Download PDF
Synlett 2012(2): 171-184  
DOI: 10.1055/s-0031-1289894
ACCOUNT
© Georg Thieme Verlag Stuttgart ˙ New York

Fused Polycyclic Aromatic Compounds with Near Infrared Absorption and Emission

Chongjun Jiaoa, Jishan Wu*a,b
a Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
b Institute of Materials Research and Engineering, A*Star, 3 Research Link, Singapore 117602, Singapore
Fax: +6567791691; e-Mail: chmwuj@nus.edu.sg;

Further Information

Publication History

Received 18 July 2011
Publication Date:
23 November 2011 (online)

 PDF Download Permissions and Reprints All articles of this category

Biographical Sketches

Chongjun Jiao was born in Zhenjiang, P. R. of China, in 1981. He obtained his Bachelor degree in 2004 and his Master’s degree in 2007 both from Soochow University. Then he moved to the National University of Singapore in 2007 to pursue his PhD under supervision of Dr. Jishan Wu. His research focused on the design and synthesis of stable and soluble NIR dyes and their applications in materials science.
Jishan Wu was born in Hubei province, P. R. of China, in 1975. He received a Bachelor degree in Chemistry from Wuhan University (1997) and a Master’s degree in Polymer Sciences from Changchun Institute of Applied Chemistry, Chinese Academy of Sciences (2000). He did his PhD research at the Max-Planck Institute for Polymer Research under the supervision of Professor Klaus Müllen (2000-2004) and after graduation he worked as a project leader in the same group. He moved to the University of California at Los Angeles in 2005 and worked with Professor Sir J. Fraser Stoddart on supramolecular chemistry. Dr. Wu joined the Department of Chemistry of the National University of Singapore in 2007 as an Assistant Professor. So far, he has published more than 90 peer-reviewed articles and book chapters and his work received high ­citations (> 2100, H index = 26). He received the NUS Young Investigator Award (2007), the Singapore National Young ­Scientist Award (2010) and the Departmental Young Chemist Award (2011).

Abstract

In this account, we provide an overview of our own efforts in the exploration of fused polycyclic aromatic compounds with NIR absorption and emission.

1 Introduction

2 Rylene

3 Periacenes

3.1 Bisanthene

3.2 Peripentacene

4 Zethrenes

4.1 Zethrene

4.2 Heptazethrene

5 Hybrid Fused Systems

5.1 Fused Porphyrin

5.2 Fused BODIPY

6 Concluding Remarks

Key words

near-infrared dyes - periacenes - zethrenes - porphyrins - BODIPY

1 Introduction

Dye chemistry is one of the most explored areas in industrial organic chemistry, yet it still provides many exciting challenges and surprises. Current developments in the field of electronics and in the area of bioimaging have boosted interest in the development of next-generation functional dyes. Recently, there has been rising interest is the design and synthesis of so-called near-infrared (NIR) dyes, [¹-³] which function (absorption and/or emission) in the NIR spectral region ranging from 700 nm to 2000 nm owing to their diverse applications. For instance, for practical applications such as solar cells, the materials should have good light-harvesting capability not only in the UV-vis spectral range, but also at the NIR range given that sunlight possesses 50% of its radiation energy in the infrared region. On the other hand, biological samples have low background fluorescence signals, and a concomitant high signal-to-noise ratio in the NIR region. Moreover, NIR light can penetrate into sample matrices deeply due to low light scattering. Thus, NIR dyes are required for various advanced technologies, including high-contrast bio-imaging, [4] optical recording, [5] NIR laser filter, [6] NIR photography, [7] and solar cells. [8] Up to now, a comprehensive range of common NIR compounds (e.g., cyanines, polyenes, rare earth compounds, quantum dots) have been known, some of which have even been commercialized. However, many commercially available NIR dyes such as cyanine and polyene dyes suffer from inherent drawbacks due to their insufficient photostability. [9] In addition, some NIR dyes/pigments also suffer from poor solubility for practical applications. To circumvent such issues, design and synthesis of soluble and stable organic NIR dyes are highly desirable for organic chemists.

Apart from commercialized dyes, another extraordinary class of compounds, named polycyclic aromatic compounds such as polycyclic aromatic hydrocarbons (PAHs), porphyrins, and BODIPY (4,4-difluoro-4-bora-3a,4a-diaza-s-indacene), are of great significance due to their particular electronic and self-assembling properties, which can be exploited in organic electronic devices such as field-effect transistors and solar cells. [¹0-¹¹] Compared with traditional cyanine and polyene dyes, polycyclic aromatics usually exhibit excellent chemical stability and photostability. In addition, polycyclic aromatics hold advantages such as low toxicity and ease of functional and structural tunability over quantum dots. Therefore, polycyclic aromatic compounds appear to be promising candidates for NIR dyes. However, most polycyclic aromatics are only capable of capturing UV or visible light. Promotion of their absorptions into NIR region can normally be achieved by extension of the π-conjugation or by construction of a push-pull motif or in rare cases by quinoidization. [²a] All the three methods will definitely produce a decrease in the HOMO-LUMO band gaps, and a concomitant bathochromic shift of their absorption and emission bands. Another troublesome issue for polycyclic aromatic compounds is their poor solubility because of the strong intermolecular interaction between large planar π-systems [¹²] and this problem becomes more and more serious upon an increase of the π-conjugation and the molecular size. In order to resolve the solubility problem, the attachment of long alkyl chains and/or induced distortion from planarity of the aromatic core, in most cases, is necessary. [¹³]

Since the establishment of our laboratory in 2007, we have been working on the design and synthesis of fused polycyclic aromatic compounds with NIR absorption and emission. In this account, we will give an overview of our own efforts in the exploration of various soluble and stable NIR dyes, including rylene, periacene, zethrene and hybrid fused systems. Their synthetic routes and interesting properties will be summarized and highlighted. Besides the achievements of our research group in this field, we also include the fundamental introduction to related polycyclic aromatic compounds to provide readers with necessary background knowledge.

2 Rylenes

A representative family in dye chemistry are rylene and its derivatives due to their excellent chemical and photostability. [¹4] For instance, perylene bisimide has a brilliant red color with the absorption maximum at 540 nm and terrylene bisimide was reported to have green color with the absorption maximum at 650 nm. [¹5] For their higher homologues, namely quaterrylene [¹6] , pentarylene [¹7] and hexarylene [¹7] , the absorptions are bathochromically shifted into the NIR region with large molar extinction coefficients (Figure  [¹] ). A general problem for the higher-order rylene dyes is their poor solubility even though some bulky groups are attached to the peri positions of the terminal naphthalene units. Substitution by bulky phenoxy groups at the bay positions effectively improves solubility and processability of the higher rylene dyes and also results in a moderate bathochromic shift of their absorption spectra. [¹7-¹8] However, ring cyclization reaction, which is required for preparation of higher-order rylenes, sometimes suffered from the dealkylation of the phenoxy group under strong basic conditions. [¹8]

Figure 1 Structures of rylene

A different strategy recently developed to construct the rylene backbone is based on fused N-annulated perylene analogues reported by Wang’s group. [¹9] In parallel, we have also been working on poly(peri-N-annulated perylene) and its carboximide derivative 8 that can be regarded as perfect nanosized graphene ribbons containing nitrogen atoms annulated onto the armchair edge. [²0]

Scheme 1 Synthetic route towards bis-N-annulated quaterrylenebis(dicarboximide)

As shown in Scheme  [¹] , the key intermediate 3 was prepared based on selective nitration at bay position of perylene 1a and the subsequent annulation was promoted by triethyl phosphite. The attachment of a flexible aliphatic chain was then carried out at this stage in the presence of KOH and KI to give 4 and one bromine atom was subsequently introduced to the peri position to afford 5. Although perylene monoimide can be readily obtained by one-step reaction from the cheap perylene tetracarboxylic dianhydride, [²¹] there is no efficient way to synthesize an N-annulated perylene dicarboxylic imide (NPDI) such as 6. We and other researchers recently found that arene dicarboxylic anhydride could be prepared by Friedel-Crafts reaction of oxalyl chloride with active aromatic compounds (e.g. anthracene), followed by oxidation of the formed diketone to a carboxylic anhydride group in good yields. [²²] Herein, this method was used and the key intermediate compound 6 was prepared in a convenient synthetic route following sequential Friedel-Crafts reaction, oxidation, and imidization. Ni(cod)2-mediated Yamamoto homo-coupling then gave easy access to the perylene dicarboxylic imide dimer 7 in nearly quantitative yield.

Our next challenge was to find efficient approaches with the goal of accomplishing the ring-closure reaction. As a matter of fact, the last cyclization reaction was performed under different conditions such as (1) KOH in ethanol with glucose; [²³] (2) t-BuOK/DBN in diglyme; [²4] (3) FeCl3 in nitromethane and dichloromethane; [²5] and (4) K2CO3 in ethanolamine. [²6] It was found that the first three methods all failed and complicated mixtures were obtained probably due to the decomposition of NPDI units. Fortunately, method (4) via the mild base K2CO3 promoted cyclization worked well and our target compound 8 was obtained in 35% yield, probably following an arylide anion mechanism. Subsequent studies revealed that the yield of 8 is dependent on the concentration of K2CO3 in ethanolamine and a maximum yield of 70% could be achieved when the ratio of K2CO3 to ethanolamine is 1:3.

Highly electron-rich poly(peri-N-annulated perylene) has proven to be unstable in solution, [¹6] whereas dye 8 displays high thermal and photostability owing to the introduction of electron-withdrawing carboximide groups, which lowers the HOMO energy level and makes 8 very stable upon exposure to light as well as to oxygen. In common with unmodified quaterrylene dicarboxylic imide, the absorption maximum of 8 is situated at 760 nm. [¹6] However, dye 8 has remarkably good solubility (100 gl -¹, CH2Cl2), a high fluorescence quantum yield of 0.55 in dichloromethane, and a large molar extinction coefficient ( = 260000 M-¹cm-¹). All these properties are superior to those of other quaterrylene dyes.

Figure 2 Structures of bisanthene, peritetracene and peripentacene

3 Periacenes

Extension of the conjugation along the long molecular axis of perylene 1a to its higher homologues is well known and such an extension affords the higher-order rylene molecules with longer absorption in wavelength. Theoretical calculations indicate that laterally extended perylene derivatives (also called ‘periacenes’), namely bisanthene, peritetracene, peripentacene (Figure  [²] ), would also lead to good candidates for NIR-absorbing dyes in terms of absorption wavelength. [²7] Only two aromatic sextet benzenoid rings can be drawn for periacenes, thus the higher-order periacenes are expected to have a high chemical reactivity.

3.1 Bisanthene

The synthesis of bisanthene was first reported by Clar [²8] and the improved synthesis was recently developed by Bock’s s group in 2006. [²9] The parent bisanthene was a blue-color compound with absorption maximum at 662 nm in benzene. However, it showed very poor stability due to its high-lying HOMO energy level, allowing the addition reaction with singlet oxygen at the active meso-positions to take place. [³0]

The necessary research to obtain stable bisanthene derivatives was systematically conducted by our group applying different strategies. The first method is the attachment of electron-withdrawing dicarboxylic imide groups onto the zig-zag edges (Scheme  [²] ), [²²a] which is capable of lowering the relatively high lying HOMO level of bisanthene. In common with the preparation of NPDI 6, the key intermediate compound 13 was first prepared from anthracene by stepwise Friedel-Crafts reaction and oxidation. To obtain monobrominated 15, we then encountered the synthetic challenge of exclusive bromination of 13, and the formation of isomers of 14 was inevitable. Fortunately, separation of 15 from the other isomer was successful after imidization, as intermediate 15 exhibits enhanced solubility with respect to 14. Compound 15 then underwent Ni(cod)2-mediated Yamamoto homocoupling reaction to give anthracene dicarboxylic imide dimer 16. After base-promoted cyclization by t-BuOK and 1,5-diazabicyclo[4.3.0]non-5-ene (DBN), the fully fused bisanthene bis(dicarboximide) 17 was successfully synthesized in moderate yield (Scheme  [²] ). Compared with the parent bisanthene 9, compound 17 displayed good photostability and no significant change could be observed as a solution of 17 was exposed to air. In addition, a significant red shift of 300 nm at the absorption maximum was observed for 17 compared with perylene bisimide, suggesting the great effect of lateral extension on the absorption behavior.

Scheme 2 Synthetic route towards bisanthene bis(dicarboximide)

The second method is to introduce aryl or alkyne substituents onto the most reactive meso positions of the bisanthene (Scheme  [³] ) so as to not only stabilize the active bisanthene core but also cause an additional bathochromic shift to the NIR spectral region. [³¹] Based on these considerations, our group synthesized three meso-substituted bisanthenes 20a-c using nucleophilic addition of aryl or alkyne Grignard reagents to the bisanthenequinone 18 followed by reductive aromatization of the formed diol. Owing to the extended π-conjugation of the bisanthene core through the aryl and triisopropylsilylethynyl moieties in 20, the absorption maxima of 20a-c in toluene more or less shift to longer wavelength. In addition, solutions of 20a-c are stable for weeks under ambient conditions, showing higher stability than their parent bisanthene 9. More importantly, impressive photoluminescence quantum yields of 0.81, 0.80 and 0.38 were achieved for dyes 20a-c, respectively, endowing them with appropriate properties for practical applications.

Scheme 3 Synthesis of meso-substituted bisanthene derivatives

Scheme 4 Synthesis of bisanthene quinone

Generally speaking, the band gap of quinoidal π-conjugated systems is lower than that of their parent aromatic analogues. In this principle, quinoidization of PAHs thus may be a feasible option for obtaining compounds with longer wavelength absorption. In 2009 our group reported successful preparation of quinoidal bisanthene 23 (Scheme  [4] ), which can be regarded as a rare case of soluble and stable polycyclic aromatic compounds with a quinoidal character. [³²] The synthesis of this new dye 23 began with the reaction between Grignard reagent 21 and bisanthenequinone 18. The obtained alcohol 22 then underwent desilylation by tetrabutylammonium fluoride (TBAF) and dehydration of the formed phenol by POCl3 in pyridine to give the desired quinoidal bisanthene 23. A solution of 23 in dichloromethane displayed well-resolved absorption bands between 500 and 800 nm with the absorption maximum at 690 nm and is very stable at ambient conditions upon exposure to UV light for several days. Interestingly, quinoidal bisanthene 23, showing good electrochemical amphotericity with four reversible one-electron transfer steps, can both be oxidized by an oxidant and be reduced by a reductant to form four stabilized singly and doubly charged species.

Scheme 5 Synthesis of laterally extended bisanthene

Since bisanthene has a diene character at the bay positions, the oxidative Diels-Alder reaction taking place along the bay position of bisanthene by using dienophiles therefore allows to ‘grow’ the bisanthene core. [²9] [³³] Particularly, this design concept was successfully applied to construct laterally extended bisanthene adducts 27 and 28. [³4]

The synthesis of target compounds 27 and 28 is depicted in Scheme  [5] . Bisanthene 20a initially underwent one-fold oxidative Diels-Alder reaction with 1,4-naphthaquinone 24 in refluxing nitrobenzene for 24 h to generate bisanthene adduct 25. In contrast, prolongation of the reaction time to 2 d could lead to the formation of 2:1 cycloadduct 26. Treatment of 25 and 26 with the Grignard reagent of 3,5-di-tert-butylbromobenzene followed by reductive aromatization with NaI/NaH2PO2 gave the desired 27 and 28, respectively. Alternatively, compound 28 also can be prepared from 27 that subsequently underwent the next round reaction sequence of Grignard reaction, and reductive aromatization.

Early studies by Müllen and coworkers revealed that bay annulation of rylenes leads to hypsochromic shifts of absorption compared to their parent rylene molecules. [³5] In our case, owing to the additionally higher conjugation along the long molecular axis in 27 and 28 than in 20a, the ‘net’ effect is that the absorption spectra of 27 and 28 in toluene display a slight bathochromic shift with both absorption maxima at 696 nm. Interestingly, no bathochromic shift was observed for the more π-extended compound 28 compared with compound 27. This can be explained by the more twisted structure in 28 and the presence of more sextet rings in compound 28. In addition, the emission spectra of 27 and 28 showed the mirror symmetry with their absorption spectrum, and relatively high photoluminescence quantum yields of 0.45 and 0.38 were measured for dyes 27 and 28, respectively. Such data are unusual for fused polycyclic aromatic compounds.

Scheme 6 Preliminary studies towards peripentacene

3.2 Peripentacene

Theoretical calculations by Jiang et al. showed that HOMO-LUMO gap changes for the periacenes drop quickly with the increase in molecular size from 1.87 eV for perylene to 0.08 eV for peripentacene. [³6] As a consequence, synthesis of higher periacenes is extremely challenging and little chemistry has been developed for this purpose. The preliminary studies towards peripentacene were carried out in our group. [³7-³8] As shown in Scheme  [6] , the bispentacenequinone 30, obtained from dimerization of pentacenyl monoketone 29, was subjected to nucleophilic addition of triisopropylsilylethynyl lithium reagent followed by reductive aromatization with NaI/NaH2PO2 to give the cruciform 6,6′-dipentacenyl 31. Compound 31 exhibits two face-to-face π-stacking axes in the single crystal and this allows two-directional isotropic charge transport. FET mobilities up to 0.11 cm²V- ¹s- ¹ were obtained based on vapor-deposited thin films. [³7] In addition, fused bispentacenequinone 32, which can be regarded as a precursor for the synthesis of peripentacene derivatives, was prepared with formation of two new C-C bonds via oxidative photocyclization of 30.³8 However, the subsequent nucleophilic reaction of compound 32 with excess Grignard reagent of 1-bromo-3,5-di-tert-butylbenzene in anhydrous THF followed by acidification in air did not generate the desired 1,2-addition adduct. Alternatively, an unexpected Michael 1,4-addition product 33 was obtained and confirmed by single-crystal analysis. Further treatment of 33 with excess Grignard reagent followed by acidification in air gave the tetraaryl-substituted fused bispentacenequinone 34. Single-crystal analysis reveals that there are α,β-unsaturated ketone structures in the fused bispentacenequinones 33 and 34, which may explain the unusual Michael additions.

4 Zethrenes

Zethrenes, which are not as well investigated as the other aromatic compounds such as periacenes and rylenes, can be regarded as an interesting class of polycyclic aromatic compounds in which the two naphthalene units are fused with two or more hexagonal rings into a Z-shape (Figure  [³] ). Theoretical calculations predicted that zethrenes also have diradical character at ground state and they will show interesting non-linear optical properties and near-infrared absorption. [³9] However, successful syntheses of zethrene and its derivatives were seldom reported due to their low accessibility and high sensitivity in the presence of oxygen and light, especially in dilute solution.

Figure 3  Resonance structures of zethrene and heptazethrene

Scheme 7 Synthesis of zethrene bisimide

4.1 Zethrene

Recent progress in the synthesis and characterization of zethrene was made by Tobe et al. [40] and Y.-T Wu et al. [4¹] Inspired by our previous studies, we independently reported electron-withdrawing dicarboximide groups substituted zethrene derivative 38, which is expected to display superior chemical as well as photostability and bathochromic shift of absorption spectra to the far-red and NIR region. [4²] In the design of zethrene 38, 5,6-dibromoacenaphthenequinone 35 was chosen as starting material, which was transformed into 4,6-dibromo-1,8-naphthalimide 36 following oxidation and subsequent imidization. Compound 38 was then synthesized in a one-pot Stille cross-coupling between 36 and bis(tri-n-butylstannyl)acetylene followed by simultaneous transannular cyclization of the intermediate compound 37 (Scheme  [7] ). As expected, the obtained zethrene bisimide 38 displays advantages over unmodified zethrene, including good chemical stability and photostability with half-life times determined as 4320 min under irradiation of ambient light, red-shifted absorption with the absorption maximum at 648 nm and enhanced fluorescence quantum yield up to 54%. Attempted bromination at the 7,14-positions by NBS in DMF gave the oxidized product zethrenequinone 39 probably due to the butadiene character of the central fixed double bonds. However, no spectroscopic evidence could be found to support the diradical feature in zethrene 38.

4.2 Heptazethrene

An exciting breakthrough in the field of zethrene chemistry has been made very recently and the higher homolog in the zethrene family, that is, heptazethrene, has been synthesized and well characterized for the first time by our group. [4³] The design consideration of 43 is still based on our early studies that electron-withdrawing dicarboximide with bulky diisopropylphenyl group can effectively lower the HOMO energy level of the zethrene core and suppress dye aggregation. As shown in Scheme  [8] , successive coupling between 36 with trimethylsilylacetylene (TMSA) and then triisopropylsilylacetylene (TIPSA) gave compound 40 under standard Sonogashira reaction conditions. After selectively removing the TMS group, oxidative coupling was adopted to generate 41. It is found that the careful choice of alkyne coupling conditions is of great importance at this stage since the general Glaser conditions using oxygen as oxidant led to 41 in extremely low yield whereas palladium-promoted coupling reaction with p-benzoquinone as oxidant in the absence of oxygen proved to be effective to increase the isolated yield of 41. Based on these findings, subsequent deprotection and intramolecular alkyne coupling using the same palladium-catalyzed conditions afforded the key intermediate 42, which eventually underwent simultaneous transannular cyclization to give the target compound 43 as dark green solid.

Scheme 8 Synthesis of heptazethrene bis(dicarboximide)

The absorption spectrum of 43 showed four bands from the near-infrared to the far-red spectral region with maxima at 827, 747, 701 and 641 nm. Noteworthy is that 43 exhibits a singlet biradical character in the ground state, which was evidenced by the combination of low-temperature ¹H NMR spectra and variable-temperature absorption spectral measurement. The impressive stability of the biradical form can be explained by the following reasons: (1) tendency of the central six-membered ring to maintain the aromaticity character; (2) stabilization of biradicals throughout the highly extended conjugation backbone, as well as by the electron-withdrawing imide groups.

5 Hybrid Fused Systems

5.1 Fused Porphyrin

Porphyrin is one of the most widely investigated nitrogen-containing macrocyclic aromatic compounds. Porphyrins and metalloporphyrins generally possess an intense absorption in the 400-450 nm region called the Soret band and relative weak absorptions in comparison to the Soret band in the 500-700 nm region named Q band. Considering their limited absorption behavior in the NIR region, efforts have been made towards the design and synthesis of π-extended porphyrins with low symmetry, including fused porphyrin tapes [44] and aromatic ring fused porphyrins. [45]

The singly linked porphyrin-aromatic ring dyads usually can be prepared by palladium-catalyzed Suzuki coupling reactions between appropriate aromatic compounds and porphyrin building blocks. The determining factor for the synthesis of aromatic ring-fused porphyrins is how to promote intramolecular cyclodehydrogenation of these dyads. There are three challenges for the aforementioned cyclization: (1) stability; (2) solubility; (3) limited knowledge of fusion reaction. Therefore, how to obtain stable and soluble NIR dyes based on hybrid fused porphyrin compounds remains a challenging target.

As a part of our efforts to synthesize polycyclic aromatic compounds related NIR dyes, our initial attempts were made by fusing the electron-rich N-annulated perylene (NP) unit to a porphyrin core, [46] given that the Sc(OTf)3-DDQ system has proven to be effective in promoting the ring fusion of electron-rich metalloporphyrins (e.g., Zn porphyrin) to the second electron-rich component. [47] Various substitutions such as bulky 4-tert-butylphenyl, 3,5-di-tert-butylphenyl, 3,5-di-tert-butylbenzyl, 2,6-diisopropylphenyl, and branched aliphatic chains were introduced because theses groups not only surmount the solubility problem but also suppress the aggregation of the chromophores in solution.

Scheme 9 Synthesis of NP-fused porphyrin

Our synthesis plan is illustrated in Scheme  [9] . Suzuki coupling between 44 and N-annulated perylene boronic ester 45, followed by subsequent Zn metalation gave the singly linked precursor. The combination of Sc(OTf)3 and DDQ was then applied to promote the ring closure and eventually led to the desired compound 46. Synthesis of 49 followed a similar synthetic route. However, when the branched alkyl chains (i.e., R¹ and R²) were used as the substituents, a complicated mixture was obtained after cyclodehydrogenation and the purification and characterization of the final product turned out to be very difficult because of the strong aggregation tendency of the obtained perylene-fused porphyrins. Fortunately, pure hybrid molecule 49 was successfully obtained when bulky 3,5-di-tert-butylphenyl (R³) and 3,5-di-tert-butylbenzyl (R4) groups were used.

The absorption maximum of 46 was found at 775 nm ( = 40000 M-¹cm-¹) while extension of the π-conjugation length of 46 along the long molecular axis by fusion of two porphyrin units with one N-annulated perylene in 49 obviously demonstrated a more red-shifted and intense NIR absorption with absorption maximum at 981 nm ( = 120000 M-¹cm-¹). Both NP-fused porphyrin compound 46 and bis-porphyrin fused N-annulated perylene 49 showed detectable NIR photoluminescence with emission maxima at 800 nm and 982 nm, respectively. It is noteworthy that photoluminescence quantum yield was measured as 5.6% for compound 46. Considering the relatively low photoluminescence quantum yield of porphyrin tapes and other fused porphyrin compounds, [44] [45] [48] such an enhancement in the NIR absorption/emission of 46 and 49 is impressive.

Scheme 10 Synthesis of perylene monoimide-fused porphyrin

Despite the narrow band gap, dyes 46 and 49 can be stored as a solids for more than several months, but their solutions gradually decompose upon UV irradiation. In order to develop more stable porphyrin-based NIR dyes, perylene monoimide, possessing a strong electron-withdrawing imide group, was considered to be a rational building block to replace the relatively electron-rich N-annulated perylene unit in the perylene-fused porphyrin compounds. Hence, different reagents are required to be used for cyclization due to the different electronic properties of perylene segments. Our approach towards perylene monoimide fused porphyrins utilized an oxidative dehydrogenation protocol that makes use of FeCl3 as Lewis acid as well as oxidant. [49] The synthesis started from the Suzuki coupling of perylene monoimide boronic ester 50 and porphyrin monobromide 44 (Scheme  [¹0] ). With the requisite intermediate 51 in hand, we investigated the effect of the metal center and found that the presence of Ni in the porphyrin is necessarily based on the following considerations: (1) matching the energy level of the porphyrin core to that of perylene monoimide so that fusion of these two subunits becomes possible; (2) avoiding demetalation during the FeCl3-promoted oxidative cyclodehydrogenation reaction. The general oxidative dehydrogenation conditions using 2 or 3 equivalents FeCl3 at room temperature did not result in cyclized product, while cyclization of 51 with 10-fold excess FeCl3 at reflux conditions simultaneously generated doubly and triply linked porphyrin-perylene monoimide 52 and 53 in a one-pot reaction. To the best of our knowledge, this is the first example to demonstrate that not only the peri positions but also the meta position of perylene monoimide can be involved in the cyclodehydrogenation to form a highly π-conjugated system.

Scheme 11 Synthesis of perylene anhydride fused porphyrin for DSCs

Both 52 and 53 show broad absorption spectra that cover the entire visible and a part of the NIR spectral regions with absorption maxima at 803 nm ( = 91000 M-¹cm-¹) for 52 and 897 nm ( = 59000 M-¹cm-¹) for 53. More importantly, air-saturated solutions of 52 and 53 show no significant changes in their absorption spectra upon exposure to sunlight for months. Even under irradiation of UV light for 48 h, 95% of their initial optical density remains unchanged. In contrast, the half-lives of 46 and 49 were only estimated as 244 and 547 min, respectively, under the same UV irradiation conditions.

The extraordinary stabilities of 52 and 53 provide possibilities for practical applications. In subsequent studies, preliminary investigation into dye-sensitized solar cells (DSCs) [50] taking advantage of perylene-fused porphyrin backbone as stable NIR dyes was carried out. [5¹] Although attempts to use these fused dyes in DSCs were independently made by Yeh’s and Imahori’s groups, [5²-5³] DSCs based on fused porphyrin systems exhibiting an NIR response beyond 900 nm have never been reported. In the design of these dyes in terms of applications for DSCs, appropriate structural modifications are necessary, including (1) saponification of the imide group in the presence of a strong base to form anhydride that serves as an anchoring group; [54] (2) replacement of aliphatic chains with bulky 3,5-di-tert-butylphenyl groups to eliminate the aggregation of the chromophores beneficial for DSCs; [50d] and (3) introduction of an electron-donating 4-(dimethyl­amino)phenylethynylene onto the meso position of porphyrin in 63 to result in a more red-shifted NIR absorption and a concomitant fast electron injection from the excited dye to the conduction band of TiO2 in DSCs.

The synthesis of 56 commenced with the preparation of the perylene monoimide fused porphyrin, via a synthetic route similar to the one used for 52 and 53. The imide group was eventually saponified to the corresponding anhydride in the presence of KOH. Preparation of electron-donating 4-(dimethylamino)phenylethynylene substituted 63 followed a slightly revised procedure (Scheme  [¹¹] ). The key intermediate 59 was f irst prepared by sequential Miyaura reaction, Suzuki coupling, bromination and trans-metallation. FeCl3-promoted oxidative dehydrogenation of 59 then afforded fused 60 together with some chlorinated side products. The presence of a bromine atom in 60 allowed us to further functionalize this hybrid molecule by means of palladium-catalyzed coupling reactions and the incorporation of an electron-donating 4-(dimethylamino)phenylethynylene group was conducted at this stage by Sonogashira-Hagihara coupling reaction to generate ‘push-pull’-type molecule 62, which finally underwent saponification to give target molecule 63.

In common with 52 and 53, the absorption spectra of both 56 and 63 cover the entire visible and a part of the NIR spectral region. When dyes 56 and 63 were utilized for DSCs, broader IPCE spectra covering the entire visible range and even extention into the NIR region up to 1000 nm were achieved. Nevertheless, moderate power conversion efficiencies of 1.26% and 1.36% were measured for 56 and 63, respectively. Such low efficiencies might be accounted for by the mismatch of LUMO energy levels of dyes with the Fermi level of TiO2 and by strong aggregation of dyes. These overall results illustrate that creating NIR-absorbing organic dyes that provide high efficiency remains a challenging task. As a result, not only absorption coverage, but also the other aspects, including energy level, suppression of aggregation and lifetime of excitation, have to be optimized.

5.2 Fused BODIPY

In pursuit of stable NIR dyes based on polycyclic aromatic compounds, BODIPY is an outstanding candidate due to its unusual and remarkable properties, such as high fluorescence quantum yield, large molar extinction coefficients, outstanding chemical, thermal and photochemical inertness. [¹¹c] [55] Three approaches have been reported with the aim of promoting the absorption and emission of BODIPY into far-red and even NIR region, currently including: (1) extension of π-conjugation by fusing a rigid ring to the pyrrole unit; [56] (2) functionalization at the α- and/or meso position to generate a ‘push-pull’ motif; [57] and (3) replacement of the meso-carbon atom with a nitrogen atom. [58] Besides these, our group recently has reported for the first time that fusion of polycyclic aromatic compounds (i.e., perylene, porphyrin) onto the zig-zag edge (i.e., meso and β-positions) of a BODIPY core resulted in a series of fused BODIPY derivatives 70, 74 and 78 as NIR dyes. [59-60]

Scheme 12 Synthesis of fused BODIPYs

As we have been studying fused polycyclic aromatic compounds as NIR dyes in the past several years, we envisioned that use of the bulky groups on polycyclic aromatic compounds is crucial for the purpose of suppressing aggregation. Furthermore, it was found that the presence of such groups is even necessary in the BODIPY moiety during our studies. Based on these considerations, we initiated a synthetic strategy aiming at the preparation of 3,5-di-tert-butylphenyl substituted pyrrole 66. Scheme  [¹²] depicts the synthetic route that was used for preparation of compound 66. Selective bromination of Boc-protected pyrrole 64 followed by Suzuki coupling of Boc-protected bromopyrrole with arylboronic ester gave the protected 2-arylpyrroles 65. Removal of the Boc protecting group by using TFA generated a significant amount of decomposed compounds. Sodium methoxide, however, gently removed the protection group [6¹] and afforded the key building block 66. On the other hand, the perylene aldehyde 68 was prepared by lithiation of monobrominated N-annulated perylene 67 followed by reaction with anhydrous DMF. TFA-catalyzed condensation of the obtained aldehyde 67 with 2 equivalents 66 led to the corresponding dipyrromethane derivative in good yield. Subsequent oxidation by DDQ and complexation with BF3˙OEt2 afforded the N-annulated perylene-BODIPY dyad 69. The final step was carried out by using FeCl3 to promote intramolecular ring fusion of 69 to give perylene-fused BODIPY 70. Fusing other polycyclic aromatic compounds such as porphyrin to the BODIPY core was attempted as well. The singly linked porphyrin-dipyrromethane dyad 72 was prepared by condensation of pyrrole derivative 66 with porphyrin aldehyde 71 promoted by p-toluene sulfonic acid monohydrate (TsOH˙H2O) rather than TFA. In our case, only a small amount of desired product 72 could be detected under general TFA-catalyzed conditions probably owing to the steric hindrance coming from the porphyrin subunit. While TsOH×H2O was found to accelerate such an acid-promoted transformation to generate 72 and this strategy also reduced the degree of side reactions formation. Oxidation of 72 by DDQ and subsequent complexation with BF3˙OEt2 afforded the porphyrin-BODIPY dyad 73. FeCl3 that has been successfully used to prepare numerous fused compounds was utilized herein and the synthesis of the fully fused product 74 was achieved in 72% yield. Bis-BODIPY fused porphyrin 78 with higher conjugation was thus synthesized in an overall yield of 34% in three steps, via the same synthetic route used for 74. Due to the requirement to form more C-C bonds in 78 in comparison to 74, increasing the reaction temperature turned out to be effective to promote intramolecular ring fusion to completion and the fully fused compound 78 was eventually separated in 15% yield.

Owing to the extended conjugation results from effective fusion, compounds 70, 74 and 78 displayed remarkable NIR absorption behaviors. It is worth noting that bis-­BODIPY fused porphyrin 78 possesses NIR absorption deeply beyond 1000 nm, which is the longest NIR absorption maximum ever observed for all BODIPY derivatives. Furthermore, perylene-fused BODIPY 70 exhibited detectable NIR emission with the emission maximum at 830 nm. More importantly, all the fused BODIPY dyes exhibited excellent photostability. Comparison of fused BODIPY dyes 70 with the other homologous fused dyes, such as poly(peri-N-annulated perylene) and NP-fused porphyrin 46, revealed that the fused BODIPY unit is the most effective building block to stabilize the highly conjugated system so far. Such a crucial finding stimulates us to prepare more soluble and stable polycyclic aromatic compounds fused BODIPY dyes with tunable NIR absorption and emission.

6 Concluding Remarks

It is obvious that synthesis of organic dyes with NIR absorption and emission is among the most widely studied research areas in organic chemistry. This account is an attempt to deal with design concepts and synthetic methods towards a variety of NIR dyes based on fused polycyclic aromatic compounds that have been made over the past four years in our group. By the means of state-of-the-art designs and appropriate modifications, troublesome solubility and stability problems have been resolved for these highly conjugated systems. The fundamental structure-property relationships have also been presented. We hope that all the chemistry discussed in this account will stimulate fellow researches to boost further developments in this area.

Acknowledgment

The work mentioned in this account was made by former and present members of Jishan Wu’s group and our collaborators. All of their names are listed in the references below. We gratefully ­acknowledge financial support from Singapore DSTA DIRP Project (DSTA-NUS-DIRP/2008/03), NRF Competitive Research Program (R-143-000-360-281), NUS Young Investigator Award (R-143-000-356-101), A*Star BMRC-NMRC joint grant (no. 10/1/21/19/642) and IMRE Core Funding (IMRE/10-1P0509).

9

Gregory P. High-Technology Applications of Organic Colorants Plenum New York 1991

9

Gregory P. High-Technology Applications of Organic Colorants Plenum New York 1991

   Permissions and Reprints All articles of this category

Figure 1 Structures of rylene

Scheme 1 Synthetic route towards bis-N-annulated quaterrylenebis(dicarboximide)

Figure 2 Structures of bisanthene, peritetracene and peripentacene

Scheme 2 Synthetic route towards bisanthene bis(dicarboximide)

Scheme 3 Synthesis of meso-substituted bisanthene derivatives

Scheme 4 Synthesis of bisanthene quinone

Scheme 5 Synthesis of laterally extended bisanthene

Scheme 6 Preliminary studies towards peripentacene

Figure 3  Resonance structures of zethrene and heptazethrene

Scheme 7 Synthesis of zethrene bisimide

Scheme 8 Synthesis of heptazethrene bis(dicarboximide)

Scheme 9 Synthesis of NP-fused porphyrin

Scheme 10 Synthesis of perylene monoimide-fused porphyrin

Scheme 11 Synthesis of perylene anhydride fused porphyrin for DSCs

Scheme 12 Synthesis of fused BODIPYs