Open Access Article
This Open Access Article is licensed under a
Creative Commons Attribution 3.0 Unported Licence

Gold-catalyzed allene cycloisomerization for pyrrole synthesis: towards highly fluorinated BODIPY dyes

Linda Lempke a, Tobias Fischer b, Jérémy Bell b, Werner Kraus b, Knut Rurack *b and Norbert Krause *a
aOrganic Chemistry, Dortmund University of Technology, Otto-Hahn-Str. 6, D-44227 Dortmund, Germany. E-mail: norbert.krause@tu-dortmund.de; Fax: (+)49 231 755 3884
bDept. 1 Analytical Chemistry; Reference Materials, BAM Federal Institute for Materials Research and Testing, Richard-Willstätter-Str. 11, D-12489 Berlin, Germany. E-mail: knut.rurack@bam.de; Fax: (+)49 30 8104 1157

Received 26th December 2014 , Accepted 5th February 2015

First published on 5th February 2015


Abstract

A novel synthetic strategy toward highly fluorinated BODIPY dyes with exceptional photostabilities relying on sustainable gold catalysis has been developed. A key to the tailored pyrrole precursors is the gold catalysis performed in ionic liquids as the reaction medium, allowing a facile recycling of the catalysts. The dyes prepared are well-matching with the spectral windows of popular rhodamine dyes and possess high brightness while showing a distinctly higher photostability than the rhodamines especially in aprotic solvents.


Introduction

Sustainable metal catalysis has attracted general attention during the last few years. Meanwhile also for gold catalysis powerful sustainable methods are known.1 The use of ionic liquids (ILs) or micelles as the reaction medium enables the stabilization of catalysts, preventing their undesired reduction to metallic gold. Previously, the unique stability and recyclability of these catalyst solutions were demonstrated.2

Novel sustainable synthetic routes are often a key to success in advancing compounds of a certain class of specialty chemicals, for instance with unprecedented substitution patterns or improved reaction yields. In this regard, boron–dipyrromethene (BODIPY) dyes as one of the fastest growing classes of dyes3 are very attractive targets. The popularity of BODIPY dyes is rooted in their outstanding absorption and fluorescence properties and high photo- and chemical stabilities, having resulted in a wide variety of applications. Besides bioimaging,4 fluorescent indicators,5 photodynamic therapy,6 and organic photovoltaics,7 highly fluorinated BODIPYs have been recently found to constitute a very promising class of multifunctional labeling agents.8 One of the bottlenecks for novel BODIPYs is the availability of tailor-made pyrroles which constitute their core building blocks.

The expertise of our groups in sustainable metal catalysis2 and the development of high-performance dyes9 prompted us to embark on a collaborative approach toward new fluorinated BODIPY dyes 4 by gold-catalyzed cycloisomerization of α-aminoallenes 1 to 2,5-dihydropyrroles 2 in ionic liquids (Scheme 1). Although fluorinated BODIPYs possess many promising features, only a very limited number of fluorinated BODIPYs is known in the literature.8,10 Furthermore, the synthesis of pyrroles by cyclization reactions allows extensive structural variation and represents a valuable alternative compared to common strategies.11


image file: c4ob02671c-s1.tif
Scheme 1 Synthesis of BODIPY dyes.

Results and discussion

The cycloisomerization was carried out in ionic liquids (Tables 1–3). Different gold catalysts and ionic liquids were examined for this purpose (Scheme 2).
image file: c4ob02671c-s2.tif
Scheme 2 Gold catalysts and ionic liquids used in this work.
Table 1 Influence of ionic liquids on the gold-catalyzed cycloisomerization of α-aminoallene 1a

image file: c4ob02671c-u1.tif

Entry Ionic liquid A (mol%) T (°C) t (h) Conv.a (%)
a Conversion was determined by 1H-NMR spectroscopy.
1 [BMIM][PF6] 1 25 24 100
2 [BMIM][PF6] 2 25 2 100
3 [BMIM][PF6] 2 60 0.25 100
4 [BMIM][HSO4] 2 60 0.25 100
5 [EMIM][HSO4] 2 60 1 100
6 [BMIM][HSO4] 60 24 0
7 [EMIM][HSO4] 60 24 0
8 [BMIM][BF4] 2 25 24 0
9 [EMIM][BF4] 2 25 24 0


Table 2 Influence of the gold catalyst on the cycloisomerization of α-aminoallene 1a in [BMIM][PF6]

image file: c4ob02671c-u2.tif

Entry [Au] Additive t Conv.a (%)
a Conversion was determined by 1H-NMR spectroscopy. b Toluene (ca. 20 vol%) was added. c Reaction temperature of 50 °C.
1 A 2 h 100
2 AuBr3 48 h 30
3 Ph3PAuNTf2 1 h 100
4b Ph3PAuNTf2 5 min 100
5 Ph3PAuCl AgSbF6 1 h 100
6 Ph3PAuCl 24 h 0
7 B AgSbF6 30 min 100
8 B 24 h 0
9b B AgSbF6 5 min 100
10c B AgSbF6 10 min 100
11 AgSbF6 24 h 20


Table 3 Recycling of the catalyst/solvent system

image file: c4ob02671c-u3.tif

Entry Allene Run Yielda (%)
a Isolated yield. b 12% of 1a was recovered.
1 1a 1 91
2 1a 2 91
3 1a 3 98
4 1a 4 95
5 1a 5 80b
6 1b 1 90
7 1b 2 92


The cyclization conditions were optimized with allene 1a and cationic gold catalyst A (Table 1). Best results were achieved with [BMIM][PF6] and [BMIM][HSO4].

With 1 mol% of catalyst A in [BMIM][PF6], complete conversion was observed after 24 h, while 2 mol% led to a shorter reaction time of 2 h (entries 1 & 2). The acidic ILs [BMIM][HSO4] and [EMIM][HSO4] were used at 60 °C to attain a lower viscosity. Short reaction times of 15 min and 1 h were obtained (entries 4 & 5), which are comparable to [BMIM][PF6] at the same temperature (entry 3). The acidic ILs do not mediate the cycloisomerization in the absence of the gold catalyst (entries 6 & 7).12,13 In [BMIM][BF4] and [EMIM][BF4] no reaction was observed after 24 h at RT (entries 8 & 9). This may be due to a limited solubility of the gold catalyst in these more hydrophilic ionic liquids.

Allene 1a and ionic liquid [BMIM][PF6] were used for screening the gold catalyst (Table 2). Except for AuBr3 (entry 2), all catalysts examined gave short reaction times between 30 min and 2 h (entries 1, 3, 5 & 7). Both precatalyst B and Ph3PAuCl were transformed into the corresponding cationic gold species with AgSbF6. Without the addition of a silver salt no reaction was observed (entries 6 & 8), whereas the silver salt alone induces only a very slow cyclization (entry 11). Beside the possibility of warming up (entry 10), the addition of a certain amount of toluene to form a solvent film led to decreased reaction times of 5 min (entries 4 vs. 3, 9 vs. 7). This acceleration is probably caused by the two-phasic character of the reaction mixture. While the catalyst/solvent system is a viscous solution, the allene is a solid and affords an emulsion during the reaction. The toluene seems to act as a phase-transfer agent. These conditions were chosen for recycling experiments (Table 3).

Previous investigations with AuBr3 in [BMIM][PF6] showed an extremely low catalyst leaching using hexane as an extraction solvent.2a The pyrrolines 2 are not soluble in nonpolar solvents like hexane. Unfortunately, polar solvents like Et2O cause a dramatic catalyst leaching. Therefore, the recycling was tested for 1a with toluene as the extraction solvent and Ph3PAuNTf2 as the gold catalyst. Excellent reactivity was observed over four runs (Table 3, entries 1–4). In all runs, the reactions were stopped after 5 min, leading to quantitative conversions and high yields. A loss of reactivity was observed in the fifth run (entry 5), causing a lower yield of 80% (12% of allene 1a were also recovered). Allene 1b was cyclized under similar conditions with 90% yield in 5 min (entry 6). A second run was carried out with a similar result (entry 7).

The cyclization products are converted into pyrroles using KOtBu in THF (Scheme 1).14 According to common strategies, these were directly used in the BODIPY synthesis.10a In contrast to the literature method, however, oxidation times were extended and a second complexation step was employed, finally providing the dyes in good yields of 68 and 70%.

The molecular structures of 4a,b were determined by single-crystal X-ray diffraction analysis and the corresponding molecular configurations are shown in Fig. 1. Selected crystallographic data and structure refinement parameters are listed in Table S1. The observed geometric parameters of all crystal structures are generally comparable to those of other BODIPYs.154a crystallizes in a monoclinic space group and 4b in a triclinic space group with similar crystal packing. There are no significant interactions like hydrogen bonds or π–π interactions in the crystal lattice of both dyes. Only a weak π–π interaction occurs in 4a with a plane-to-plane distance of 4.34 Å between the same pyrrole rings (N1, C1, C2 ,C3, C4) of two neighbouring pyrromethene fragments of the BODIPY core.


image file: c4ob02671c-f1.tif
Fig. 1 ORTEP representation of 4a (left) and 4b (right) with atomic labeling shown with 30% probability displacement ellipsoids.

The BODIPY skeleton formed by three conjugated heterocyclic rings is almost planar, with an rms deviation of 0.1260 and 0.1016 for 4a and 4b. Due to steric repulsion of the methyl groups attached to C1 and C7, the pentafluorophenyl moiety is strongly twisted out of the BODIPY mean plane, with dihedral angles of 88.35(05)° for 4a and 76.73(10)° for 4b, which is as expected more pronounced than for 1,7-H- and 3,5-diaryl-substituted dyes.10e,15c An overlay of the two molecular configurations shows the differences in molecular conformation, i.e., the arrangement of the substituents attached at C3 and C5 (Fig. S1). Whereas in 4b the two phenyl rings are almost parallel with a dihedral angle of 28.09(22)°, the angle amounts to 86.24(07)° in 4a. Because the arrangement of the molecules is head-to-tail, the short contacts from the pentafluorophenyl ring to the phenyl rings at C3,5 lead to a bowl-like conformation of 4a (Fig. 2). Such a bowl-type arrangement of head-to-tail aligned 3,5,8-phenyl-substituted BODIPYs has also been very recently reported by Wakamiya.15c


image file: c4ob02671c-f2.tif
Fig. 2 Short contacts in 4a lead to a bowl-like conformation.

The spectroscopic properties of 4a and 4b are comparable with very little dependence on the solvent. The spectral envelopes shown in Fig. 3 resemble classical BODIPY features. The high fluorescence quantum yields (Φf > 0.85), long fluorescence lifetimes (τf ∼ 5.5 ns) and high absorption coefficients (ελmax ∼ 50000 M–1 cm–1), independent of solvent polarity and proticity, distinguish the dyes as bright emitters (comprehensive spectroscopic data of 4a,b are shown in the ESI). Compared to their parent dye, 8-pentafluorophenyl-1,7,3,5-tetramethyl-BODIPY that carries methyl groups instead of the trifluoromethylphenyl residues on the 3,5-positions,8 the absorption maxima are shifted by ca. 40 nm and the absorption bands are significantly broadened. The Stokes shifts are increased from ca. 300 to 1000 cm–1. In view of the color rules for BODIPY π-system substitution,16 these changes can be attributed to the extension of the fluorophores’ π-system with the additional trifluorophenyl moieties and their enhanced rotational freedom. The comparatively small spectral differences between 4a and 4b despite the introduction of two additional CF3 groups in the first one are tentatively ascribed to the enhanced steric hindrance in the case of 3,5-phenyl substitution on the BODIPY core.17 That results in considerable (average) torsion angles around the C1–C1′ bond 〈θC2–C1–C1′–C2′〉 = 39° for 4a and 38° for 4b (for the atom numbering pattern, see Scheme 1, bottom panel) and hence reduced electronic coupling between the BODIPY core and phenyl rings, diminishing the electron withdrawing effect of these substituents. These experimental findings are well reflected by quantum chemical calculations employing the density functional theory (DFT) method, revealing differences in the S1←S0 transition in the gas phase of only ca. 5 nm (Table S2, ESI). The latter are also the only oscillator-strong transitions in the 350–500 nm region for both dyes, involving HOMO and LUMO which are both centered almost exclusively on the 3,5-diphenyl-dipyrrin fragments for 4a,b (Fig. S2, ESI). Furthermore, with the corresponding triplet transitions being shifted for ca. 1.25 eV, the theoretical studies fully support the favorable spectroscopic properties found experimentally.


image file: c4ob02671c-f3.tif
Fig. 3 Absorption and fluorescence spectra of 4a (black) and 4b (red) in ethanol; the corresponding spectra of rhodamine 101 (blue) are included for comparison.

Dyes 4a and 4b absorb and emit in the green visible range, matching well the spectral region of prominent rhodamine dyes such as rhodamine 101, 6G, B or TRITC18 and fitting perfectly to the output of green laser sources (e.g., the prominent 532 or 543 nm lasers, Fig. 3). Compared to rhodamines the zwitterionic yet net uncharged BODIPYs are commonly well soluble in organic solvents across the entire polarity range,3,19 broadening the areas of applications and offering potent alternatives to the search for hydrophobic rhodamine derivatives.20 The most important feature of the newly synthesized fluorinated BODIPYs 4a,b with regard to application is their excellent photostability, in particular when used in aprotic solvents. Fig. 4 exemplarily combines the photobleaching curves of 4a,b and rhodamine 101 upon irradiation at 532 nm.


image file: c4ob02671c-f4.tif
Fig. 4 Photobleaching curves for irradiation at 532 nm (1 W laser power) in acetonitrile (4a: orange, 4b: blue, rhodamine 101: red) and ethanol (4a: magenta, 4b: green, rhodamine 101: black).

Conclusions

In the work presented here, two highly fluorinated and photo-stable BODIPY dyes with excellent spectroscopic properties have been prepared via gold catalysis in ionic liquids. While the use of sustainable gold catalysis is often strictly limited to the applied system, the possibility of recycling the catalyst solution was extended to more challenging allenes. Here, [BMIM][PF6] was found to be an excellent solvent system for gold catalysis and recycling. The fluorinated dyes possess an enormous potential for further applications such as functional group labeling or the staining of lipophilic compartments, for instance in cell imaging. Current work in our laboratories is directed toward both the synthesis of other functional BODIPYs and the application of the title dyes.

Acknowledgements

Financial support by SusChemSys (which is co-financed by the Regional Development Fund (Investing in Your Future) of the European Union and the state of North Rhine-Westphalia), the European Metrology Research Programme (EMRP) (project IND15 SurfChem) and the Adolf-Martens-Fund is gratefully acknowledged. The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union.

Notes and references

  1. (a) X. Liu, L. He, Y.-M. Liu and Y. Cao, Acc. Chem. Res., 2014, 47, 793 CrossRef CAS PubMed; (b) C. Winter and N. Krause, Green Chem., 2009, 11, 1309 RSC; (c) C. Hviid Christensen and J. K. Nørskov, Science, 2010, 327, 278 CrossRef PubMed.
  2. (a) Ö. Aksın and N. Krause, Adv. Synth. Catal., 2008, 350, 1106 CrossRef; (b) S. R. K. Minkler, B. H. Lipshutz and N. Krause, Angew. Chem., Int. Ed., 2011, 50, 7820 CrossRef CAS PubMed; (c) S. R. K. Minkler, N. A. Isley, D. J. Lippincott, N. Krause and B. H. Lipshutz, Org. Lett., 2014, 16, 724 CrossRef CAS PubMed; (d) X. Liu, Z. Pan, X. Shu, X. Duan and Y. Liang, Synlett, 2006, 1962 CAS; (e) X. Moreau, A. Hours, L. Fensterbank, J.-P. Goddard, M. Malacria and S. Thorimbert, J. Organomet. Chem., 2009, 594, 561 CrossRef PubMed.
  3. (a) G. Ulrich, R. Ziessel and A. Harriman, Angew. Chem. Int. Ed., 2008, 47, 1184 CrossRef CAS PubMed; (b) A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891 CrossRef CAS PubMed.
  4. K. Umezawa, D. Citterio and K. Suzuki, Anal. Sci., 2014, 30, 327 CrossRef CAS.
  5. N. Boens, V. Leen and W. Dehaen, Chem. Soc. Rev., 2012, 41, 1130 RSC.
  6. A. Kamkaew, S. H. Lim, H. B. Lee, L. V. Kiew, L. Y. Chung and K. Burgess, Chem. Soc. Rev., 2013, 42, 77 RSC.
  7. A. Bessette and G. S. Hanan, Chem. Soc. Rev., 2014, 43, 3342 RSC.
  8. (a) M. Hecht, T. Fischer, P. Dietrich, W. Kraus, A. Descalzo, W. Unger and K. Rurack, ChemistryOpen, 2013, 2, 25 CrossRef CAS PubMed; (b) T. Fischer, P. M. Dietrich, C. Streeck, S. Ray, A. Nutsch, A. Shard, B. Beckhoff, W. E. S. Unger and K. Rurack, Anal. Chem., 2015 DOI:10.1021/ac503850f.
  9. (a) P. Ashokkumar, H. Weißhoff, W. Kraus and K. Rurack, Angew. Chem. Int. Ed., 2014, 53, 2225 CrossRef CAS PubMed; (b) Y.-H. Yu, A. B. Descalzo, Z. Shen, H. Röhr, Q. Liu, Y.-W. Wang, M. Spieles, Y.-Z. Li, K. Rurack and X.-Z. You, Chem. – Asian J., 2006, 1, 176 CrossRef CAS PubMed; (c) M. Kollmannsberger, K. Rurack, U. Resch-Genger and J. Daub, J. Phys. Chem. A, 1998, 102, 10211 CrossRef CAS.
  10. For pentafluorophenyl-substituted BODIPYs see: (a) G. Vives, C. Giansante, R. Bofinger, G. Raffy, A. Del Guerzo, B. Kauffmann, P. Batat, G. Jonusauskas and N. D. McClenaghan, Chem. Commun., 2011, 47, 10425 RSC; (b) O. Galangau, C. Dumas-Verdes, R. Méallet-Renault and G. Clavier, Org. Biomol. Chem., 2010, 8, 4546 RSC. For aryl- and fluorine-substituted BODIPYs see: (c) V. Lakshmi and M. Ravikanth, Chem. Phys. Lett., 2013, 564, 93 CrossRef CAS PubMed; (d) V. Lakshmi and M. Ravikanth, J. Org. Chem., 2011, 76, 8466 CrossRef CAS PubMed; (e) A. Burghart, H. Kim, M. B. Welch, L. H. Thoresen, J. Reibenspies and K. Burgess, J. Org. Chem., 1999, 64, 7813 CrossRef CAS; (f) L. H. Thoresen, H. Kim, M. B. Welch, A. Burghart and K. Burgess, Synlett, 1998, 11, 1276 CrossRef PubMed.
  11. (a) V. Estévez, M. Villacampa and J. C. Menéndez, Chem. Soc. Rev., 2010, 39, 4402 RSC; (b) V. Estévez, M. Villacampa and J. C. Menéndez, Chem. Soc. Rev., 2014, 43, 4633 RSC. For 2,4-substituted pyrroles see: (c) Z. Chen, B. Lu, Z. Ding, K. Gao and N. Yoshikai, Org. Lett., 2013, 15, 1966 CrossRef CAS PubMed; (d) H.-Y. Wang, D. S. Mueller, R. M. Sachwani, H. N. Londino and L. L. Anderson, Org. Lett., 2010, 12, 2290–2293 CrossRef CAS PubMed; (e) L. Meng, K. Wu, C. Liua and A. Lei, Chem. Commun., 2013, 49, 5853 RSC; (f) Z. Shi, M. Suri and F. Glorius, Angew. Chem., Int. Ed., 2013, 52, 4892 CrossRef CAS PubMed.
  12. N. Krause, M. Laux and A. Hoffmann-Röder, Tetrahedron Lett., 2000, 41, 9613 CrossRef CAS.
  13. N. Krause, A. Hoffmann-Röder and J. Canisius, Synthesis, 2012, 12, 1759 Search PubMed.
  14. T. Xu, X. Mu, H. Peng and G. Liu, Angew. Chem., Int. Ed., 2011, 50, 8176 CrossRef CAS PubMed.
  15. (a) C. L. Picou, E. D. Stevens, M. Shah and J. H. Boyer, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1990, 46, 1148 CrossRef; (b) D.-C. Wang, C. He, J.-L. Fan, W.-W. Huang and X.-J. Peng, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2007, 63, o2900 CAS; (c) A. Wakamiya, T. Murakami and S. Yamaguchi, Chem. Sci., 2013, 4, 1002 RSC.
  16. Y.-W. Wang, A. B. Descalzo, Z. Shen, X.-Z. You and K. Rurack, Chem. – Eur. J., 2010, 16, 2887 CrossRef CAS PubMed.
  17. In addition it has to be kept in mind that the phenyl ring attenuates the effects of a CF3 group, i.e., the step from 4-CF3C6H5 to 3,5-(CF3)2C6H4 has much less influence in terms of for instance the Hammett constant than the step from a p-CF3 to two m-CF3 groups, cf. e.g. G. A. Olah, A. L. Berrier, M. Arvanaghi and G. K. S. Prakash, J. Am. Chem. Soc., 1981, 103, 1122 CrossRef CAS; Y. Borguet, X. Sauvage, G. Zaragoza, A. Demonceau and L. Delaude, Organometallics, 2010, 29, 6675 CrossRef.
  18. (a) P. C. Beaumont, D. G. Johnson and B. J. Parsons, J. Chem. Soc., Faraday Trans., 1993, 89, 4185 RSC; (b) H. Brismar, O. Trepte and B. Ulfhake, J. Histochem. Cytochem., 1995, 43, 699 CrossRef CAS PubMed.
  19. Another advantage of BODIPY dyes is the straightforwardness of tailoring water solubility through substitution of the fluorine atoms of the BF2 group, see, e.g: (a) A. M. Courtis, S. A. Santos, Y. Guan, J. A. Hendricks, B. Ghosh, D. M. Szantai-Kis, S. A. Reis, J. V. Shah and R. Mazitschek, Bioconjugate Chem., 2014, 25, 1043 CrossRef CAS PubMed; (b) S.-L. Niu, G. Ulrich, P. Retailleau, J. Harrowfield and R. Ziessel, Tetrahedron Lett., 2009, 50, 3840 CrossRef CAS PubMed.
  20. L. F. Mottram, S. Forbes, B. D. Ackley and B. R. Peterson, Beilstein J. Org. Chem., 2012, 8, 2156 CrossRef CAS PubMed.

Footnote

Electronic supplementary information (ESI) available: Full experimental details, including X-ray structure analysis, optical spectroscopy, computational studies, and NMR spectra. CCDC 1040186 and 1040187. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ob02671c

This journal is © The Royal Society of Chemistry 2015
Click here to see how this site uses Cookies. View our privacy policy here.