115 (s, N–H), 2982 (m, C–H), 2233 (w, C≡N), 1734 (s, C=O, ester), 1516 (s, C=S, thiolactam). MS (EI, 70 eV): m/z (%) 227 ([M+ + H], 11), 226 ([M]+, 74), 211 ([M+ - CH3], 13), 195 ([M+ - CH3O], 15), 167 ([M+ - C2H3O2], 8), 153 ([M+

- C3H5O2], 100). 9a. TLC (silica gel, 65:35 n-hexane/methylacetate): Rf = 0.61. 1H NMR (200 MHz, CD3Cl):δH, ppm 1.23 (3H, s, 4-CH3), 1.70 (3H, s, 5-CH3), 2.80 (2H, q, AB-system, 2J = 6.0 Hz, 3-CH2), 3.09 (2H, q, AB-system, 2J = 5.8 Hz, 4-CH2), 3.74 (3H, s, O-CH3), 8.24 (1H, s, br, NH). Anal. calcd. for C10H14N2O2S: C, 53.08%; H, 6.24%; N, 12.38%; S 14.17%. Found: C, 52.86%; H, 5.83%; N, 12.43%; S, 13.66%. 9b. TLC (silica gel, 65:35 n-hexane/methylacetate): Rf = 0.55. 1H NMR (200 MHz, CD3Cl): δH, ppm 1.44 (3H, s, 4-CH3), 1.66 (3H, s, 5-CH3),2.67 (2H, q, AB-system, 2J = 6.2 Hz, 3-CH2), 3.04 (2H, q, AB-system, 2J = 6.0 Hz, 4-CH2), 3.71 (3H, s, O-CH3),9.18 (1H, s, br, NH).

The confi gurations of cyanopyrrolidone 18a,b and thione 9a,b were determined via differential 1H NOE spectroscopy. In the NOE tests, the protons irradiated by a second radio-frequency fi eld were the 5-methyl-protons and the value of the correlated time between molecules (τm) was 2 s. Intensity of the NOE effect was expressed as increment percentage of the peak area of the observed protons. The results are shown in Tables 1 and 2.

4-methyl-3-ethoxycarbonyl-1H-pyrrole-2-carboxylic acid (26). To a solution of 3.45 g (150 mmo1) of natrium in 300 mL of dry ethanol were added 21.92 g (150 mmo1) of diethyl oxalate 24 and 14.10 g (160 mmo1) of ethyl acetate at 0 °C over 1 h. After 2 h under argon at 40 °C, the solvent and the superfl uous ethyl acetate were removed by rotary evaporation. The residue was treated with an ice cold solution of 100 mL of glacial acetic acid and 150 mL of water, and then extracted with diethyl ether (3 × 100 mL). The aqueous layer was exacted again with diethyl ether (2 × 100 mL). The combined organic layers were dried (Na2SO4), fi ltered and concentrated to give 19.47 g (103.5 mmol, 69%) of diethyl oxaloacetate 25 as colorless crystals. The obtained diethyl oxaloacetate 25 was dissolved in 200 mL of CH2Cl2 and then treated with 13.14 g (120 mmol) of acetonylamine hydrochlo-ride (prepared in our own laboratory). NaOH (1.0 M) was added to the mixture until the pH was 8.5–8.8. After 2 h under refl ux, the mixture was cooled, acidifi ed with HCl (37%) until the pH was 2.0, and laid aside for 1 h. The aqueous layer was extracted with CH2Cl2 (3 × 100 mL). The combined organic layers were dried (Na2SO4), fi l-tered and concentrated. Recrystallization from CH2Cl2/


96 B.C. HU ET AL.

petrol-ether afforded 15.37 g (78.0 mmol, 52% from 24)of pyrrole 26 as colorless crystals: mp 132–133 °C. UV-vis (CH3OH): λmax, nm (ε) 206 (11480), 265 (6210). 1HNMR (200 MHz, CD3Cl): δH, ppm 1.47 (3H, t, 3J = 7.2 Hz, CO2CH2CH3), 2.29 (3H, s, 3-CH3), 4.48 (2H, q, 3J = 7.1 Hz, CO2CH2CH3), 6.82 (1H, s, =CH), 9.95 (1H, br, NH), 14.75 (1H, br, COOH). IR (KBr): ν, cm-1 3420 (s, N–H), 3014 (s, N–H), 2895 (s, C–H), 1778 (s, C=O, ester), 1682 (s, C=O, acid), 1605 (m, C=C). MS (EI, 70 eV): m/z (%) 197 ([M]+, 48), 152 ([M+ - CO2H], 37), 107 ([M+ - CO2H - OC2H5], 100), 79 ([M+ - CO2H - CO2C2H5], 14). Anal. calcd. for C9H11NO4: C, 54.80%; H, 5.63%; N, 7.11%. Found: C, 54.65%; H, 5.74%; N, 7.23%.

4-methyl-1H-pyrrole-3-carboxylic ethyl ester (27).A solution of 13.00 g (66 mmol) of pyrrole 26 in 150 mL of DMF was treated with 16.80 g of morpholine. After 5 h under refl ux, the mixture was cooled, fi ltered and washed with DMF. The fi ltrate was concentrated, dried and recrystallized from CH2Cl2/petrol-ether to give 6.57 g (42.9 mmol, 65%) of pyrrole 27 as colorless crystals: mp 151–152 °C. UV-vis (CH3OH): λmax, nm (ε) 210 (14215), 276 (5630). 1H NMR (200 MHz, CD3Cl): δH, ppm 1.56 (3H, t, 3J = 8.2 Hz, CO2CH2CH3), 2.39 (3H, s, 3-CH3),4.58 (2H, q, 3J = 7.3 Hz, CO2CH2CH3), 6.24 (1H, s, 5-H), 8.38 (1H, s, 2-H), 10.97 (1H, br, NH). IR (KBr): ν, cm-1

3430 (s, N–H), 2914 (s, C–H), 1724 (s, C=O, ester), 1613 (m, C=C). MS (EI, 70 eV): m/z (%) 153 ([M]+, 67), 124 ([M+ - C2H5], 54), 108 ([M+ -OC2H5], 100), 80 ([M+ - CO2C2H5], 19). Anal. calcd. for C8H11NO2: C, 62.71%; H, 7.24%; N, 9.15%. Found: C, 62.54%; H, 7.35%; N, 9.28%.

1,5-dihydro-3-methyl-4-ethoxycarbonyl-3H-pyr-rol-2-one (28). A solution of 6.12 g (40 mmol) of pyr-role 27 in 100 mL of redistilled pyridine was quickly treated with 15 mL of H2O2 (30%) at rt. After 2 h under refl ux, the mixture was cooled and concentrated. The residue was dissolved in 100 mL of CH2Cl2 and the pH was adjusted with NaOH (1.0 M) to 8. The organic layer was concentrated, dried and recrystallized from CH2Cl2/petrol-ether to give 3.92 g (23.2 mmol, 58%) of pyrroli-none 28 as colorless crystals: mp 148–149 °C. UV-vis (C2H5OH): λmax, nm(ε) 216 (17326), 283 (7148). 1H NMR (200 MHz, CD3Cl): δH, ppm 1.93 (3H, t, 3J = 7.2 Hz, CO2CH2CH3), 2.27 (3H, s, 3-CH3), 4.33 (2H, q, 3J =6.4 Hz, CO2CH2CH3), 4.50 (2H, q, AB-system, 2J =5.8 Hz, 5-H2), 7.88 (1H, br, NH). IR (KBr): ν, cm-1 3450 (s, N–H), 3056 (m, N–H), 2973 (s, C–H), 1732 (s, C=O, ester), 1642 (s, C=O, lactam), 1608 (m, C=C). MS (EI, 70 eV): m/z (%) 169 ([M]+, 54), 154 ([M+ - CH3], 26), 124 ([M+ - OC2H5], 100), 96 ([M+ - CO2C2H5], 78). Anal. calcd. for C8H11NO3: C, 56.78%; H, 6.56%; N, 8.28%; Found: C, 56.63%; H, 6.68%; N, 8.40%.

1,5-dihydro-3-methyl-3H-pyrrol-2-one (10). A solu-tion of 6.77 g (40 mmol) of pyrrolinone 28 in 80 mL of dry methanol was treated with a solution of 5.61 g (100 mmol) of KOH in 20 mL of dry methanol at rt. After 2 h under refl ux, the mixture was cooled and fi ltered. The fi l-trate was extracted with CH2Cl2 (3 × 100 mL). The com-bined organic layers were dried (Na2SO4), fi ltered and concentrated. Recrystallization from CH2Cl2/petrol-etherto give 2.83 g (29.2 mmol, 73%) of pyrrolinone 10 as colorless crystals: mp 63–64 °C. UV-vis (C2H5OH): λmax,nm (ε) 276 (13380). 1H NMR (200 MHz, CD3Cl): δH,

Table 1. Determination of configurations of 18a and 18b via 1H NOE test

Irradiation of a second radio-frequency field Observed data of NOE

Area increment of Compound δ, ppm Position δ, ppm Position absorption peak, %

18a 1.70 5-CH3 1.20 4-CH3 0

3.06 4-CH2 3.4

18b 1.68 5-CH3 1.48 4-CH3 9.8

3.01 4-CH2 0

Table 2. Determination of configurations of 9a and 9b via 1H NOE test

Irradiation of a second radio-frequency field Observed data of NOE

Area increment ofCompound δ, ppm Position δ, ppm Position absorption peak, %

9a 1.70 5-CH3 1.23 4-CH3 0

3.09 4-CH2 4.7

9b 1.66 5-CH3 1.44 4-CH3 11.2

3.04 4-CH2 0



ppm 2.28 (3H, s, 3-CH3), 4.48 (2H, m, 5-H2), 6.45 (1H, m, 4-H), 8.32 (1H, br, NH). IR (KBr): ν, cm-1 3440 (s, N–H), 3180 (s, N–H), 2945 (s, C–H), 1694 (s, C=O, lac-tam), 1602 (m, C=C). MS (EI, 70 eV): m/z (%) 97 ([M]+,49), 68 ([M+ - CH3N], 100). Anal. calcd. for C5H7NO: C, 61.82%; H, 7.27%; N, 14.43%. Found: C, 61.68%; H, 7.19%; N, 14.61%.

4,5-dimethyl-1H-pyrrole-2-carboxylic ethyl ester (35). (a) Preparation of the reagent I: a solution of 13.01 g (100 mmol) of ethyl acetoacetate 29 in 60 mL of gla-cial acetic acid was treated with a solution of 8.16 g (118 mmol) of NaNO2 in 20 mL of water at 5 °C. After 2 h at 0 °C, the obtained solution (I), which contained hydrox-imino-ester 30, was laid aside at rt overnight. (b) Prep-aration of reagent II: a solution of 4.6 g (200 mmol) of natrium in 60 mL of dry ethyl ether was treated with a mixture of 14.8 g (200 mmol) of ethyl formate 32 and 14.4 g (200 mmol) of butanone 33 at 5 °C. After 2 h under refl ux, the mixture was cooled and fi ltered to give a yel-lowish solid (II), which contained aldehyde-ketone 34.A suspension of 13.73 g (210 mmol) of zinc powder in 60 mL of water was treated with the reagents I and II at rt. After 1 h under refl ux, the mixture was poured into 200 g of crushed ice, stirred and fi ltered. The yellow precipi-tate was recrystallized from CH2Cl2/ethylacetate to give 10.03 g (60 mmol, 60%) of dimethylpyrrole 35 as col-orless crystals: mp 111–112 °C. UV-vis (CH3OH): λmax,nm (ε) 221 (10670), 263 (5943). 1H NMR (200 MHz, CD3Cl): δH, ppm 1.25 (3H, t, 3J = 5.8 Hz, CO2CH2CH3),2.17 (3H, s, 4-CH3), 2.31 (3H, s, 5-CH3), 4.37 (2H, q, 3J= 7.0 Hz, CO2CH2CH3), 6.75 (1H, s, =CH), 8.72 (1H, br, NH). IR (KBr): ν, cm-1 3306 (s, N–H), 2983 (s, C–H), 2923 (s, C–H), 1675 (s, C=O, ester), 1595 (m, C=C). MS (EI, 70 eV): m/z (%) 167 ([M]+, 87), 138 ([M+ - C2H5],100), 122 ([M+ - OC2H5], 76), 94 ([M+ - CO2C2H5], 18). Anal. calcd. for C9H13NO2: C, 64.63%; H, 7.84%; N, 8.38%. Found: C, 64.47%; H, 7.78%; N, 8.52%.

4,5-dimethyl-3-bromo-1H-pyrrole-2-carboxylic ethyl ester (36). A solution of 8.36 g (50 mmol) of dimethyl pyrrole 35 in 60 mL of CCl4 was treated with a solution of 8.68 g of Br2 (55 mmol) in 40 mL of CCl4 at rt. After 2 h under refl ux, the mixture was cooled and fi l-tered. After the solvent being removed by rotary evapora-tion, the residue was chromatographed on silica gel (95:5 CH2Cl2/ethylacetate) to afford 10.66 g (43.5 mmol, 87%) of bromide 36 as yellowish crystals: mp 136–137 °C. UV-vis (CH3OH): λmax, nm (ε) 239 (12734), 287 (8152). 1H NMR (200 MHz, CD3Cl): δH, ppm 1.34 (3H, t, 3J =6.8 Hz, CO2CH2CH3), 2.03 (3H, s, 5-CH3), 2.29 (3H, s, 4-CH3), 4.33 (2H, q, 3J = 7.4 Hz, CO2CH2CH3), 10.24 (1H, br, NH). IR (KBr): ν, cm-1 3313 (s, N–H), 2983 (s, C–H), 2917 (s, C–H), 1678 (s, C=O, ester), 1604 (w, C=C). MS (EI, 70 eV): m/z (%) 245 ([M]+, 82), 216 ([M+ - C2H5],14), 200 ([M+ - OC2H5], 100), 172 ([M+ - CO2C2H5], 25), 121 ([M+ - Br - OC2H5], 34), 93 ([M+ - Br - CO2C2H5],91). Anal. calcd. for C9H12NO2Br: C, 44.08%; H, 4.94%; N, 5.72%. Found: C, 44.12%; H, 4.78%; N, 5.81%.

4-methyl-5-formyl-1H-pyrrole-2-carboxylic ethyl-ester (37). To a mixture of 50 mL of CCl4, 45 mL of gla-cial acetic acid and 40 mL of water 3.34 g (20 mmol) of dimethylpyrrole 35 and 44.8 g (80 mmol) of CAN were added at 0 °C. After 1 h at 0 °C, the mixture was poured into 300 mL of water, washed with saturated NaHCO3

solution (100 mL) and then extracted with CH2Cl2 (3 × 100 mL). The combined organic layers were dried (Na2SO4), fi ltered and concentrated. Recrystallization from CH2Cl2/ethylacetate to give 2.93 g (16.2 mmol, 81%) of form-ylpyrrole 37 as colorless crystals: mp 106–107 °C. UV-vis (CH3OH): λmax, nm(ε) 230 (11085), 276 (5267). 1H NMR (200 MHz, CD3Cl): δH, ppm 1.72 (3H, t, 3J = 8.8 Hz, CO2CH2CH3), 2.47 (3H, s, 4-CH3), 4.58 (2H, q, 3J =7.6 Hz, CO2CH2CH3), 6.33 (s, 1H, =CH), 9.38 (1H, br, NH), 10.57 (1H, s, CHO). IR (KBr): ν, cm-1 3442 (s, N–H), 2932 (s, C–H), 2884 (m, C–H), 1725 (s, C=O, aldehyde), 1618 (s, C=O, ester), 1592 (m, C=C). MS (EI, 70 eV): m/z (%) 181 ([M]+, 75), 152 ([M+ - C2H5 or CHO], 88), 136 ([M+ - OC2H5], 100), 108 ([M+ - CO2C2H5], 27), 64 (34). Anal. calcd. for C9H11NO3: C, 59.64%; H, 6.12%; N, 7.73%. Found: C, 59.47%; H, 6.18%; N, 7.85%.

3-bromo-4-methyl-5-formyl-1H-pyrrole-2-carbox-ylic ethyl ester (11). To a mixture of 50 mL of CCl4,45 mL of glacial acetic acid and 40 mL of water were added 4.90 g (20 mmol) of bromide 36 and 44.8 g (80 mmol) of CAN at 0 °C. After 1 h at 0 °C, the mixture was poured into 300 mL of water, washed with saturated NaHCO3 solu-tion (100 mL) and then extracted with CH2Cl2 (3 × 100 mL). The combined organic layers were dried (Na2SO4),fi ltered and concentrated. Recrystallization from CH2Cl2/ethylacetate to give 4.66 g (18.0 mmol, 90%) of bromo-formylpyrrole 11 as colorless crystals: mp 134–135 °C. UV-vis (CH3OH): λmax, nm (ε) 264 (13127), 311 (6735). 1H NMR (200 MHz, CD3Cl): δH, ppm 1.40 (3H, t, 3J =6.8 Hz, CO2CH2CH3), 2.32 (3H, s, 4-CH3), 4.41 (2H, q, 3J = 7.2 Hz, CO2CH2CH3), 9.78 (1H, br, NH), 10.43 (1H, s, CHO). IR (KBr): ν, cm-1 3320 (s, N–H), 3210 (s, N–H), 2983 (s, C–H), 2931 (s, C–H), 1700 (s, C=O, alde-hyde), 1680 (s, C=O, ester), 1615 (w, C=C). MS (EI, 70 eV): m/z (%) 259 ([M]+, 82), 230 ([M+ - C2H5 or CHO], 12), 214 ([M+ - OC2H5], 41), 186 ([M+ - CO2C2H5], 100), 157 (186 - CHO, 8), 107 (186 - Br, 19). Anal. calcd. for C9H10NO3Br: C, 41.70%; H, 3.89%; N, 5.41%. Found: C, 41.52%; H, 3.78%; N, 5.60%.

Synthesis of bromo-formylpyrrole 11 from formyl-pyrrole 37. A solution of 1.81 g (10 mmol) of formylpyr-role 37 in 15 mL of CCl4 was treated with a solution of 1.74 g of Br2 (11 mmol) in 10 mL of CCl4 at rt. After 2 h under refl ux, the mixture was cooled and fi ltered. After the solvent being removed by rotary evapora-tion, the residue was chroma tographed on silica gel (9:1 CH2Cl2/ethylacetate) to afford 1.51 g (5.8 mmol, 58%) of bromo-formylpyrrole 11 as colorless crystals.

5-(2,5-dihydro-4-methyl-5-oxo-1H-pyrrol-2-ylidenmethyl)-3-bromo-4-methyl-1H-pyrrol-2- carboxylic ethyl ester (38). To a solution of 2.33 g


98 B.C. HU ET AL.

(24 mmol) of pyrrolinone 10 and 6.22 g (24 mmol) of bromo-formylpyrrole 11 in 60 mL of dry THF were added 7.48 mL (7.61 g, 50 mmol) of DBU and 3.70 g of molec-ular sieve (3 Å). After 8 h under argon, the mixture was cooled, washed successively with saturated NaH2PO4,NaHCO3 and NaCl solution, and then fi ltered to sepa-rate the molecular sieve. The fi ltrate was extracted with 150 mL of CHCl3. The aqueous layer was extracted again with CHCl3 (3 × 100 mL). The combined organic lay-ers were dried (Na2SO4), fi ltered and concentrated. The brown residue was recrystallized from CHCl3/n-hexane to give 5.32 g (15.7 mmol, 65%) of lactam 38 as fi ne yel-low needle-like crystals: mp 241–242 °C. TLC (neutral aluminum oxide, 7:1 CH2Cl2/CH3COOC2H5): Rf = 0.34. UV-vis (CH3OH): λmax, nm (ε) 251 (13870), 275 (6120), 368 (18860), 404 (19720). 1H NMR (200 MHz, CDCl3):δH, ppm 1.46 (3H, t, 3J = 6.6 Hz, CO2CH2CH3), 1.98 (3H, s, 4-CH3, ring B), 2.29 (3H, s, 4-CH3, ring C), 4.46 (2H, q, 3J = 6.9 Hz, CO2CH2CH3), 8.82 (1H, s, methine-bridge CH), 9.26 (1H, s, 3-H, ring B), 10.71 (1H, s, br, NH, ring C), 11.24 (1H, s, br, NH, ring B). IR (KBr): ν, cm-1 3320 (m, N–H), 3084 (w, N–H), 2980 (s, C–H), 1757 (s, C=O, ester), 1685 (s, C=O, lactam), 1616 (s, C=C). MS (ESI, +c 40 eV): m/z (%) 361 ([M + Na+], 16), 316 ([M - OC2H5

+ Na+], 64), 288 ([M -CO2C2H5 + Na+], 50), 282 ([M - Br + Na+], 37), 252 ([M - Br - 2CH3 + Na+], 25), 186 ([M - CO2C2H5 - Br+], 100), 108 (41). Anal. calcd. for C14H15N2O3Br: C, 49.70%; H, 4.47%; N, 8.29%. found: C, 49.54%; H, 4.56%; N, 8.17%.

5-(2,5-dihydro-4-methyl-5-thioxo-1H-pyrrol-2-ylidenmethyl)-3-bromo-4-methyl-1H-pyrrol-2- carboxylic ethyl ester (39). (a) Preparation of the reagent P4S10/NaHCO3: to 40 mL of dry THF were added 1.34 g (3 mmol) of P4S10 and 3.02 g (36 mmol) of NaHCO3 at rt. The mixture was stirred at rt for 30 min to give a yel-low solution. (b) Sulfi dation process: to a solution of 2.03 g (6 mmol) of lactam 38 in 30 mL of dry THF the reagent P4S10/NaHCO3 prepared in (a) was added under argon and vigorous stirring. After 45 min under refl ux and 1 h at rt, the mixture was transferred to 150 mL of CH2Cl2. The aqueous layer was extracted again with CH2Cl2 (3 × 100 mL). The combined organic layers were dried (Na2SO4), fi ltered and concentrated. Column chro-matography (silica gel, 10:1 CH2Cl2/ethyl acetate) gave an orange-yellow eluate which contained the product. The eluate was concentrated and recrystallized from CH2Cl2/n-hexane to give 1.73 g (4.89 mmol, 82%) of thiolactam 39 as orange crystals: mp 214–215 °C. UV-vis (C2H5OH): λmax,nm (ε) 285 (15575), 290 (14210), 446 (19180). 1H NMR(200 MHz, CDCl3): δH, ppm 1.51 (3H, t, 3J = 6.9 Hz, CO2CH2CH3), 1.97 (3H, s, 4-CH3, ring B), 2.36 (3H, s, 4-CH3, ring C), 4.56 (2H, q, 3J = 7.2 Hz, CO2CH2CH3), 8.91 (1H, s, methine-bridge CH), 9.37 (1H, s, 3-H, ring B), 10.54 (1H, s, br, NH, ring C), 11.37 (1H, s, br, NH, ring B). IR (KBr): ν, cm-1 3410 (m, N–H), 2982 (s, C–H), 2905 (m, C–H), 1724 (s, C=O, ester), 1603 (s, C=C), 1228 (s, C=S, thiolactam). MS (ESI, +c 40 eV): m/z (%) 377 ([M + Na+],

13), 345 ([M - S + Na+], 27), 332 ([M - OC2H5 + Na+], 64), 304 ([M - CO2C2H5 + Na+], 53), 298 ([M - Br + Na+], 36), 236 ([M - Br - S - 2CH3 + Na+], 19), 202 ([M - CO2C2H5

-Br+], 100), 124 (28). Anal. calcd. for C14H15N2O2SBr: C, 47.46%; H, 4.27%; N, 7.91%; S, 9.03%. found: C, 47.65%; H, 4.39%; N, 7.76%; S, 9.18%.

(5-cyano-2,3,4,5-tetrahydro-4,5-dimethyl-4-meth-oxycarbonylmethyl-1H-pyrrol-2-ylidene) acetate t-butyl ester (43). A suspension of 1.00 g (4.42 mmol) of thi-olactam 9a and 2.44 g (17.65 mmol) of powdery K2CO3

in 50 mL of dry acetonitrile was treated under argon with 1.18 g (4.87 mmol) of dibenzoyl peroxide at 0 °C for 15 min. After 30 min under argon at 0 °C, 1.06 g (5.29 mmol) of malonic ester 12 were injected. The mixture was continuously stirred at 0 °C for 30 min and at rt for 4 h, washed with saturated NaHCO3 solution (100 mL) and then extracted with 100 mL of CH2Cl2. The aqueous layer was extracted again with CH2Cl2 (3 × 80 mL). The combined organic layers were dried (Na2SO4), fi ltered and concentrated to give thioiminoester 41 as a yellow oily liquid. Under argon the thioiminoester 41 was dis-solved in 20 mL of freshly redistilled P(OC2H5)3. After being stirred for 8 h at 80 °C, the solvent was removed by rotary evaporation. The remained yellow oil, namely the enamine 42, was dissolved in 20 mL of dry THF. To this solution under argon were added 2 mL of freshly redistilled piperidine and 0.2 g of Pd(PPh3)4. After 2 h at rt, the mixture was neutralized with 60 mL of 2 N HCl and extracted with 100 mL of CH2Cl2. The aqueous layer was extracted again with CH2Cl2 (3 × 80 mL). The combined organic layers were dried (Na2SO4), fi ltered and concentrated. The residue was purifi ed by column chromatography (silica gel/neutral aluminum oxide, 9:1 CH2Cl2/petrol-ether, Rf = 0.30). The obtained yellowish product-contained fraction was purifi ed again by column chromatography (silica gel, 9:1 petrol-ether/ethyl acetate, Rf = 0.41) to afforded a colorless oil, whose purity was supervised via HPLC. After being laid aside for one week at -18 °C, the oil was crystallized to give 0.79 g (2.56 mmol, 58%) of enamine 43 as colorless crystals: mp 61–62 °C. HPLC (polygosil 60–10, 9:1 petrol-ether/ethyl acetate, 1.5 mL/min, UV 270 nm, tR = 3.8 min). UV-vis (C2H5OH): λmax, nm(ε) 270 (24360). 1H NMR (200 MHz, CDCl3): δH, ppm 1.12 (3H, s, 4-CH3), 1.46 (9H, s, t-bu-tyl), 1.53 (3H, s, 5-CH3), 2.27 (1H, d, AB-system, A-part, 2J = 7.4 Hz, 3-CH2), 2.58 (1H, d, AB-system, A-part, 2J = 7.4 Hz, 3-CH2), 2.72 (1H, d, AB-system, A-part, 2J = 5.6Hz, O=C-CH2), 3.04 (1H, d, AB-system, B-part, 2J = 5.6Hz, O=C-CH2), 3.72 (3H, s, O-CH3), 4.62 (1H, s, =CH), 7.67 (1H, s, br, NH). IR (KBr): ν, cm-1 3360 (m, N–H), 3026 (w, N–H), 2972 (s, C–H), 2230 (m, C≡N), 1678 (s, C=O), 1620 (s, C=C). MS (70 eV): m/z (%) 309 ([M + H+], 5), 308 ([M]+, 29), 251 ([M+ - C4H9], 48), 236 ([M+

- C4H9 - CH3], 25), 220 ([M+ - C4H9 - OCH3], 100), 148 ([M+ - CO2C4H9 - CO2CH3], 26), 121 (9). Anal. calcd. for C16H24N2O4: C, 62.32%; H, 7.84%; N, 9.08%. Found: C, 62.47%; H, 8.03%; N, 8.92%.



5-{5 - [ ( t -buty loxycarbony l ) (5 -Cyano-2 ,3 ,4,5-tetrahydro-4,5-dimethyl-4-methoxycarbonylm-ethyl-1H-pyrrol-2-ylidene)methylthio]-4-methyl-2H-pyrrol-2-ylidenmethyl}-3-bromo-4-methyl-1H-pyr-rol-2-carboxylic ethyl ester (46). A solution of 0.65 g (2.1 mmol) of enamine 43 in 15 mL of dry CH2Cl2 was treated in the dark with 0.38 g (2.2 mmol) of N-bromo-succinimide under argon at 0 °C. After 20 min at 0 °C and 2 h at rt, the mixture was concentrated and dried to produce a purplish brown raw bromide, which was imme-diately and without purifi cation dissolved in 15 mL of dry acetonitrile. To this solution were under argon added 1.25 mL (1.28 g, 8.4 mmol) of DBU and 0.71 g (2.0 mmol) of thiolactam 39. After 2 h at rt, the mixture was transferred to 150 mL of saturated aqueous NaHCO3 solution and extracted with 100 mL of CH2Cl2. The aqueous layer was extracted again with CH2Cl2 (3 × 80 mL). The combined organic layers were dried (Na2SO4), fi ltered and concen-trated. Purifi cation by column chromatography (neutral aluminum oxide, 3:1 CH2Cl2/petrol-ether, Rf = 0.35) afforded 1.12 g (1.70 mmol, 85%) of sulfi de 46 as yellow crystals: mp 184–185 °C. UV-vis (CH3OH): λmax, nm (ε)270 (27854), 436 (18012). 1H NMR (200 MHz, CDCl3):δH, ppm 1.05, 1.55 (6H, 2s, 2CH3, ring A), 1.39 (9H, s, t-butyl), 1.47 (3H, t, 3J = 6.7 Hz, CO2CH2CH3), 1.92 (3H, s, CH3, ring B), 2.28 (3H, s, CH3, ring C), 2.47, 2.82 (2H, AB-system, J = 9.3 Hz, CH2, ring A), 2.90, 3.17 (2H, AB-system, J = 10.8 Hz, -H2CCOO), 3.78 (3H, s, CH3,CO2CH3), 4.53 (2H, q, J = 7.1 Hz, CO2CH2CH3), 8.85 (1H, s, methine-bridge CH), 9.30 (1H, s, =CH, ring B), 10.26, 11.71 (2H, 2s, br, NH). IR (KBr): ν, cm-1 3298 (m, N–H), 3043 (w, N–H), 2980 (s, C–H), 2218 (m, C≡N),1768 (s, C=O), 1688 (m), 1610 (m, C=C), 1583 (s). MS (ESI, +c 30 eV): m/z (%) 683 ([M + Na+], 17), 626 ([M - C4H9 + Na+], 34), 610 ([M - CO2C2H5 + Na+], 9), 604 ([M – Br + Na+], 17), 599 ([M -C4H9– HCN + Na+], 26), 553 ([M - C4H9 - CO2C2H5 + Na+], 19), 547 ([M - C4H9 – Br + Na+], 13), 451 ([M - C4H9 -CO2C2H5 - Br+], 100), 436 (451 - CH3, 31), 424 (451 - HCN, 14), 374 (451 - CO2 -SH, 31). Anal. calcd. for C30H37N4O6SBr: C, 54.53%; H, 5.65%; N, 8.48%; S, 4.84%. Found: C, 54.68%; H, 5.60%; N, 8.37%; S, 4.93%.

5-{5-[(5-cyano-2,3,4,5-tetrahydro-4,5-dimethyl-4-methoxycarbonylmethyl-1H-pyrrol-2-ylidene)methyl]-4-methyl-2H-pyrrol-2-ylidenmethyl}-3-bromo-4-methyl-1H-pyrrol-2-carboxylic ethyl ester (7). To a mixture of 80 mL of dry CH2Cl2 and 6 mL of CF3COOH under argon were added 1.00 g (1.51 mmol) of sulfi de 46 and 2.34 g (12.13 mmol) of P(CH2CH2CN)3.After 3 h under refl ux, the mixture was cooled, carefully transferred to 300 mL of ice-cold saturated NaHCO3

solution (CO2-giving off) and then extracted with 100 mL of CH2Cl2. The aqueous layer was extracted again with CH2Cl2 (3 × 80 mL). The combined organic layers were dried (Na2SO4), fi ltered and concentrated. The residue was purifi ed by column chromatography (silica gel/neutral aluminum oxide, 2:1 CH2Cl2/petro-ether, Rf = 0.42) to

give three fractions. Firstly the orange product fraction, secondly a middle fraction composed of the product and P(CH2CH2N)3 and fi nally the pure colorless phosphorous compound were eluted. The middle fraction was puri-fi ed again by column chromatography (neutral alumi-num oxide, 4:1 petrol-ether/ethyl acetate, Rf = 0.33) to elute the product. The combined product fractions were recrystallized from CH2Cl2/n-hexane to give 0.383 g (0.73 mmol, 48%) of the tricycle 7 as fi ne orange-red needle-like crystals: mp 135–136 °C. UV-vis (C2H5OH):λmax, nm (ε) 238 (13082), 290 (27964), 447 (25332), 473 (18650). 1H NMR (200 MHz, CDCl3): δH, ppm 1.12, 1.65 (6H, 2s, 2CH3, ring A), 1.49 (3H, t, J = 7.2 Hz, CH3,ethylester), 1.90 (3H, s, CH3, ring B), 2.32 (3H, s, CH3,ring C), 2.52, 2.84 (2H, AB-system, J = 8.7 Hz, CH2,ring A), 2.89, 3.20 (2H, AB-system, J = 10.5 Hz, H2C-COO-), 3.77 (3H, s, CH3, methylester), 4.55 (2H, q, J = 7.5 Hz, CH2, ethylester), 8.51 (1H, 1s, methine-bridge CH between ring A and ring B), 8.93 (1H, 1s, methine-bridge CH between ring B and ring C), 9.32 (1H, s, =CH, ring B), 10.20–11.93 (2H, m, very broad, NH). IR (KBr):ν, cm-1 3440 (w, sharp, N–H), 3360 (s, br, N–H), 2972 (s, C–H), 2210 (w, C≡N), 1754 (s, C=O), 1600 (s, C=C). MS (ESI, +c 25 eV): m/z (%) 551 ([M + Na+], 15), 520 ([M - OCH3 + Na+], 14), 506 ([M -OC2H5 + Na+], 8), 478 ([M - CO2C2H5 + Na+], 23), 472 ([M - Br + Na+], 31), 376 ([M - CO2C2H5 - Br+], 100), 361 (376 - CH3, 17), 349 (376 - HCN, 23). Anal. calcd. for C25H29N4O4Br: C, 56.80%; H, 5.53%; N, 10.61%. Found: C, 56.69%; H, 5.60%; N, 10.52%.

CONCLUSION

In this paper, we describe the successful construc-tion of the A-B-C-tricycle skeleton 7 which incorporates most of the appropriate functionality for conversion to the structural model of tolyporphin 3. The three precur-sors to this skeleton were synthesized in a concise and stereocontrolled manner and were then connected via a two-time Eschenmoser sulfi de contraction. Although the fi nal tolyporphin compound has not been gained, very signifi cant progress has been made toward this goal, and the feasibility of the approach has clearly been demon-strated. With the important intermediate 7 at hand, we feel in a good position to complete total syntheses of 3and other tolyporphins.

REFERENCES

1. Prinsep MR, Caplan FR, Moore RE, Patterson GML and Smith CD. J. Am. Chem. Soc. 1992; 114:385–387.

2. Morliere P, Maziere JC, Santus R, Smith CD, Prinsep MR, Stobbe CC, Fenning MC, Gol-berg JL and Chapman JD. Cancer Res. 1998; 58:3571–3578.


100 B.C. HU ET AL.

3. Prinsep MR, Patterson GML, Larsen LK and Smith CD. Tetrahedron 1995; 51: 10523–10530.

4. Prinsep MR, Patterson GML, Larsen LK and Smith CD. J. Nat. Prod. 1998; 61: 1133–1136.

5. Minehan TG and Kishi Y. Tetrahedron Lett. 1997; 38: 6811–6814.

6. Wang WG and Kishi Y. Org. Lett. 1999; 1:1129–1132.

7. Minehan TG and Kishi Y. Angew. Chem. 1999; 111:975–977.

8. Hu BC, Lu CX and Liu ZL. Chin. J. Org. Chem.2004; 24: 270–280.

9. Montforts FP. Angew Chem, 1981; 93: 795–796. 10. Roth M, Dubs P, Goetschi E and Eschenmoser A.

Helv. Chim. Acta 1971; 54: 710–713. 11. Montforts FP and Schwartz UM. Liebigs Ann.

Chem. 1985; 1228–1253. 12. Eschenmoser A and Winter CE. Science 1977; 196:

1410–1420. 13. Bahrs J. Ein homochiraler Baustein zum Aufbau

von Bonellin und Chlorophylla aus optisch aktiven

Campher, PhD Thesis: Bremen University, Bremen, 1992.

14. Kirchner U. Darstellungen und nucleophile Sub-stitutionsreaketionen von Aza-Allylsystemen, PhD Thesis: Bremen University, Bremen, 1993.

15. Scheeren JW, Ooms PHJ and Nivard RJ. Synthesis1973; 149–155.

16. Hu BC and Lu CX. Chin. J. Appl. Chem. 2004; 21:1165–1169.

17. Cai CJ, Hu BC and Lu CX. Chin. J. Org. Chem.2005; 25: 1311–1317.

18. Hudlicky M. Oxidation in Organic Chemistry,American Chemical Society: Washington DC, 1990; 112.

19. Fischli A and Eschenmoser A. Angew. Chem. 1967; 6: 866–868.

20. Goetschi E, Hunkeler W, Wild HJ, Schneider P, Fuhrer W, Gleason J and Eschenmoser A. Angew. Chem. 1973; 12: 910–912.

21. Ofner S, Rasetti V, Zehnder B and Eschenmoser A. Helv. Chim. Acta. 1981; 64: 1431–1433.


DOI: 10.1142/S1088424610001775



INTRODUCTION

Surface modifi cation by self-assembly of sulfur deriv-atives onto gold has been massively employed in the last decade in a wide range of applications, including chemi-cal and biochemical sensors [1], optoelectronic devices [2, 3] and electrocatalysis [4]. The main motivation for using self-assembled monolayers (SAMs) for the attachment of specifi c molecules on the surface, is their well-known stability, ease of preparation, organiza-tion and molecular control. With the view of designing electrocatalytic systems based on SAMs, the aim of the study presented here is to be able to transfer the recog-nized catalytic properties of individual molecules, such as porphyrins, to the electrode surface in an organized

monolayer array. In this sense, efforts must be employed to synthesize and investigate new molecules bearing appropriate groups for an effi cient one-step linkage to the surface. A great number of investigations have been reported on SAMs of porphyrins with distinct tail and anchor groups [5–9], most of them employing as macro-cycles tetraphenylporphyrin derivatives and a thiol or a disulfi de as anchor groups. The catalytic activity of the immobilized porphyrins towards the oxygen reduc-tion reaction with the formation of hydrogen perox-ide or directly to water, has been discused based on the different metal centres [10] and also on the differ-ent macrocycle orientations [4]. In previous studies we demonstrated that specially synthesized Co [11] and Fe [12] porphyrin containing one disulfi de binding group could be successfully immobilized on gold, preserving their catalytic activity. One novelty of the work was to address the formation of self-assembled monolayers using a different class of porphyrins than tetraphenylpor-phyrin derivatives, structurally more related to the hemes

Synthesis and self-assembly of a novel cobalt(II) porphyrin lipoic acid derivative on gold

Christoph S. Eberlea, Ana S. Vianab, Franz-Peter Montfortsa and Luisa Maria Abrantes*b

a Universität Bremen, Institut für Organische Chemie, Leobener Str. NW2 C, 28359 Bremen, Germanyb CQB, Departamento de Quimica e Bioquimica, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1794-016 Lisboa, Portugal

Received 10 February 2009Accepted 4 May 2009

ABSTRACT: A novel Co(II) porphyrin lipoic acid derivative was synthesized starting from the commerically available red blood pigment hemin. The disulfi de functionalities of the lipoic acid moieties allowed its immobilization on gold by a self-assembly method. The Co(II) porphyrin self-assembled monolayer (SAMs) on gold (111) surfaces were characterized electrochemically through monolayer reductive desorption and evaluation of the redox properties of the immobilized molecules in organic medium, and by scanning tunneling microscopy (STM). It was found that after assembly the Co(II) porphyrin is electroactive exhibiting the typical redox processes observed for its precursor without the appended lipoic acid in solution. A coverage of 2.7 × 10-10 mol.cm-2 has been estimated assuming that four electrons (one per each sulfur atom) are involved in the process. The porphyrin-modifi ed gold electrodes exhibit catalytic acitivity demonstrated towards the reduction of molecular oxygen in acidic solution.

KEYWORDS: Co(II) porphyrin lipoic acid derivative, self-assembled monolayers, electrochemistry, scanning tunneling microscopy, electrocatalytic activity.


*Correspondence to: Luisa Maria Abrantes, email: [email protected], tel: +351 21-7500890, fax: +351 21-7500088


102 C.S. EBERLE ET AL.

and recalling natural systems, also displaying a metal- centred catalytic performance. In the current work the synthesis and characterization of a self-assembled mono-layer formed from a novel Co(II) porphyrin, with dis-tinct anchor groups, also based on biomimetic porphyrin derivatives, onto Au (111) electrodes is investigated. For this purpose the Co(II) porphyrin 5 (Scheme 1) with two covalently linked lipoic acid moieties was synthesized. The two dithiolane groups per porphyrin unit should allow a very strong binding to the gold surface by four sulfur gold linkages. Surface attachment of the metallo-porphyrin has been studied by cyclic voltammetry through the electrochemical behaviour of the modifi ed electrodes in organic solution and by electrochemical reductive desorption. The electrocatalytic properties of the modifi ed electrodes have been tested for the oxygen reduction reaction. Scanning tunneling microscopy has been used for the morphological characterization of the functionalized gold surface with the Co(II) porphyrin lipoic acid derivative.


Synthesis of Co(II) porphyrin lipoic acid ester derivative

The deuteroporphyrin bis-alcohol 3 [13] was prepared in few reaction steps according to literature procedures

from the red blood pigment hemin 1 (Scheme 1). The cobalt insertion [Co(II) porphyrin 4] followed standard procedures for porphyrin complexation using Co(II) acetylacetonate as complexing reagent. The alcohol functions of the Co(II) porphyrin were subsequently esterifi ed with (R)-lipoic acid. The enantiomerically enriched (R)-lipoic acid was obtained by enzymatic resolution of a racemic mixture of the lipoic acids [14]. Enantiomerically pure (enriched) lipoic acid was cho-sen to achieve a higher order of SAMs on the metal surface. For the esterifi cation step the lipoic acid was activated by n-propyl phosphonic acid anhydride, which leads to the desired Co(II) porphyrin lipoic acid deriva-tive 5. The activation reagent has the advantage that it could be removed easily from the reaction mixture by aqueous extraction, which facilitated the whole work procedure.

Electrochemical characterization of self-assembled monolayers

The successful immobilization of metalloporphyrin lipoic acid derivative 5 was primarily confi rmed through electrochemical monolayer reductive desorption (Fig. 1). The fi rst cycle reveals a sharp peak at -1.100 V (Table 1) attributed to the desorption of the dithiolane groups from the gold substrate, similar to the observation for alkyl-thiols on gold [15]. The desorption process occurs at a potential value of approximately -0.210 V more negative

Scheme 1. Synthesis of Co(II) deuteroporphyrin bis-lipoic acid ester 5. Reagents and conditions: (a) Co(II) acetylacetonate, THF, refl ux, 24 h, 65%. (b) (R)-lipoic acid, n-propylphosphonic acid anhydride (PPA), DMAP, NEt3, THF, 0 °C to rt, 12 h, 52%


NOVEL COBALT(II) PORPHYRIN LIPOIC ACID DERIVATIVE ON GOLD 103

than the one observed for Co(II) porphyrin disulfi de derivative 6, previously investigated by us [11]. The signifi cant difference in the electrochemical desorp-tion is fully reproducible, clearly confi rming a stronger linkage of the molecule to the gold surface, once it can be attached via four sulfur atoms, or/and second show-ing a better molecular packing arising from stronger chain-chain interaction in compound 5 compared to the slightly shorter disulfi de derivative [11]. This observa-tion discloses the high surface stability for the novel Co(II) porphyrin SAM under study. As shown in Table 1, a coverage of 2.7 × 10-10 mol.cm-2 was estimated from the integration of the cathodic peak observed in the fi rst cyclic voltammogram (Fig. 1), assuming that four electrons (one per each sulfur atom) are involved in the molecule desorption from gold. In spite of the different

desorption potential, similar surface coverages (Γ) could be achieved for the two Co(II) porphyrin derivatives. According to the estimated coverage, each Co(II) por-phyrin derivative claims an average area of 61 Å2, which is closer to the predicted value for molecules displaying a perpendicular orientation, relative to the surface plane (30 Å2), rather than a parallel position (170 Å2), indicating a relatively packed and oriented monolayer. As shown in Fig. 1, the second cycle shows a less intense and nega-tive reduction peak, at -1.015 V, as a result of a partially re-adsorption of the dithiolane groups on the reverse sweep of the fi rst cycle, a behaviour typically observed for SAMs of thiols on gold [16].

The redox properties of the modifi ed electrode in organic medium were investigated in order to check whether the surface immobilization affects the electro-chemical response of the catalytic centre and compared to the electrochemical behavior of its Co(II) precursor in solution (Scheme 1, compound 4), as illustrated in Figs 2 and 3, respectively. The ill-defi ned peaks obtained for the modifi ed gold electrode are due to the extremely low concentration (2.3 × 10-10 mol.cm-2) of Co(II) deuteropor-phyrin (DP) moiety on the surface inherent to the method of self-assembly. Notwithstanding, one can detect some of the typical processes of synthetic metalloporphy-rin and of the precursor 4; Table 2 shows the potential values for the oxidation and reduction peaks revealed in Figs 2 and 3. The assignment of the peaks was per-formed in agreement to previous studies involving the electrochemical characterization of synthetic metal-loporphyrins [17, 18]. It can be observed that both the porphyrin ring and the cobalt centre are still electrochem-ically active upon immobilization through the dithiolane functionalities.

Fig. 1. Cyclic voltammetry of gold (111) modifi ed with Co(II) porphyrin lipoic acid derivative in 0.1 M NaOH, ν = 20 mV.s-1

Table 1. Comparision of the electrochemical desorption data of immobilized metalloporphyrin derivatives

Modified electrode Epred, V n e-/molecule Γ, mol.cm-2

Au/Co(II) porphyrin lipoic acid derivative (5) -1.10 4 2.4 × 10-10

Au/Co(II) porphyrin disulfide derivative (6) -0.89 2 2.5 × 10-10

Fig. 2. Cyclic voltammetry of modifi ed gold (111) with Co(II) porphyrin lipoic acid derivative in 0.1 M tetrabutylammonium perchlorate/propylene carbonate solution; (a) ν = 10 mV.s-1; (b) ν = 50 mV.s-1



Scanning Tunnelling Microscopy

STM has been uttered in the morphological charac-terization of the modifi ed electrode. Figure 4 depicts the images of the modifi ed electrode (b and c) in com-parison with a bare gold substrate (a). After a fl ame-annealing procedure the substrate exhibits terraces with a vertical distance of 0.24 nm, which is characteristic for gold (111) atomic steps. On the modifi ed electrode, in spite of the clearly visible monoatomic steps, there is a marked increase in the surface roughness with the appearance of pits, one gold atom dip, typically found as fi ngerprint for sulfur adsorption on gold [11], thus confi rming the immobilization of the porphyrin deriva-tive via the dithiolane moieties. The larger size of the terminal porphyrin groups in this SAM compared

to pure alkylthiol monolayers must be the reason for the absence of large ordered domains with the SAM, which are absolutely required for the observation of molecular resolution. Notwithstanding, a smaller scan of the functionalized gold surface (as shown in Fig. 4c) depicts some bright spots, which according to the size might well correspond to individual porphyrin molecules. Due to the high noise levels observed in the images, it is not possible to infer directly the organiza-tion and packing of the molecules.

Electrocatalysis

As it is well known, the metalloporphyrins are rec-ognized as catalytic centres for a number of redox reac-tions [19], including the reduction of molecular oxygen

Fig. 4. Scanning Tunnelling Microscopy images of bare gold (a) and after being modifi ed with the lipoic acid Co-porphyrin deriva-tive (b) and (c); z = 1–2 nm in all images. The inset in fi gure (a) corresponds to the cross section signed on the image

Fig. 3. Cyclic voltammetry of 0.5 mM Co(II) porphyrin precursor 4 on bare gold (111) in 0.1 M tetrabutylammonium perchlorate/propylene carbonate solution; ν = 50 mV.s-1

Table 2. Comparison of the electrochemical processes of the Co(II) porphyrin precursor 4 and the immobilized Co(II) porphyrin derivative 5 on gold (111)

Electrochemical process I DP/DP•- II Co(II)/Co(III) III DP/DP•+ IV DP/DP2+

Co(II) porphyrin precursor (4) E½I = -1.01 V Ep

ox,II = 0.16 V Epox,III = 0.38 V E½

IV = 0.98 V

Epred,II = -0.03 V

Modified gold with Co(II) porphyrin (5) Epred,I = -0.97 V Ep

red,II = -0.09 V Epox,III = 0.33 V Ep

ox,IV = 1.06 V



[4, 10, 11, 20] directly to water or partially to hydrogen peroxide. In order to ensure that the immobilized Co(II) porphyrin was still reactive, a preliminary study has been performed in a 0.5 M H2SO4 aerated solution, towards the reduction of molecular oxygen (Fig. 5). After surface modifi cation a clear reduction peak can be detected at -0.035 V, which is attributed to the molecular oxygen reduction. In a deoxygenated solution this process is not detected. A bare gold electrode does not show any defi ned peak in the same potential window, and the large cathodic current observed in Fig. 5 is most probably due to the hydrogen evolution reaction, possibly with some contribution of oxygen reduction. It is worth noting that it is not surprising that on the modifi ed gold electrode the hydrogen evolution reaction is shifted towards more negative potentials due to the presence of a relatively well-packed porphyrin monolayer.

Taking into account the results reported before for a similar Co(II) porphyrin it should be expected that hydro-gen peroxide is being formed in the reaction. However a more detailed electrochemical study must be undertaken in order to unravel the mechanism and number of elec-trons involved in the oxygen reduction reaction. Work is now in progress to study the redox properties and electro-catalytic ability of various metal-centered porphyrins in monolayers onto Au (111).

EXPERIMENTAL

General remarks

Starting materials were either prepared according to literature procedures or were purchased from Fluka, Merck, Acros Organics or Sigma Aldrich and used with-out further purifi cation. Tetrabutylammonium perchlo-rate (TBAP), sodium hydroxide, and absolute ethanol had analytical grade quality and were used as received. Propylene carbonate was distilled in vacuo prior to use.

Ultra pure water was obtained from a purifi cation sys-tem Sartorious, nominal resistivity 18.2 MΩcm at 25 °C. Deuteroporphyrin-(IX)-dimethylester 2 was prepared as reported in reference [19]. All reactions were carried out under argon. All solvents were purifi ed and dried by standard methods. 1H NMR spectra: Bruker DPX-200 spectrometer; AVANCE NB-360 MHz and DPX-200 ADVANCE (υ: 1H = 200 MHz); all chemical shifts were referenced to TMS lock signal (δH = 0). MS and HR-MS: Finnigan MAT 8200 spectrometer (EI (70 eV) and ESI). IR: Perkin-Elmer Paragon 500 FT-IR spectrometer. UV-vis: Varian Cary 50 spectrophotometer. Optical activity: Perkin-Elmer polarimeter 243. Column chromatographic separations were performed on silica gel (32–63 μm, 60 Å, ICN) or aluminum oxide (activity grade II–III, neutral, ICN).

Substrates. 200 nm of gold evaporated on glass with a pre-layer of 2–4 nm of chromium (Gold Arran-dee, Gmbh) were cleaned in “piranha” solution for few minutes, thourougly rinsed with ethanol and water and fl ame-annealed in an oxygen-gas fl ame to produce a fl at surface with a predominant (111) crystallographic ori-entation, as confi rmed by STM. An average value for roughness factor of the gold substrates of 1.2 has been estimated by the iodine chemisorption method [21] and is in conformity with the values reported by other authors [22, 23] for similar thin-layer gold electrodes.

Electrode modifi cation. The gold substrates were immersed in a 1 mmol dm-3 ethanol solution of the Co(II) porphyrin derivative for ~20 h at RT. After self-assembly, the electrodes were rinsed with copious amounts of etha-nol and water.

Electrochemistry. All glassware was cleaned using a chromic acid solution followed by thorough rinsing with purifi ed water. Cyclic voltammetry was performed using a PARSTAT 2263 (Perkin Elmer). A one compart-ment Tefl on cell, fi tted with a Pt foil counter electrode and a saturated calomel reference electrode (SCE) was used. The Au slides were clamped against an O-ring, which defi ned the geometric area of the working electrode as 0.57 cm2. The electrolyte solutions were degassed 1 hour with nitrogen (99.9999%) prior to each experiment.

Scanning Tunnelling Microscopy. The measure-ments were carried out with a Nanoscope IIIa Multimode Microscope (Digital Instruments, Veeco) fi tted with a STM head, at room temperature and in air. Pt/Ir STM tips were mechanically cut before each experiment.

Synthesis

Preparation of [13,17-bis(3-hydroxypropyl)-2,7,12,18-tetramethyl-21H,23H-porphyrinato]-cobalt(II) 4. 13,17-bis(3-hydroxypropyl)-2,7,12,18-tetramethylporphyrin 3 [14] (299 mg, 0.62 mmol) and Co(II) acetylacetonate (957 mg, 3.72 mmol, 6 equiv.) in 80 mL THF were refl uxed with exclusion of light under argon for 24 h. After evaporation

Fig. 5. Cyclic voltammogram of bare gold and gold modifi ed with Co(II) porphyrin lipoic acid derivative in aerated and deoxygenated 0.5 M H2SO4 at 20 mV.s-1



of the reaction mixture it was partitioned between dichloro-methane and brine. The aqueous phase was extracted again with dichloromethane and the combined organic extracts were fi ltered over dry cotton wool. After evapo-ration of the solvents the residue was chromatographed over aluminium oxide (activity grade II–III, neutral, ICN) with methanol. After removal of the eluent and drying the residue in vacuo using an oil pump pure Co(II) porphyrin bis-alcohol 4 was obtained. Yield: 216 mg (0.40 mmol, 65%), mp > 300 °C, TLC (aluminium oxide, methanol) Rf

= 0.93. 1H NMR (CDCl3): δ, ppm 2.30 (m, 4 H, J = 6.60 Hz, 13′′-, 17′′-H), 3.42, 3.46, 3.55, 3.58 (4s, 12 H, 2-, 7-, 12-, 18-CH3), 3.74 (t, 4H, J = 5.80 Hz, 13′-, 17′-H), 4.08 (dt, 6H, J = 5.79 Hz, 13′′′-, 17′′′-H, 2 OH), 9.12, 9.15, 9.91, 9.96, 10.01, 10.48 (6 s, 6 H, 3-, 5-, 8-, 10-, 15-, 20-H). IR (KBr): ν, cm-1 3422, 2918, 2841, 1629, 1575, 1557, 1507, 1457, 1435, 1383, 1248, 1122, 1036, 986, 844, 795. UV-vis (CH2Cl2): λmax, nm (log ε, M-1.cm-1) 392 (5.16), 416 (5.28), 527 (4.39), 554 (4.36). MS (EI, 70 eV, direct inlet): m/z 539 (25) [M]+, 494 (15), 449 (17) and other fragments. HRMS (EI, C30H32N4O2

58Co): calcd. 538.18572; found 538.18576.Preparation of {13,17-bis{3-(5′-[(3′′R)-(1,2-dithiolan-

3′′-yl]-pentanoyl)-oxy-propanyl}-2,7,12,18-tetramethyl-21H,23H-porphyrinato}-cobalt(II) 5. Cobalt(II) deuter-oporphyrin bis-alcohol 4 (25 mg, 46 μmol), (3′R)-5-(1′,2′-dithiolan-3′-yl)pentanoic acid [(R)-lipoic acid] 1 (126 mg, 0.61 mmol) and dimethylaminopyridine (DMAP) (13 mg, 0.11 mmol) were dissolved in 30 mL dry tetrahydrofurane. After 10 min stirring at room temperature under argon, 3 mL of dry triethylamine were added to the reaction mix-ture at 0 °C. Then 1 mL of a solution of 50% n-propyl phosphonic acid anhydride (PPA) in ethyl acetate was slowly added through a syringe and the reaction mixture was fi rst stirred for further 45 min at 0 °C then overnight at room temperature under argon. To this mixture was added 50 mL of dichloromethane and the mixture was extracted with H2O and NaHCO3. All organic phases were collected and fi ltered through cotton wool. After removal of the sol-vent in an rotatory evaporator, the crude product was chro-matographed on silica gel (32–63 μm, 60 Å, ICN) eluting with CH2Cl2/MeOH/n-hexane (19:1:1). Evaporation of the eluents and drying of the product fraction in vacuo of an oil pump gave an dark red oil. Isothermic recrystallization from CH2Cl2/n-pentane gave the product as reddish brown crystals. Yield: 22 mg (24 μmol, 52%), mp > 164.2 °C, TLC (silica gel, CH2Cl2/MeOH/n-hexane (19:1:1)) Rf = 0.63. 1HNMR (CDCl3/pyridine-d5): δ, ppm 1.0–2.3 and 2.8–3.0 (m, 30H, lipoic acid part, 13′-, 13′′-, 17′-, 17′′-H), 2.50 (s, 4H, lipoic acid part) 3.51, 3.52, 3.54, 3.60 (4s, 12H, 2-, 7-, 12-, 18-CH3), 4.13 (s, 4H, 13′′′-, 17′′′-H), 9.24, 9.27, 9.99, 10.05, 10.09, 10.12 (6s, 6H, 3-, 5-, 8-, 10-, 15-, 20-H). IR (KBr): ν, cm-1 2922, 2856, 1777, 1634, 1505, 1436, 1383, 1228, 1126, 1096, 1062, 912, 801, 737. [α]D

20 +3.9 (c = 0.002 in CH2Cl2). UV-vis (CH2Cl2): λmax, nm (log ε, M-1.cm-1) 396 (5.32), 404 (5.48), 470 (4.41), 532 (4.56), 570 (4.51). MS (ESI, CH2Cl2/MeOH(1:10)): m/z positive mode 915 [M + H]+.

CONCLUSION

The successful immobilization of Co(II) porphyrin lipoic acid derivative was confi rmed by the electrochemi-cal behaviour of the surface-confi ned moiety in organic medium, from electrochemical reductive desorption and by STM. Electrochemical reductive desorption potential showed that the novel Co(II) porphyrin lipoic acid bis-esters 5 form packed and more stable SAMs compared to SAMs of a formerly investigated Co(II) porphyrin disulfi de derivative [3]. The charge corresponding to the monolayer desorption process strongly supports that all four sulfur atoms are involved in the self-assembly process, thus signifi cantly increasing SAM stability and consistent with a coverage of 2.7 × 10-10 mol.cm-2. The immobilized Co(II) complexes revealed to be catalytic for the oxygen reduction reaction, showing that this class of porphyrins has potential applications in this fi eld. Whether hydrogen peroxide or water are formed during the reaction, still needs to be disclosed with more detailed electrochemical studies.

Acknowledgements

The authors are grateful to CRUP/DAAD (Action A-1/06) and DFG Mo 274/24-1 for fi nancial support. We thank Mr. J. Stelten for NMR investigations, Dr. Thomas Dülcks and Mrs. D. Kemken for recording the MS spectra.

REFERENCES

1. Fragoso A, Laboria N, Latta D and O’Sullivan CK. Anal. Chem. 2008; 80: 2556–2563.

2. Bonifazi D, Enger O and Diederich F. Chem. Soc. Rev. 2007; 36: 390–414.

3. Marois J-S and Morin J-F. Langmuir 2008; 24:10865–10873.

4. Hutchison JE, Postlethwaite TA, Chen CH, Hath-cock KW, Ingram RS, Ou W, Linton RW, Murray RW, Tyvoll DA, Chung LL and Collman JP. Lang-muir 1997; 13: 2143–2146.

5. Jiao J, Nordlund E, Lindsey JS and Bocian DF. J. Phys. Chem. C 2008; 112: 6173–6180.

6. Watcharinyanon S, Nilsson D, Moons E, Shaporenko A, Zharnikov M, Albinsson B, Martensson J and Johanssona LSO. Phys. Chem. Chem. Phys. 2008; 10: 5264–5275.

7. Hutchison JE, Postlethwaite TA and Murray RW. Langmuir 1993; 9: 3277–3283.

8. Shimazu K, Takechi M, Fuji H, Suzuki M, Saiki H, Yoshimura T and Uosaki K. Thin Solid Films 1996; 273: 250–253.

9. Lu X, Jin J, Jingwang K, Lv B, Liu Hongde and Geng Z. Mater. Chem. Phys. 2002; 77: 952–957.

10. Nishimura N, Ooi M, Shimazu K, Fujii H and Uosaki K. J. Electroanal. Chem. 1999; 473: 75–84.



11. Viana AS, Leupold S, Montforts FP and Abrantes LM. Electrochim. Acta. 2005; 50: 2807–2813.

12. Cordas CM, Viana AS, Leupold S, Montforts FP and Abrantes LM. Electrochem. Commun. 2003; 5:36–41.

13. Wedel M, Walter A and Montforts FP. Eur. J. Org. Chem. 2001; 9: 1681–1687.

14. Fadnavis NW and Koteshwar K. Tetrahedron: Asymmetry 1997; 8: 337–339.

15. Finklea HO. Self-assembled Monolayers on Elec-trodes, in Encyclopedia of Analytical Chemistry,Meyers RA. (Ed.), John Wiley & Sons Inc: Chich-ester, USA, 2000.

16. Viana AS, Kalaji M and Abrantes LM. Electrochim. Acta 2002; 47: 1587–1594.

17. Cosnier S, Gondran C, Wessel R, Montforts FP and Wedel M. Sensors 2003; 2: 213–222.

18. Imahori H, Norieda H, Ozawa S, Ushida K, Yamada H, Azuma T, Tamaki K and Sakata Y. Langmuir1998; 14: 5335–5338.

19. Porphyrins and Metalloporphyrins, Smith KM. (Ed.) Elsevier: Amsterdam, 1975.

20. Pérez-Morales M, Miguel G, Munoz E, Martín-Romero T and Camacho L. Electrochim. Acta 2009; 54: 283–289.

21. Rodrigues F, Mebrahtu T and Soriaga MP. J. Elec-troanal. Chem. 1987; 253: 283–289.

22. Zhong CJ, Zak J and Porter MD. J. Electroanal. Chem. 1997; 421: 9–13.

23. Walzack MM, Alves CA, Lamp BD and Porter MD. J. Electroanal. Chem. 1995; 396: 103–114.


DOI: 10.1142/S1088424610001787



INTRODUCTION

Porphyrins and phthalocyanines (Pcs) represent a major class of dyes and are applied extensively in bio-logical systems and in material science because of their unique molecular assembly properties and excellent optical and electrical properties [1,2]. A lot of work has been done on the modifi cation of the Pcs and porphyrins so that they may be used for various applications [3–5]. For example, substituted Pcs containing hydrocarbon chains and/or crown ether (CE) units self-assembled into highly organized columnar aggregates are used for one-dimensional charge and ion transport [6].

Various tetrathiafulvalene (TTF)-modifi ed Pcs and porphyrins have been developed since the fi rst report

of a symmetrically-functionalized Pc with eight (TTF) units appeared in 1996 [7–19]. These compounds exhib-ited interesting optical and electrical properties such as photoinduced electron transfer followed by fl uorescence. This behavior results from the combination of Pc or por-phyrin units and TTF units into a single molecule. The unique redox properties of TTF is characterized by a sequential and reversible oxidation to a radical cation (TTF+•) and a dication (TTF+2) [20–22]. Interestingly, a tetra(thiafulvalene-crown-ether) phthalocyanine was found to self-assemble into helical tapes (nanometers in width and micrometers in length) and showed potentially novel electronic and structural properties [23]. Although TTF annulated macrocycles have received less atten-tion, they are believed to be more attractive candidates for various applications than ensembles of TTF and Pcs or porphyrins linked by spacers because they can self-assemble into various organized aggregates [24–29]. Normally, Pcs or porphyrins that are directly annulated

Synthesis, characterization and electrochemistry of the novel metalloporphyrazines annulated with tetrathiafulvalene having pentoxycarbonyl substituents

Fengshou Lenga, Ruibin Houa, Longyi Jina, Bingzhu Yin*a and Ren-Gen Xiong*b

a Key laboratory of Organism Functional Factors of Changbai Mountain, Yanbian University, Ministry of Education, Yanji 133002, Chinab Ordered Matter Science Research Center, College of Chemistry and Chemical Engineering, Southeast University, Nanjing 210096, China

Received 13 April 2009Accepted 24 June 2009

ABSTRACT: Three novel tetrathiafulvalene-annulated metalloporphyrazines with electron-withdrawing pentoxycarbonyl groups at the periphery were synthesized via the cyclotetramerization of dipentyl 6,7-dicyanotetrathiafulvalen-2,3-dicarboxylate in the presence of corresponding metal salts (Zn(OAc)2·2H2O,Cu(OAc)2·2H2O, and NiCl2) and in pentanol. Molecular structures were fully characterized by 1H NMR, FT-IR, UV-vis, MALDI-TOF mass spectra and elemental analysis. These newly synthesized macrocyclic dyes were suffi ciently stable in air during the purifi cation process and also in further experiments. Electron-withdrawing substituents reduced the ability of tetrathiafulvalene groups to form radical cations. Solution electrochemical data showed one reductive and three oxidative processes within a -2000 mV to +2200 mV potential window. The four couples observed were assigned to Pz-2/Pz-3 (I), TTF+•/TTF (II), TTF+2/TTF+• (III) and Pz-1/Pz-2 (IV).

KEYWORDS: tetrathiafulvalenedinitrile, metalloporphyrazines, electrochemistry, tetramerization, aggregation.

*Correspondence to: Bingzhu Yin, email: [email protected], tel: +86 433-2732298, fax: +86 433-2732456 and Ren-gen Xiong, email: [email protected]


NOVEL METALLOPORPHYRAZINES ANNULATED WITH TETRATHIAFULVALENE HAVING PENTOXYCARBONYL 109

with four TTF units do not show luminescence because of the presence of fused electron-donating TTF units. Their fl uorescent forms can, however, be achieved if the TTF units are oxidized using various chemi-cal oxidants or an electrochemical method. Hence they could be regarded as electro-switchable fl uores-cent molecules. Electro-switched luminescence has been observed in a mono TTF-annulated porphyrin and tetrakis-TTF-Pc molecules [25–26, 30]. The syn-thesis of tetrakis-TTF-porphyrin resulted in a mixture of the neutral porphyrin and its corresponding radi-cal cation. We have recently also reported the synthe-sis of novel porphyrazines (Pzs) annulated with four TTF units that also contain electron-donating alkyl-thio groups. These molecules spontaneously form TTF group radical cations during the separation process [31]. This hindered an accurate assessment of their photophysical and electrochemical properties.

To investigate the effect of peripheral substituents on the stability of TTF annulated Pzs we introduced electron-withdrawing pentoxycarbonyl groups onto TTF annulated Pzs. Three novel TTF-annulated Pz dyes, as shown in Scheme 1, were thus synthesized and we evalu-ated their photophysical and electrochemical properties.

EXPERIMENTAL

General

NMR spectra were recorded in CDCl3 with a Bruker AV-300 Spetrometer (300 MHz for 1H and 75 MHz for

13C), and chemical shifts were referenced relative to tetra-methylsilane (δH/δC = 0). The UV-vis spectra were recorded on a Hitachi U-3010 spectrophotometer in CHCl3 (c = 2 × 10-5 M). Mass spectrometry was performed on a Hewl-ett Packard 1100-HPLC/MSD instrument (APCI mode). HRMS data were obtained by a JEOL JMS SX 102A appa-ratus. MALDI-TOF-MS data were obtained by a Shimadzu AXIMA-CFRTM plus mass spectrometry, using a 1,8,9- anthracenetriol (DITH) matrix. Cyclic voltammetry was carried out on a Potentiostat/Galvanostat 273A instrument in CH3CN-CH2Cl2 with 0.1 M Bu4PF6 as supporting electro-lyte and the scan rate was 100 mV.s-1. Counter and working electrodes were made of Pt and Glass-Carbon, respectively, and the reference electrode was calomel electrode (SCE). IR spectra were recorded on a Shimadu FT-IR 1730 instru-ment (KBr pressed disc method). Mass spectrometry was performed on a Hewlett Packard 1100-HPLC/MSD instru-ment. The ESR spectra were recorded using a FA200GEOL spectrometer at 25 °C. Crystal data were measured on a Rigaku SCXmini diffractometer with Mo K radiation (λ = 0.71073 Å) by scan mode at 293(2) K.

Synthesis

All reagents and solvents were of commercial qual-ity and distilled or dried when necessary using standard procedures. All reactions were carried out under argon atmosphere. Starting materials 1 [32], 4,5-dicyano-1,3-dithiol-2-one [33] and 2a [34] were prepared according to methods described in referenced literature.

Dipentyl 2-thioxo-1,3-dithiole-4,5-dicarboxylate (2b). A mixture of 1 (1000 mg, 4 mmol) and K2CO3 (100 mg,

S

SO

NC

NC

S

SS

CO2CH3

CO2CH3P(OEt)3

toluene/refluxS

SNC

NC S

S CO2R

CO2R

M(OAc)2

PeOH/refluxN

N

N

NN

N

N

N

SS

S

S

S S

S

S

SS

S

S

S S

S

S

RO2C CO2R

CO2R

CO2R

RO2C CO2R

RO2C

RO2C

M

R = pentyl4 M = Zn 5 M = Cu6 M = Ni

1 2a R = butyl2b R = pentyl

3a R = butyl3b R = pentyl

3b

ROH

K2CO3 S

SS

CO2R

CO2R

Scheme 1. Synthetic route for tetrathiafulvalene-annulated metalloporphyrazines (4–6)


110 F. LENG ET AL.

0.15 mmol) in 10 mL pentanol was stirred for 72 h at room temperature. Methanol produced was removed uninter-ruptedly under reduced pressure. The reaction mixture was concentrated in vacuum to yield yellow oil, which was puri-fi ed by column chromatography (silica gel, CH2Cl2/P.E. = 1:1, Rf = 0.74) to give pure 2b as yellow oil. Yield 1160 mg (80.3%).1H NMR (300 MHz; CDCl3; Me4Si): δH, ppm 0.94 (3H, s, -CH2-CH3), 1.37 (4H, s,-CH2-CH2-CH3), 1.73 (2H, t, J = 6.6 Hz, -CH2-CH2-), 4.30 (2H, t, J = 6.6 Hz, -O-CH2-).

13C NMR (75 MHz; CDCl3; Me4Si): δC, ppm 13.92 (-CH2-CH3), 22.23 (-CH2--CH3), 27.87 (-CH2-CH2-CH2-), 27.96 (-CH2-CH2-), 67.44 (O-CH2-), 138.36 (C=C), 157.56 (C=O), 207.44 (C=S). MS (APCI): m/z 362.0448 (calcd. for [M + H]+ 363.53).

Dialkyl 2,3-dicyanotetrathiafulvalenes-6,7-di-carboxylate (3a-3b). According to literature procedure [34], an equimolar mixture of 4,5-dicyano-1,3-dithiol-2-one (2 mmol) and 2 (2 mmol) in 60 mL toluene was added dropwise to a refl uxing mixture of 10 mL P(OEt)3 and 20 mL toluene. The mixture was refl uxed for 1 h. After cooling to room temperature, the solvent was removed in vacuum and the red oily residue obtained was chromatographed on a silica gel column (CH2Cl2/P.E. = 1/1, v/v) to afford dark red solid 3. Compound 3a: recrystallization from petro-leum ether gave 3a as dark red needles. Yield 122 mg (13.6%), mp 66–67 °C. Anal. calcd. for C18H18N2O4S4: C, 47.56; H, 3.99; N, 6.16. Found: C, 47.55; H, 3.71; N, 6.00. 1H NMR (300 MHz; CDCl3; Me4Si): δH, ppm 0.96 (3H, t, J = 7.2 Hz, -CH2-CH3), 1.34–1.47 (2H, m, -CH2-CH3),1.64–1.73 (2H, m, -CH2-CH2-), 4.26 (2H, t, J = 6.6 Hz, -O-CH2-CH2-).

13C NMR (75 MHz; CDCl3; Me4Si): δC,ppm 13.62 (-CH2-CH3), 18.99 (-CH2-CH3), 30.27 (-CH2-CH2-), 67.07 (COO-CH2-CH2-), 101.94 (C-CN), 108.85 (NC-C=C-CN), 118.71 (-S-C=C-S-), 121.10 (S-C=C-S),131.76 (OOC-C=C-COO-), 158.64 (-C-COO-). IR (KBr pellets): νmax, cm-1 2959 (C-H), 2870 (C-H), 2216 (CαN),1734 (C=O), 1705 (C=O), 1579, 1294, 1252, 1026. MS (APCI): m/z 477.0 (calcd. for [M + Na]+ 477.61). Com-pound 3b: recrystallization from petroleum ether gave 3bas a dark red needles. Yield 214 mg (22.2%), mp 64.7 °C. Anal. calcd. for C20H22N2O4S4: C, 49.77; H, 4.59; N, 5.80. Found: C, 49.55; H, 4.73; N, 5.57. 1H NMR (300 MHz; CDCl3; Me4Si): δH, ppm 0.92 (3H, t, J = 3.0 Hz ,-CH2-CH3), 1.35 (4H, br, -CH2-CH2-CH3), 1.70 (2H, br, -CH2-CH2-), 4.25 (2H, t, J = 6.6 Hz, -CH2-COO-). 13C NMR (75 MHz; CDCl3; Me4Si): δC, ppm 13.95 (-CH2-CH3), 22.23 (-CH2-CH3), 27.84 (-CH2-CH2-), 28.81 (-CH2-CH2-),67.36 (COO-CH2-CH2-), 101.97 (C-CN), 108.88 (NC-C=C-CN), 118.72 (-S-C=C-S-), 120.93 (-S-C=C-S-),131.79 (OOC-C=C-COO-), 158.64 (-C-COO-). IR (KBr pellets): νmax, cm-1 2954 (C-H), 2860 (C-H), 2210 (CαN),1740 (C=O), 1701 (C=O), 1290, 1224. MS (APCI): m/z483.0 (calcd. for [M + H]+ 483.66).

{2,3,7,8,12,13,17,18-tetrakis[6,7-bis(alkyl-oxycarbonyl)tetrathiafulvalene]porphyrazines} M(II) (M = Zn for 4, Cu for 5 and Ni for 6). A mixture of 97 mg 3b (0.2 mmol) and the appropriate metal salt (0.1 mmol)

(Zn(OAc)2·2H2O for 4, Cu(OAc)2·2H2O for 5, anhydrous NiCl2 for 6) was heated in 10 mL n-pentanol at 145 °C for 4 h. The color of reaction solution changed from dark red to black. The solvent was concentrated in vacuum to give 4 as black powder. The solid obtained was puri-fi ed on silica gel column chromatography (gradient of CH2Cl2/MeOH = 1/0 → 50:1, v/v) to give 4. Compound 4: reprecipitation from CH2Cl2-MeOH gave 4 as dark blue powder. Yield 56 mg (56%), mp 260 °C (decom-posed). Anal. calcd. for C80H88N8O16S16Zn: C, 48.14; H, 4.44; N, 5.61. Found: C, 48.32; H, 4.23; N, 5.80. 1HNMR (300 MHz; CDCl3; Me4Si): δH, ppm 0.95 (24H, br, -CH2-CH3), 1.39 (32H, br, -CH2-CH2-CH3), 1.68 (16H, br, -CH2-CH2-), 4.22 (16H, br, -CH2-COO-). 13C NMR (75 MHz; CDCl3; Me4Si): δC, ppm 13.97 (-CH2-CH3),22.39 (-CH2-CH3), 27.97 (-CH2-CH2-), 66.66 (COO-CH2-CH2-), 130–134 (br), 157-160 (br). IR (KBr pellets): νmax, cm-1 2955–2856 (C-H), 1730 (C=O), 1575, 1462, 1250, 1094. UV-vis (CHCl3): λmax, nm (log ε) 291 (4.53), 358 (4.52), 605 (4.35). MS (MALDI-TOF): m/z 1996.81 (calcd. for [M + H]+ 1997.05). Compound 5: reprecipita-tion from CH2Cl2-MeOH gave 5 as dark purple powder. Yield 52 mg (52.4%), mp 250 °C (decomposed). Anal. calcd. for C80H88CuN8O16S16: C, 48.18; H, 4.45; N, 5.62. Found: C, 48.27; H, 4.36; N, 5.54. 1H NMR (300 MHz; CDCl3; Me4Si): δH, ppm 0.92 (24H, br, -CH2-CH3), 1.35 (32H, br, -CH2-CH2-CH3), 1.68 (16H, br, -CH2-CH2-),4.32 (16H, br, -CH2-COO-). 13C NMR (75 MHz; CDCl3;Me4Si): δC, ppm 13.92 (-CH2-CH3), 22.26 (-CH2-CH3),27.88 (-CH2-CH2-), 28.07 (-CH2-CH2-), 66.73 (COO-CH2-CH2-), 130-133 (br), 157-160 (br). IR (KBr pellets): νmax, cm-1 2955-2862 (C-H), 1726 (C=O), 1576, 1460, 1246, 1088. UV-vis (CHCl3): λmax, nm (log ε) 290 (4.78), 317 (4.76), 567 (4.33), 601 (4.31). MS (MALDI-TOF): m/z 1994.44 (calcd. for [M + H]+ 1995.18). Compound 6:reprecipitation from CH2Cl2-MeOH gave 6 as dark pur-ple powder. Yield 24.5 mg (24.6%), mp 250 °C (decom-posed). Anal. calcd. for C80H88N8NiO16S16: C, 48.30; H, 4.46; N, 5.63. Found: C, 48.19; H, 4.45; N, 5.44. 1HNMR (300 MHz; CDCl3; Me4Si): δH, ppm 1.05 (24H, br, -CH2-CH3), 1.48 (32H, br, -CH2-CH2-CH3), 1.80 (16H, br, -CH2-CH2-), 4.25 (16H, br, -CH2-COO-). 13C NMR (75 MHz; CDCl3; Me4Si): δC, ppm 14.13 (-CH2-CH3),22.47 (-CH2-CH3), 28.08 (-CH2-CH2-), 29.69 (-CH2-CH2-), 66.56 (COO-CH2-CH2-). IR (KBr pellets): νmax,cm-1 2955–2862 (C-H), 1730 (C=O), 1578, 1642, 1230, 1113. UV-vis (CHCl3): λmax, nm (log ε) 291 (4.78), 325 (4.79), 557 (4.55), 593 (4.40). MS (MALDI-TOF): m/z1988.49 (calcd. for [M + H]+ 1989.48).


Synthesis and characterization

The synthesis of target compounds 4–6 is shown in Scheme 1. The fi rst step in the synthetic procedure is the



synthesis of dicyanotetrathiafulvalene 3 that contains electron-withdrawing ester groups. The cross-coupling reaction of 4,5-dicyano-1,3-dithiol-2-one and dialkyl 1,3-dithiol-2-thione-4,5-dicarboxylate (2a–b) in a mix-ture of triethyl phosphite and toluene at refl ux and under argon leads to key intermediates 2,3-dicyano-6,7-bis-(pentyloxycarbonyl) TTF (3a–b) in 14% and 22% yields, respectively.

Red crystals that were suitable for X-ray diffraction analysis of intermediate 3a was obtained by the slow evaporation of a petroleum ether (30–60 °C) solution of 3a at room temperature. Compound 3a crystallizes in a monoclinic space group (P21/c) and the 1, 3-dithiole-4,5-dicarbonitrile unit is almost coplanar with the ester sub-stituent dithiole ring [dihedral angle = 2.9(2)°] (Fig. 1). In the crystal structure, intermolecular S…O interactions [S3...O2 = 3.013 Å] create dimers between neighboring molecules in the ac plane. In addition, a weak ··· stack-ing interaction exists between adjacent dithiole rings, with centroid-centroid distances of 4.075 Å, and chains formed along the b axis (Figs S1–S2, see Supporting information section).

Using the building block 3b, we synthesized TTF annulated metallo-Pzs (4–6) by a metal templated tetramerization in 1-pentanol and obtained yields of 56% for Zn-TTF-Pz 4, 52% for Cu-TTF-Pz 5 and 15% for Ni-TTF-Pz 6, respectively. As expected, the TTF annulated Pz dyes were suffi ciently stable for purifi cation and for further experiments. The compounds are soluble in com-mon organic solvents such as hexane, benzene, toluene, CHCl3, CH2Cl2, DMF, acetone and THF. The compounds are insoluble in alcohols and DMSO. We were unable to demetallate the metallo-Pzs 4–6 using acidic reaction conditions because the ester groups decomposed [35].

1H NMR spectra of 4–6 in CDCl3 displayed four very broad signals which could be explained by considering that slow tumbling results from aggregation in concen-trated solutions (Fig. S2) [24, 33]. The MALDI-TOF mass spectra of 4–6 showed peaks at m/z = 1996.81 (M+ = 1996.05) for 4, 1994.44 (M+ = 1994.18) for 5 (Fig. S3) and 1988.49 (M+ = 1988.48) for 6. Results from ele-mental analysis confi rmed the proposed structures of compounds 4–6.

Photophysical properties

UV-vis spectra of TTF annulated Pzs 4–6 in CH2Cl2

are shown in Fig. 2. Compounds 4–6 show typical Pz electronic spectra, consisting of two strong absorption regions. Broadening of the Q and B bands is attributed to n-π* transitions of non-bonding electrons, that are associated with peripheral S and N atoms [36], and also attributed to aggregation of the macrocyclic system. The B band (or Soret band) of metallo-Pz rings in the ultra-violet region for compound 4 arises from deep π LUMOtransitions (between a2u and eg orbitals). TTF absorption in this region thus leads to superimposed bands. In the visible region, 4 had an intense Q band at 605 nm with a shoulder at around 574 nm and this corresponds to mono-meric and aggregated Pz in chloroform, respectively. The indistinct shoulder band at about 574 nm, arising from the dimer, is attributed to a π→π* transition from the HOMO to the LUMO of the Pz2- ring. For 5 the Q band of the dimer at higher energy was centered at 566 nm and the Q band arising from the monomer was at a lower energy and centered at 601 nm as a weak shoulder band. The absorption bands of 6 are similar to those of 5 except that the sharper component of the Q band at higher energy is less intense in 5 and the monomer shoulder band is stronger in 5. This indicates that there is a higher degree of face-to-face interaction in 6 compared with the inter-actions in 4 and 5 and this is typical of metallated sym-metrically substituted Pzs with D4h symmetry [37].

In order to address the donor properties of target com-pounds, 4 was selected and dope with TCNQ in CH2Cl2.However no CT band was observed at 600–1000 nm region. This might be attributed to the presence of eight strongly electron-withdrawing pentoxycarbonyl groups onto TTF annulated Pzs. Whereas, doping of F4TCNQ to a CH2Cl2 solution of 4 (2 × 10-5 M) resulted in the appear-ance of two new CT absorption bands, centered on λmax = 756 nm and λmax = 863 nm in the UV-vis spectrum (Fig. 3). These new bands corresponds to the SOMO-LUMO transition of the cation radical species of the TTF moieties [23]. The formation of F4TCNQ-•/TTF+• charge-transfer

Fig. 1. ORTEP plot of the molecule of 3a

Fig. 2. Absorption spectra of 4–6 in CHCl3 (2 × 10–5 M)


112 F. LENG ET AL.

complex in the mixture F4TCNQ-4 (4:1) in CH2Cl2 was also confi rmed by the FT-IR and ESR spectra. A FT-IR spectrum showed the nitrile stretch of the F4TCNQ radi-cal anion at 2212 cm-1 [38] compared to the neutral state of 2222 cm-1. In addition, an electron paramagnetic resonance (ESR) spectrum of 4 was centered around

g = 2.007 and 2.002, which are in the region characteris-tic of both a TTF radical cation [39] and a F4TCNQ radi-cal anion [38]. These results show that in solution some CT takes place between the TTF unit(s) and F4TCNQ(Fig. S4).

Cyclic voltammetric (CV) studies were performed on compounds 4–6 and their intermediates 3a–b in a mixture of CH3CN-CH2Cl2. The data are collected in Table 1. As seen in Table 1, the cyclic voltammograms of 3a–b reveal that all redox processes are quasi-reversible. Each oxidation wave of the precursor compounds 3aand 3b is assigned to a one-electron process. By com-parison to tetrathiafulvalene [40], both oxidation peaks were shifted signifi cantly to higher oxidation potentials (1.034 and 1.275 V for 3a and 1.035 and 1.278 V for 3b, respectively) because of the electron-withdrawing effect of the cyano and alkoxycarbonyl groups. For the electrochemical characterization of compounds 4–6,the cyclic voltammograms were indistinct because of aggregation in solution. This behavior has previously been reported in the literature [26, 41]. Differential pulse voltammetry (DPV) provided more detailed informa-tion. Figure 4 shows cyclic (a) and differential pulse vol-tammograms (b) of 4 within a -2000 mV to +2200 mV potential window. Compound 4 shows redox processes

Fig. 3. Absorption spectra of 4 (2 × 10-5 M) in the CH2Cl2 after addition of different equiv. F4TCNQ

Table 1. Cyclic voltammetric data for M-TTF-Pzs 4–6 and intermediates 3a–b in a mixture of CH2Cl2 and CH3CN (4:1, v/v)

Compound E11/2(ΔE)/V E2

1/2(ΔE)/V E31/2(ΔE)/V E4

pa/V

TTF[40] 0.36 0.77

3a 1.034 (0.073) 1.275 (0.089)

3b 1.035 (0.077) 1.278 (0.087)

4 -0.813 (0.273) 0.914 (0.088) 1.308 (0.150) 1.786

5 -0.802 (0.283) 0.972 (0.106) 1.304 (0.137) 1.784

6 -0.836 (0.283) 0.998 (0.102) 1.313 (0.106) 1.784

Fig. 4. Cyclic (a) and differential pulse voltammogram (b) of compound 4 in CH3CN-CH2Cl2 (1:4, v/v) containing 0.1 M Bu4PF6.Scan rate was 100 mV.s-1



at E1/2 = -0.813 (I), 0.914 V (II), 1.308 V (III) and 1.786 V (IV) vs the SCE and all processes are not completely reversible in terms of ΔE, the shift of Ep at different scanning rates and the ratio of anodic to cathodic peak currents (ipa/ipc) [42]. Process I which is assigned to a reduction of the porphyrazine ring system (Pz-2/Pz-3) is irreversible because of the excessive anodic to cathodic peak separation (ΔE > 200 mV) and the difference of anodic to cathodic peak currents. Processes II and III are quasi-reversible because the anodic to cathodic peak cur-rents are near unity while ΔE for II and III are 88 mV and 150 mV, respectively. These processes represent the two four-electron oxidation processes of the TTF moi-eties. Interestingly, 4 shows similar behavior to its phtha-locyanine analog [25] with negative shifts in oxidation potentials for the radical cations of TTF (0.12 V). The splitting of the fi rst oxidation wave (process II) arises from the molecular system where the donor moieties interact through conjugation and/or through space. On the other hand, compound 4 is distinct from its phthalo-cyanine analogs and an irreversible one-electron wave is observed at 1.786 V (IV), which is assigned to oxidation of the Pz ring system (Pz-1/Pz-2). The redox processes of compounds 5 and 6 are similar to 4 except that their fi rst oxidation waves are not split.

In summary, we developed an effi cient synthetic route to the novel porphyrazines that are annulated with tetrathiafulvalene having pentoxycarbonyl substitu-ents. Effi cient modulation of macrocycle stability was achieved by the introduction of electron-withdrawing pentoxycarbonyls. Ability of 4 to function as a donor for F4TCNQ was established and the formation of TTF+/F4TCNQ- charge-transfer complex was confi rmed by the UV-vis, FT-IR and ESR spectral. Compounds 5 and 6showed similar cyclic voltammetric behavior with one reduction process and three oxidation processes. These were assigned to Pz-2/Pz-3 (I), TTF+•/TTF (II), TTF+2/TTF+• (III) and Pz-1/Pz-2 (IV). The synthesis, self-as-sembly and fi lm forming properties of TTF-annulated Pzs with longer alkyls and containing oxyethylenes are under investigation.

Acknowledgements

This work was supported by National Science Foun-dation of China (Grant No. 20662010), the Specialized Research Fund for the Doctoral Program of Higher Edu-cation (Grant no. 20060184001) and the open project of State Key Laboratory of Supramolecular Structure and Materials, Jilin University. Ren-Gen Xiong only contrib-utes the X-ray determination.


Crystal data and structure of 3a, 1H NMR and MALDI-TOF-MS spectra of 5, and ESR spectrum of 4(Figs S1–S4, Table S1) are given in the supplementary

material. This material is available free of charge via the Internet at http://www.worldscinet.com/jpp/jpp.shtml.

REFERENCES

1. The Porphyrin Handbook, Vols 1–10 and 11–14, Kadish KM, Smith KM and Guilard R. (Eds.) Aca-demic Press: San Diego CA, 2000 and 2003.

2. Alexiou C and Lever ABP. Coord. Chem. Rev. 2001; 216–217: 45–54.

3. McKeown NB. Phthalocyanine Materials: Synthe-sis, Structure and Function, Cambridge University Press: Cambridge, 1998.

4. The Porphyrin Handbook, Vols 15–20, Kadish KM, Smith KM and Guilard R. (Eds.) Academic Press: San Diego CA, 2003.

5. Inabe T and Tajima H. Chem. Rev. 2004; 104:5503–5534.

6. Van Nostrum CF and Nolte RJM. Chem. Commun.1996; 2385–2392.

7. Blower MA, Bryce MR and Devonport W. Adv. Mater. 1996; 8: 63–65.

8. Cook MJ, Cooke G and Ali JF. Chem. Commun.1996; 1925–1926.

9. Wang CS, Bryce MR, Batsanov AS, Stanley CF, Beeby A and Howard JAK. J. Chem. Soc., Perkin Trans. 2 1997; 9: 1671–1678.

10. Hu YY and Shen YJ. J. Heterocycl. Chem. 2002; 39:1071–1075.

11. Hu YY, Lai GQ, Shen YJ and Li YF. J. Porphyrins Phthalocyanines 2004; 8: 1042–1046.

12. Hu YY, Lai GQ, Shen YJ and Li YF. Dyes Pigm. 2005; 66: 49–53.

13. Hu YY, Lai GQ, Shen YJ and Li YF. Monatsh.Chem. 2004; 135: 1167–1172.

14. Farren C, Christensen CA, FitzGerald S, Bryce MR and Beedy A. J. Org. Chem. 2002; 67: 9130–9139.

15. Wang CS, Bryce MR, Batsanov AS and Howard JAK. Chem. Eur. J. 1997; 3: 1679–1690.

16. Liddell PA, Kodis G, de la Garza L, Bahr JL, Moore AM, Moore TA and Gust D. Helv. Chim. Acta 2001; 84: 2765–2773.

17. Saha S, Johansson E, Flood AH, Tseng H-R, Jeffrey I, Zink JI and Stoddart JF. Chem. Eur. J. 2005; 11:6846–6858.

18. Kodis G, Liddell PA, Garza L, Moore AL, Moore TA and Gust D. J. Mater. Chem. 2002; 12: 2100–2108.

19. Sadaike S-I, Takimiya K, Aso Y and Otsubo T. Tetrahedron Lett. 2003; 44: 161–165.

20. Williams JM, Ferraro JR, Thorn RJ, Carlson KD, Geise U, Wang HH, Kini AM and Wangbo M-H. In Organic Superconductor Including Fullerenes,Prentice Hall: Engelwood Cliffs, 1992; pp 1–210.

21. Bryce MR. J. Mater. Chem. 2000; 10: 589–598. 22. Nielsen MB, Lomholt C and Becher J. Chem. Soc.

Rev. 2000; 29: 153–164.


114 F. LENG ET AL.

23. Sly J, Kasak P and Nolte RJM. Chem. Commun. 2005; 1255–1257.

24. Becher J, Brimert T, Jeppesen JO, Pedersen JZ, Zubarev R, Njørnholm T, Reitzel N, Jensen TR, Kjaer K and Levillain E. Angew. Chem. Int. Ed. 2001; 40: 2497–2500.

25. Li HC, Jeppesen JO, Levillain E and Becher J.Chem. Commun. 2003; 846–847.

26. Loosli C, Jia C, Liu S-X, Haas M, Dias M, Levillain E, Neels A, Labat G, Hauser A and Decurtins S. J. Org. Chem. 2005; 70: 4988–4992.

27. Donders CA, Liu S-X, Loosli C, Sanguinet L, Neels A and Decurtinsa S. Tetrahedron 2006; 62:3543–3549.

28. Kimura T, Watanabe D and Namauo T. Heterocycles2008; 76: 1023–1026.

29. Wang R, Liu W, Chen Y, Zuo JL and You XZ. Dyes Pigm. 2009; 81: 40–44.

30. Nielsen KA, Eric Levillain, Lynch VM, Ses-sler JL and Jeppesen JO. Chem. Eur. J. 2009; 15:506–516.

31. Chen T, Wang CL, Qiu H, Jin LY and Yin BZ. Heterocycles 2007; 71: 549–555.

32. Melby LR, Hartyzler HD and Sheppard WA. J. Org. Chem. 1974; 39: 2456–2458.

33. Chen T, Wang CL, Cong ZQ and Yin BZ. Hetero-cycles 2005; 65: 187–193.

34. Iyoda M, Enozawa H and Miyake Y. Chem. Lett. 2004; 33: 1098–1099.

35. Alzeer J, Roth PJC and Luedtke NW. Chem. Com-mun. 2009; 15: 1970–1971.

36. Kandaz M, Cetin HS, Koca A and Ozkaya AR. DyesPigm. 2006; 74: 298–305.

37. Karadeniz H, Kantekin H, Gök Y and Çelenk E. Dyes Pigm. 2008; 77: 559–563.

38. Akutagawa T, Kakiuchi K, Hasegawa T, NakamuraT, Christensen CA and Becher J. Langmuir 2004; 20: 4187–4195

39. Wudl F, Smith GM and Hufnagel EJ. J. Chem. Soc., Chem. Commun. 1970; 1453–1454.

40. Akutagawa T, Kakiuchi K, Hasegawa T, Nakamura T, Christensen CA and Becher J. Langmuir 2004; 20: 4187–4195.

41. Iyoda M, Hara K, Ogura E, Takano T, Hasegawa M, Yoshida M, Kuwatani Y, Nishikawa H, Kikuchi K, Ikemoto I and Mori T. J. Solid State Chem 2002; 168: 597.

42. Lever ABP, Milaeva ER and Speier G. In Phthalocy-anines: Properties and Applications, Vol. 3, Leznoff CC and Lever ABP. (Eds.) VCH: New York, 1993.

43. Nombona N and Nyokong T. Dyes Pigm. 2009; 80:130–135.


DOI: 10.1142/S1088424610001714



INTRODUCTION

It is well-known that - interactions between two or more porphyrin molecules in close proximity are pres-ent in photosynthetic proteins, including light-harvesting chlorophyll arrays [1] and the photosynthetic reaction center special pair [2]. The interactions have played a critical role in electron (or energy) transfer in photo-synthetic proteins. Scheidt and Lee [3] surveyed the inter-ring geometry for all structurally characterized neutral porphyrin dimers. They noted that the observed lateral shifts tend to cluster around specifi c values rather than display a continuous distribution. It has been shown that

such dimerizations have a strong effect on the chemical reactivity and spectroscopic properties of the porphyri-nato complexes [4–6].

Studies on the replacement of meso-aryl substitu-ents in metallotetraryl-porphyrins by alkyl groups have attracted much attention recently owing to the fact that the alkyl-substituents may profoundly infl uence the molecular and electronic structure as well as catalytic properties of the resulting complexes [7–28]. In the pres-ent work, we describe the synthesis and characteriza-tion of a series of meso-(tetraalkylporphinato)iron(III) chloride complexes. Their molecular structures indicate that inter-ring interactions change with alteration of the alkyl group substitutions in the meso positions. We have also investigated the effect of such inter-ring interaction on magnetic exchange interactions between the iron centers of these meso-alkyl substituted porphyrinates.

Inter-ring interactions in [Fe(TalkylP)(Cl)] (alkyl = ethyl, n-propyl, n-hexyl) complexes: control by meso-substitutedgroups

Ming Lia, Teresa J. Neala, Noelle Ehlingera, Charles E. Schulz*b andW. Robert Scheidt*a

a Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana 46556, USAb Department of Physics, Knox College, Galesburg, Illinois 61401, USA

Received 26 August 2009Accepted 27 September 2009

ABSTRACT: Syntheses, molecular structures and magnetic susceptibilities of three meso-substituted high-spin iron(III) porphyrinate complexes ([Fe(TEtP)(Cl)], [Fe(TPrP)(Cl)], and [Fe(THexP)(Cl)]) are described. It was determined that the inter-ring interactions within each dimeric unit change upon alteration of the alkyl groups at the meso positions. Magnetic exchange couplings between iron centers of the dimers are in accord with the trends in structural inter-ring geometries. Crystal data for [Fe(TEtP)(Cl)]: a = 10.1710(5) Å, b = 11.309(3) Å, c = 12.170(3) Å, = 91.774(9)°, =113.170(14)°, = 112.149(9)°, V = 1165.2(4) Å3, triclinic, P1-, Z = 2, R1 = 0.0844 and wR2 = 0.2073 for observed data. Crystal data for [Fe(TPrP)(Cl)]: a = 13.040(2) Å, b = 15.221(2) Å, c = 14.6681(9) Å, = 109.997(11)°, V = 2735.9(7) Å3, monoclinic, P21/n, Z = 4, R1 = 0.0477 and wR2 = 0.1176 for observed data. Crystal data for [Fe(THexP)(Cl)]: a = 10.246(7) Å, b = 12.834(4) Å, c = 17.420(15) Å, = 69.74(3)°, = 87.52(4)°, = 84.89(3)°, V = 2140(2) Å3, triclinic, P1-, Z = 2, R1 = 0.1024 and wR2 = 0.2659 for observed data.

KEYWORDS: iron(III)meso-tetraalkylporphinates, inter-ring interaction, magnetic exchange couplings, crystal structures.


*Correspondence to: W. Robert Scheidt, email: [email protected]


116 M. LI ET AL.

EXPERIMENTAL

General information

Dichloromethane was distilled over potassium car-bonate and hexanes were distilled over sodium benzo-phenone. All other chemicals were used as received from Aldrich or Fisher. meso-tetra-n-propylporphyrin (H2TPrP) was prepared according to Neya’s method [29], while meso-tetraethylporphyrin (H2TEtP) and meso-tet-ra-n-hexylporphyrin (H2THexP) was prepared according to Lindsey’s method [30]. Iron was inserted into the three H2-meso-tetra alkylporphyrinato derivatives by standard methods [31]. UV-vis spectra were recorded on a Perkin-Elmer Lambda 19 spectrometer and IR spectra on a Per-kin-Elmer model 883 as KBr pellets. EPR spectra were obtained at 77 K on a Varian E-12 spectrometer operating at X-band.

Structure determinations

Single crystals of [Fe(TPrP)(Cl)] [32] and [Fe(TEtP)(Cl)] were obtained by slow diffusion of hexanes into a CH2Cl2 solution of the metalloporphyrin, whereas single crystals of [Fe(THexP)(Cl)] were obtained by slow evap-oration of a concentrated CH2Cl2 solution. X-ray diffrac-tion data for all the complexes were collected on a Nonius FAST area detector diffractometer with a Mo rotating anode source ( = 0.71073 Å). Our detailed methods and procedure for small molecule X-ray data collection have been described previously [33].

The structure of [Fe(TPrP)(Cl)] was solved by Patter-son methods, while [Fe(TEtP)(Cl)] and [Fe(THexP)(Cl)]

were solved by direct methods [34]. All remaining non-hydrogen atoms were located by difference Fourier syn-thesis. The structures were refi ned against F2 using the program SHELXL-93 [35], in which all data collected were used including negative intensities. Hydrogen atoms of the porphyrin ligands and the solvent molecule were idealized with the standard SHELXL-93 idealization methods. A modifi ed [36] version of the absorption cor-rection program DIFABS and extinction were applied for [Fe(TEtP)(Cl)] and [Fe(THexP)(Cl)]. For [Fe(THexP)(Cl)], it was found that three hexyl substituents were disordered. One hexyl chain is disordered over two half-occupied positions (C(80) to C(85) and C(90) to C(95)). The last three carbon atoms of another hexyl group are disordered; C(4) to C(6) have refi ned occupancies of 0.76, while C(41) to C(61) have occupancies of 0.24. In the third disordered hexyl group, the fi nal methyl group is disordered over two positions (C(24) and C(240)) with refi ned occupation factors of 0.63 and 0.37, respectively. There is one dichloromethane molecule with an occu-pancy factor of 0.10. Brief crystallographic data for all three complexes are listed in Table 1. Complete crys-tallographic details are available from the CCDC (see Supporting information section).

Magnetic susceptibility measurements

Magnetic susceptibility measurements were obtained on ground samples (immobilized in Dow Corning sili-cone grease) in the solid state over the temperature range 6–300 K on a Quantum Design MPMS SQUID sus-ceptometer. Measurements at two fi elds (2 and 20 kG) showed that no ferromagnetic impurities were present

Table 1. Crystallographic details for [Fe(TalkylP)(Cl)] complexes

Molecule [Fe(TEtP)(Cl)] [Fe(TPrP)(Cl)] [Fe(THexP)(Cl)]Formula C28H28ClFeN4 C32H36ClFeN4 C44H60ClFeN4·0.1CH2Cl2

FW, amu 511.84 567.95 744.75a, Å 10.1710(5) 13.040(2) 10.246(7)b, Å 11.309(3) 15.221(2) 12.834(4)c, Å 12.170(3) 14.6681(9) 17.420(15)α, deg 91.774(9) 90 69.74(3)β, deg 113.170(14) 109.997(11) 87.52(4)γ, deg 112.149(9) 90 84.89(3)V, Å3 1165(4) 2735.9(7) 2140(2)Crystal system triclinic monoclinic triclinicSpace group P1 P21/n P1Z 2 4 2

Dc, g/cm3 1.459 1.379 1.156F(000) 534 1196 790μ, mm-1 0.787 0.678 0.448Radiation (λ, Å) MoKα (0.71073) MoKα (0.71073) MoKα (0.71073)Temperature, K 130(2) 130(2) 130(2)R indices [I > 2σ(I)] R1 = 0.0884, R1 = 0.0477, R1 = 0.1024 wR2 = 0.2073 wR2 = 0.1176 wR2 = 0.2659R indices (all data) R1 = 0.1060, R1 = 0.0556, R1 = 0.1647 wR2 = 0.2254 wR2 = 0.1235 wR2 = 0.3239Goodness-of-fit on F2 1.055 1.052 1.026


INTER-RING INTERACTIONS IN [FE(TALKYLP)(CL)] COMPLEXES 117

and that preferential orientation of unpaired spins was not taking place. M was corrected for the underlying porphyrin ligand diamagnetism according to previous experimentally observed values [38]; all remaining diamagnetic contri-butions ( dia) were calculated using Pascal’s constants [39, 40]. All measurements included a correction for the diamagnetic sample holder and the diamagnetic silicone grease.

RESULTS

The molecular structures of three meso-alkyl substituted (chloro)iron(III) porphyri-nate derivatives have been determined. The three iron porphyrin complexes differ in the length of the alkyl side chains with meso-substituentsof ethyl, [Fe(TEtP)(Cl)], n-propyl, [Fe(TPrP)(Cl)], and n-hexyl, [Fe(THexP)(Cl)]. A labeled ORTEP diagram of [Fe(TPrP)(Cl)] is shown in Fig. 1. Labeled ORTEP diagrams of [Fe(TEtP)(Cl)] and [Fe(THexP)(Cl)] can be found in Figs 2 and 3. Table 1 details selected crystallo-graphic details for all three complexes.

Bond distances and bond angles around the iron(III) atom in the three derivatives are listed in Table 2.

Figure 4a–c presents formal diagrams of the porphinato cores in [Fe(TEtP)(Cl)], [Fe(TPrP)(Cl)], and [Fe(THexP)(Cl)], respectively, displaying the perpendicular displace-ments of each atom from the 24-atom mean plane of the porphinato core. Figure 4 also contains the average bond lengths and bond angles (including standard deviations) for each complex. The porphyrin core of [Fe(TEtP)(Cl)] (Fig. 4a) is planar, while the porphyrin cores of [Fe(TPrP)-

(Cl)] and [Fe(THexP)(Cl)] are S4-ruffl ed (Fig. 4b,c).

Solid-state - dimer formation is seen in all three [Fe(TalkylP)(Cl)] complexes. Figure 5a–c shows edge views and top views of the closest interacting pair of porphyrin dimers for [Fe(TEtP)(Cl)], [Fe(TPrP)(Cl)], and [Fe(THexP)(Cl)], respectively. The inter-ring geometries for each dimer, including Fe···Fe and Ct···Ct distances, lateral shifts (L.S.), and mean plane separations (M.P.S.), are summarized in Table 3.

Temperature-dependent magnetic suscepti-bility measurements were carried out on the [Fe(TEtP)(Cl)] and [Fe(TPrP)(Cl)] complexes over the temperature range 6–300 K. Figure 6 shows the temperature-dependent magnetic moments for both complexes.

DISCUSSION

Crystal structures

The general trends in bond lengths and bond angles for all three [Fe(TalkylP)(Cl)] complexes are similar to those found in [Fe(TPP)(Cl)] [41] and [Fe(OEP)(Cl)] [42]. The coordination geometry of the iron cen-ters in the [Fe(TalkylP)(Cl)] complexes is characteristic of high-spin iron(III) porphy-rins. The equatorial Fe-Np bond lengths in

Fig. 1. Labeled ORTEP diagram of [Fe(TPrP)(Cl)]. Ellipsoids are drawn to illustrate 50% probability surfaces

Fig. 3. Labeled ORTEP diagram of [Fe(THexP)(Cl)]

Fig. 2. Labeled ORTEP diagram of [Fe(TEt)(Cl)]. Ellipsoids are drawn to illustrate 50% probability surfaces


118 M. LI ET AL.

Table 2. Select bond distance (Å) and angles (°) for [Fe(TalkylP)(Cl)] complexes

[Fe(TEtP)(Cl)] [Fe(TPrP)(Cl)] [Fe(THexP)(Cl)]

A. Bond lengths (Å)Fe(1)-N(1) 2.060(3) 2.063(1) 2.068(4)Fe(1)-N(2) 2.052(3) 2.054(1) 2.063(5)Fe(1)-N(3) 2.048(3) 2.055(1) 2.056(4)Fe(1)-N(4) 2.051(3) 2.062(1) 2.054(5)Fe(1)-Cl 2.2644(13) 2.2332(6) 2.2375(18)

B. Bond angles (deg)N(1)-Fe(1)-N(2) 86.66(14) 86.30(6) 87.32(17)N(1)-Fe(1)-N(4) 87.54(14) 86.88(6) 86.31(17)N(2)-Fe(1)-N(3) 87.49(14) 87.16(6) 86.41(17)N(3)-Fe(1)-N(4) 87.45(14) 86.50(6) 87.23(18)N(1)-Fe(1)-N(3) 154.47(14) 152.24(6) 152.14(17)N(2)-Fe(1)-N(4) 155.21(14) 152.31(6) 153.32(17)

Fig. 4. Unique 24-atom mean-plane diagrams of (a) [Fe(TEtP)(Cl)], (b) [Fe(TPrP)(Cl)] and (c) [Fe(THexP)(Cl)]



[Fe(TEtP)(Cl)], [Fe(TPrP)(Cl)], and [Fe(THexP)(Cl)] (2.053(5) Å, 2.058(8) Å, and 2.060(6) Å, respectively) fi t nicely into the category of typical high-spin Fe-N dis-tances ( 2.045 Å) [43]. A close inspection of bond angles reveals signifi cant differences in the Fe-N-C(a) angles when comparing [Fe(TEtP)(Cl)] to [Fe([Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] (Fig. 4). In [Fe(TEtP)(Cl)], the two types of Fe-N-C(a) angles are nearly equivalent [126.5(2)° and 125.7(4)°] and are similar to those found in [Fe(TPP)(Cl)]

[41] [126.9(8)° and 125.8(6)°] and [Fe(OEP)(Cl)] [42] [126.3(6)° and 126.5(5)°]. In contrast, [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] have two distinct types of Fe-N-C(a) angles with one of the angles being larger than the other: [128.1(2)° and 124.5(7)° for [Fe(TPrP)(Cl)] and 127.4(2)° and 124.9(3)° for [Fe(THexP)(Cl)]]. The differences in Fe-N-C(a) angles are apparently the result of the change in macrocyclic conformations among these complexes. In [Fe(TEtP)(Cl)], the macrocycle is basically planar, as

Fig. 5. Edge views and top views of the dimeric units of (a) [Fe(TEtP)(Cl)], (b) [Fe(TPrP)(Cl)] and (c) [Fe(THexP)(Cl)]. Ellipsoids are drawn to illustrate 30% probability surfaces for [Fe(TEtP)(Cl)], 65% probability surfaces for [Fe(TPrP)(Cl)], and 30% probability surfaces for [Fe(THexP)(Cl)]


120 M. LI ET AL.

Table 3. Summary of inter-ring geometries for [Fe(TalkylP)(Cl)] complexesa

Complex Fe· · ·Fe Ct· · ·Ctb M.P.S.c L.S.d Group Δe

[Fe(TEtP)(Cl)] 5.56 5.04 3.34 3.79 I 0.48[Fe(TPrP)(Cl)] 6.44 5.64 3.66 4.29 W 0.57[Fe(THexP)(Cl)] 6.36 5.77 3.64 4.48 W 0.55

a Values in Å. b Ct is the center of the 24-atom porphyrin ring. c The average mean plane separation for the two 24-atom cores of the dimer. d The lateral shift between the two 24-atom cores of the dimer. e The out-of-plane displacement relative to the 24-atom mean plane.

evidenced by an average deviation of the macrocycle car-bon atoms from the 24-atom mean plane of 0.03 Å, and a maximum displacement of 0.06 Å (Fig. 4a). The macro-cycles of [Fe(TPrP(Cl)] and [Fe(THexP)(Cl)] are ruffl ed to reduce steric interactions between the meso-alkyl and neighboring pyrrole groups. The average and maximum deviations of the macrocycle carbon atoms from the 24-atom mean plane of [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] are 0.18 Å and 0.38 Å, and 0.15 Å and 0.33 Å, respec-tively. Thus, ruffl ing of the macrocycles of [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] contributes to the signifi cant differ-ences in Fe-N-C(a) angles found in these two complexes. The effects of ring ruffl ing are also refl ected in the coor-dination environment of the iron centers; the out-of-plane displacement of the Fe atom in planar [Fe(TEtP)(Cl)]

(0.48 Å) is slightly smaller than those found in ruffl ed [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] (0.57 Å and 0.55 Å, respectively). Additionally, compared to planar [Fe(TEtP)(Cl)], an increase of the Fe-Np distances accompanied by a decrease of the Fe-Cl bond lengths is observed in ruffl ed [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] (see Table 2).

In addition to the differences in ring conformations, the inter-ring interactions among the [Fe(TalkylP)(Cl)] dimers are dramatically different. The lateral shift val-ues increase from ethyl to hexyl groups (Table 3). The lateral shift of [Fe(TEtP)(Cl)] is 3.79 Å, which falls into the intermediate (I) group, as defi ned by Scheidt and Lee [3], whereas the dimeric units of [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] show weaker interactions (Group W) with larger lateral shifts (4.29 Å and 4.48 Å, respec-tively). The differences in lateral shifts among these complexes can be attributed to the orientation of the meso-substituents of the porphyrin rings. Each porphyrin in [Fe(TEtP)(Cl)] has two ethyl groups pointing up and two pointing down in an orientation as to favor inter-ring overlap. However, [Fe(TPrP)(Cl)] and [Fe(THexP)(Cl)] have alternating up and down meso-substituents which leads to less favorable inter-ring overlap.

Magnetic susceptibility measurements

Temperature-dependent magnetic susceptibility mea-surements were carried out on [Fe(TEtP)(Cl)] and [Fe(TPrP)(Cl)] to investigate the differences in magnetic exchange between the high-spin iron(III) metal centers of these porphyrinate dimers. To date, high-spin ferric porphyrin complexes have been thoroughly investigated by using physical techniques such as EPR, Mössbauer, and NMR; however, relatively few detailed and accu-rate magnetic susceptibility measurements of high-spin

Fig. 6. Comparison of observed and calculated values of μeff/monomer vs. T for (a) [Fe(TEtP)(Cl)] and (b) [Fe(TPrP)(Cl)]. The solid lines are model calculation with the parameters-axial zero-fi eld splitting D = 6.95 cm-1, anti-ferromagnetic coupling with J = -0.45 for (a) and J = -0.17 cm-1 for (b)

Table 4. Summary of inter-ring coupling for various iron dimersa

Complex Fe· · ·Fe M.P.S.b L.S.c J, cm-1 Reference

[Fe(TPP)(B11CH12)]·C7H8 5.49 3.83 3.67 +3.0 d[Fe(OEP)(2-MeHIm)]ClO4·CH2Br2 4.28 3.31 1.49 +0.85 e[Fe(OEP)(2-MeHIm)]ClO4·CHCl3 4.60 3.42 2.15 +0.8 e[Fe(TEtP)(Cl)] 5.56 3.34 3.79 +0.45 this work[Fe(TPrP)(Cl)] 6.44 3.66 4.29 +0.17 this work

a Values in Å. b The average mean plane separation for the two 24-atom cores of the dimer. c The lateral shift between the two 24-atom cores of the dimer. d Inorg. Chem. 1987; 26: 3022–3030. Sign adjusted to +JS�1 · S�2 J. Chem. Phys.1985; 83: 5945–5952. Sign adjusted to + J S�1 · S�2



mononuclear iron(III) porphyrins have been carried out over a wide range of temperatures. For example, magne-tization studies of ferric tetraphenylporphyrins were car-ried out at fi eld strengths of up to 50 kG but were only studied between a narrow temperature range (2–20 K) [44]. Original magnetic susceptibility data for [Fe(OEP)(Cl)] [42] were reported at only a few temperatures and showed large scatter; recently, more accurate data for this compound has been reported from 4–100 K [46]. Although it is the lower temperature region that is most informative concerning the magnitude of the magnetic exchange interaction, large temperature ranges are help-ful in determining a model that more thoroughly describes the magnetochemistry of the complex.

The temperature-dependent magnetic susceptibility plot (6–300 K) for [Fe(TPrP)(Cl)] (Fig. 6) is character-istic of a high-spin iron(III) complex; the room temper-ature μeff value of 5.90 corresponds with the spin-only moment for a S = 5/2 center, and the zero-fi eld splitting parameter (6.95 cm-1) is typical for this class of com-pounds [47]. The coupling constant derived from fi tting the magnetic data for [Fe(TPrP)(Cl)] implies that there is an anti-ferromagnetic magnetic exchange interaction J S�1 · S�2 between iron centers of J = +0.17 cm-1. The temperature-dependent magnetic susceptibility plot for [Fe(TEtP)(Cl)] shows similar high-spin iron(III) char-acteristics (assuming a zero-fi eld splitting parameter identical to that found in the [Fe(TPrP)(Cl)] case), but is best described as having a magnetic exchange cou-pling constant about two times larger than that found in [Fe(TPrP)(Cl)].

As can be seen in Table 3, there is a relatively large dif-ference between the intermolecular geometric parameters of [Fe(TPrP)(Cl)] and [Fe(TEtP)(Cl)]. Most importantly, the Fe···Fe separation in [Fe(TEtP)(Cl)] (5.60 Å) is much smaller than that found in [Fe(TPrP)(Cl)] (6.44 Å). Also, as can be seen in Fig. 5a and b, the porphyrin overlap pattern is different for these two complexes. Thus, there is apparently a correlation between proximity of the iron centers of the dimeric unit and magnitude of magnetic exchange for these two complexes; stronger magnetic interaction between the iron centers of [Fe(TEtP)(Cl)] is most likely due to the closer proximity of the magnetic centers within the dimeric unit.

Thus, magnetic exchange coupling is found to be sen-sitive to modest structural changes between porphyrin dimer geometries, and this exchange interaction must be included in any detailed analysis of the data. In fact, it was reported that, in [Fe(OEP)(2-MeHIm)]ClO4 [45], consideration of this anti-ferromagnetic interaction was needed for a good fi t with the experimental data. Simi-larly, a small but signifi cant anti-ferromagnetic exchange between iron centers has been noted in [Fe(TPP)(Cl)] (J =+0.14 cm-1) [46], [Fe(OEP)(Cl)] (J = +0.02 cm-1) [46] and hemin chloride (J = +0.08 cm-1) [47]. It should be noted that these coupling parameters are determined assuming a mean fi eld model J S�1 · S�2. rather than a dimer-only

coupling scheme J S�1 · S�2 thus care must be taken in comparing the coupling strengths.

SUMMARY

We have prepared a series of meso-(tetraalkylporphinato)iron chloride complexes (alkyl = ethyl, n-propyl and n-hexyl). Their molecular structures indicate that inter-ring interactions change with alteration of the alkyl group substitutions in the meso positions. Magnetic exchange couplings between iron centers of the dimers are in accord with the trends in structural inter-ring geometries.

Acknowledgements

We thank the National Institutes of Health for support of this research under Grant GM-38401 to WRS. Funds for the purchase of the FAST area detector diffractome-ter was provided through NIH Grant RR-06709 to the University of Notre Dame. We also thank the National Science Foundation for the purchase of the SQUID equipment under Grant DMR-9703732.


The CIF fi les for the crystal structures were depos-ited at the Cambridge Crystallographic Data Centre (CCDC). The CCDC deposition numbers of [Fe(TEtP)(Cl)], [Fe(TPrP)(Cl)], and [Fe(THexP)(Cl)] are 744649, 744651 and 744650, respectively. Copies can be obtained on request, free of charge, via www.ccdc.cam.ac.uk/conts/retrieving.html or from the Cambridge Crystallo-graphic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44 1223-336-033 or email: [email protected]).

REFERENCES

1. a) Kuhlbrant W, Wang DN and Fujiyoshi Y. Nature (London) 1994; 367: 614–621. b) McDermott G, Prince SM, Freer AA, Hawthorn-Thwaite-Lawless AM, Papiz MZ, Cogdell RJ and Isaacs NW. Nature(London) 1995; 374: 517–521.

2. The Photosynthetic Reaction Center, Vol. 2, Deisen-hofer J and Norris JR. (Eds.) Academic Press: New York, 1993.

3. Scheidt WR and Lee YJ. Struct. Bonding (Berlin)1987; 64: 1–70.

4. a) Scheidt WR, Geiger DK, Lee YJ, Reed CA and Lang G. J. Am. Chem. Soc. 1985; 107: 5693–5699. b) Chen B, Safo MK, Orosz RD, Reed CA, Deb-runner PC and Scheidt WR. Inorg. Chem. 1994; 33:1319–1324.

5. a) Scheidt WR, Cheng B, Haller KJ, Mislankar A, Rae AD, Reddy KV, Song H, Orosz RD, Reed CA, Cukiernik F and Marchon JC. J. Am. Chem. Soc. 1993; 115: 1181–1183. b) Scheidt WR,


122 M. LI ET AL.

Brancato-Buentello KE, Song H, Reddy V and Cheng B. Inorg. Chem. 1996; 35: 7500–7507. c) Brancato-Buentello KE, Kang SJ and Scheidt WR. J. Am. Chem. Soc. 1997; 119: 2839–2846.

6. Brancato-Buentello KE and Scheidt WR. Angew. Chem., Int. Ed. Engl. 1997; 36: 1456–1459.

7. Song Y, Haddad RE, Jia S-L, Hok S, Olmstead MM, Nurco DJ, Schore NE, Zhang J, Ma J-G, Smith KM, Gazeau S, Pécaut J, Marchon J-C, Medforth CJ and Shelnutt JA. J. Am. Chem. Soc. 2005; 127:1179–1192.

8. Haddad RE, Gazeau S, Pécaut J, Marchon J-C, Medforth CJ and Shelnutt JA. J. Am. Chem. Soc.2003; 125: 1253–1268.

9. Ikeue T, Ohgo Y and Nakamura M. Chem. Com-mun. 2003; 220–221.

10. Ikeue T, Ohgo Y, Saitoh T, Yamaguchi T and Naka-mura M. Inorg. Chem. 2001; 40: 3423–3434.

11. Ikeue T, Ohgo Y, Uchida A, Nakamura M, Fujii H and Yokoyama M. Inorg. Chem. 1999; 38:1276–1281.

12. Nakamura M, Ikeue T, Fujii H, Yoshimura T and Tajima K. Inorg. Chem. 1998; 37: 2405–2414.

13. Nakamura M, Ikeue T, Fujii H and Yoshimura T. J. Am. Chem. Soc. 1997; 119: 6284–6291.

14. Ikeue T, Ohgo Y, Saitoh T, Nakamura M, Fujii H and Yokoyama M. J. Am. Chem. Soc. 2000; 122:4046–4076.

15. Ema T, Senge MO, Nelson NY, Ogoshi H and Smith KM. Angew. Chem., Int. Ed. Engl. 1994; 33:1879–1881.

16. Gallucci JC, Swepston PN and Ibers JA. Acta Crys-tallogr., Sect. B 1982; 38: 2134–2139.

17. Kutzler FW, Swepston PN, Yellin ZB, Ellis DE and Ibers JA. J. Am. Chem. Soc. 1983; 105:2996–3004.

18. Newcomb TP, Godfrey MR, Hoffman BM and Ibers JA. J. Am. Chem. Soc. 1989; 111: 7078–7084.

19. Jentzen W, Turowska-Tyrk I, Scheidt WR and Shel-nutt JA. Abstract of 210th ACS National Meeting; Inorganic Division, 531. August 1995.

20. Nakamura M, Ikeue T, Fujii H and Yoshimura T. J. Am. Chem. Soc. 1997; 119: 6284–6291.

21. Nakamura M, Ikeue T, Fujii H, Yoshimura T and Tajima K. Inorg. Chem. 1998; 37: 2405–2414.

22. Wolowiec S, Latos-Grazynski L, Toronto D and Marchon J-C. Inorg. Chem. 1998; 37: 724–732.

23. Senge MO, Medforth CJ, Forsyth TP, Lee DA, Olm-stead MM, Jentzen W, Pandey RK, Shelnutt JA and Smith KM. Inorg. Chem. 1997; 36: 1149–1163.

24. Meot-Ner M and Adler AD. J. Am. Chem. Soc.1975; 97: 5107–5111.

25. Song XZ, Jentzen W, Jia SL, Jaquinod L, Nurco DJ, Medforth CJ, Smith KM and Shelnutt JA. J. Am. Chem. Soc. 1996; 118: 12975–12988.

26. DiMagno SG, Wersching AK and Ross ICR. J. Am. Chem. Soc. 1995; 117: 8279–8280.

27. Mazzanti M, Marchon J-C, Shang M, Scheidt WR, Jia S and Shelnutt JA. J. Am. Chem. Soc. 1997; 119:12400–12401.

28. DiMagno SG, Williams RA and Therien MJ. J. Org. Chem. 1994; 59: 6943–6948.

29. Neya S and Funasaki N. J. Heterocyclic Chem.1997; 34: 689–690.

30. Lindsey JS, Schreiman IC, Hsu HC, Kearney PC and Marguerettaz AM. J. Org. Chem. 1987; 52:827–836.

31. Adler AD, Longo FR, Kampas F and Kim J. J. Inorg. Nucl. Chem. 1970; 32: 2443–2445.

32. The structure of [Fe(TPrP)(Cl)] has been reported earlier. The structure was determined at 298 K, with temperature effects considered, there are no signifi -cant differences. Ohgo Y, Ikeue T and Nakamura M. Acta Crystallogr., Sect. C 1999; C55: 1817–1819.

33. Scheidt WR and Turowska-Tyrk I. Inorg. Chem.1994; 33: 1314–1318.

34. Sheldrick GM. Acta Crystallogr., Sect. A 1990; A46: 467–473.

35. Sheldrick GM. Acta Crystallogr., Sect. A 2008; A64: 112–122.

36. The process is based on an adaptation of the DIFABS [37] logic to area detector geometry by Karaulov: Karaulov AI, School of Chemistry and Applied Chemistry, University of Wales, College of Cardiff, Cardiff CF1 3TB, UK, personal communication.

37. Walker NP and Stuart D. Acta Crystallogr., Sect. A1983; A39: 158–166.

38. Sutter TPG, Hambright P, Thorpe AN and Quoc N. Inorg. Chim. Acta 1992; 195: 131–132.

39. Selwood PW. Magnetochemistry, Interscience: New York, 1956; Chapter 2.

40. Earnshaw A. Introduction to Magnetochemistry,Academic: London, 1968; Chapter 1.

41. Scheidt WR and Finnegan M. Acta Crystallogr., Sect. C 1989; C45: 1214–1216.

42. Ernst J, Subramanian J and Fuhrhop J-H. Z. Natur-forsch. 1977; 32A: 1129–1136.

43. Scheidt WR and Reed CA. Chem. Rev. 1981; 81:543–555.

44. a) Behere DV, Date SK and Mitra S. Chem. Phys. Lett. 1979; 68: 544–548. b) Behere DV and Mitra S. Indian J. Chem. 1980; 19A: 505–507.

45. Gupta GP, Lang G, Scheidt WR, Geiger DK and Reed CA. J. Chem. Phys. 1985; 83: 5945–5952.

46. Marathe VR and Mitra S. J. Chem. Phys. 1983; 78:915–920.

47. Richards PL, Caughey WS, Ebersacher H, Feher G and Malley MM. J. Chem. Phys. 1967; 47:1187–1888.

CONTENTS


The key role of peripheral substituents in the chemistry of phthalocyanines and their analo gs 1Evgeny A. Lukyanets† and Victor N. Nemykin

Syntheses and structural studies of 5-pentamethyl cyclopentadienyl rhodium(III) and iridium(III) 41complexes of a Schiff-base expanded porphyrin

Luciano Cuesta, Vincent M. Lynch and Jonathan L. Sessler*

Synthesis of near-infrared absorbed metal phtha locyanine with S-aryl groups at non-peripheral po sitions 47Keiichi Sakamoto*, Eiko Ohno-Okumura, Taku Kato and Hisashi Soga

Through space singlet energy transfers in light-harvesting systems and cofacial bisporphyrin dyads 55Pierre D. Harvey*, Christine Stern, Claude P. Gros and Roger Guilard*

Photooxidation of phenol derivatives using μ-(dihy droxo) dipalladium(II) bisporphyrin complex 64Yuko Takao*, Toshinobu Ohno, Kazuyuki Moriwaki, Fukashi Matsumoto and Jun-ichiro Setsune*

Morphology and radical reactions of Cu(II) and Co(II) sul fonated phthalocyanines covalently linked 69to poly (ethylene amide)

Gustavo T. Ruiz, Alexander G. Lappin and Guillermo Ferraudi*

Porphyrin self-assembled monolayers and photo dy namic oxidation of tryptophan 81Isabelle Chambrier, David A. Russell, Derek E. Brundish, William G. Love, Giulio Jori*,

M. Magaraggia and Michael J. Cook*

Effi cient synthesis of an A-B-C-tricycle fragment for a structural model of tolyporphin 89Bing C. Hu*, Wei Y. Zhou, Zu L. Liu, Chao J. Cai and Shi C. Xu

Synthesis and self-assembly of a novel cobalt(II) porphyrin lipoic acid derivative on gold 101Christoph S. Eberle, Ana S. Viana, Franz-Peter Montforts and Luisa Maria Abrantes*

Synthesis, characterization and electrochemistry of the novel metalloporphyrazines annulated with 108tetrathiafulvalene having pentoxycarbonyl subs tituents

Fengshou Leng, Ruibin Hou, Longyi Jin, Bingzhu Yin* and Ren-Gen Xiong*

Inter-ring interactions in [Fe(TalkylP)(Cl)] (alkyl = ethyl, n-propyl, n-hexyl) complexes: control 115by meso-subs ti tuted groups

Ming Li, Teresa J. Neal, Noelle Ehlinger, Charles E. Schulz* and W. Robert Scheidt*

Transfer of Copyright Agreement and Publishing Agreement

The transfer of copyright from the Author(s) should be explicitly stated to enable the Publisher to disseminate the Work to thefullest extent. The following transfer must be signed and returned to the Publisher’s offi ce. The work will not be published until reception of the signed agreement.

World Scientifi c Publishing Co Pte Ltd5 Toh Tuck Link,

Singapore 596224.Fax: +65 6467 7667 - Email: [email protected]

The work entitled: ______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

The Author(s) guarantees that the above article is the Author(s)’ original Work, and has never been published elsewhere before and that the text, the illustrations, and any other materials included in the article do not infringe upon any existing copyright.

The Author(s) transfers to JPP, during the full term of copyright, free of charge, the exclusive rights comprised in the copyrightof the Work, including but not limited to the right to publish the Work and the material contained therein throughout the world, in all languages, and in all media of expression now known or later developed, and to license or permit others to do so. However, the following rights are reserved by Author(s):

– the right to use, free of charge, all or part of this article in future Works of their own, such as books or lectures; – the right to reproduce the article for their own purposes provided the copies are not offered for sale; – the right to make copies of all or part of the Work for the Author(s)’ use in classroom teaching.

The Author(s) indemnifi es the Publisher and the Editors of the Journal of Porphyrins and Phthalocyanines for, and holds them harmless from, any loss, expense or damage occasioned by a claim or suit by a third party for copyright infringement, or any suit arising out of any breach of the foregoing warranties as a result of publication of his article.

Note: corrections, additions or deletions introduced at the proof stage which constitute a departure from the originally submitted and accepted work will be incorporated at the Authors’ expense.

Signature* ___________________________ Date: ________________________________

Printed name: ___________________________________________________________________

*If signed by only one author on behalf of all coauthors, the additional statement must be signed.

« I represent and warrant that I am authorized to execute this copyright on behalf of all the authors of the article referred above »

Signature ____________________________ Date: ________________________________

2010; 14: 1–122 doi: 10.1142/s108842461000188x contents...gustavo t. ruiz, alexander g. lappin and...

Documents