18

4

Full PaperReceived: 22 April 2009 Revised: 16 September 2009 Accepted: 16 September 2009 Published online in Wiley Interscience: 21 October 2009

(www.interscience.com) DOI 10.1002/aoc.1568

Polyhedral oligomeric silsesquioxane-boundiminofullereneDavid J. Clarkea∗, Janis G. Matisonsa, George P. Simonb, Marek Samocc

and Anna Samocc

Polyhedral oligomeric silsesquioxane (POSS) cages containing an iminofullerene species are reported herein. Monosubstitutedbenzyl chloride POSS was synthesized, and subsequently reacted with sodium azide to form mono benzyl azide POSS. The azidewas subsequently reacted with C60 in anhydrous, degassed toluene to yield the desired POSS iminofullerene compound. Theprepared compounds were characterized by multinuclear NMR, electrospray mass spectrometry, elemental analysis, UV–vis,fluorescence and optical power limiting measurements. Copyright c© 2009 John Wiley & Sons, Ltd.

Supporting information may be found in the online version of this article.

Keywords: Power limiting; silsesquioxane; fullerene

Introduction

The use of lasers is prevalent in a variety of scientific, industrial,medical and military fields. The emission wavelength of lasers canbe tuned from the visible to the near-infrared (NIR) region andthe radiation can be emitted either as a continuous wave or in apulsed mode, with pulse duration ranging from microseconds tofemtoseconds. Protection of operating personnel and technicalcomponents against pulsed tunable lasers is a high priority. Eyeprotection is critical, as the retina is vulnerable in the visible andNIR spectral range, generating much interest in the developmentof optical limiting materials.[1,2] Promising optical limitingmaterials are those that exhibit strong nonlinear absorptions, andare also known as reverse saturable absorbers (RSA).[3] The primaryrequirement for RSA optical limiting is a large ratio of excited-stateto ground-state absorption cross section. Thus, potent reversesaturable absorbers are usually molecules with weak ground-stateabsorptions, such as metallophthalocyanines,[4 – 8] mixed metalcomplexes[9 – 12] and clusters[13] and fullerenes.[3,14 – 18]

C60 solutions particularly show strong optical-limiting behav-ior derived from such an RSA mechanism. C60 exhibits a broadabsorption spectrum, characterized by strong absorptions in theultraviolet region and weaker absorptions that extend over themajority of the visible region.[19] This weak absorption allows foroptical pumping using a broad range of laser wavelengths. The ex-cited state dynamics and the large quantum yield for intersystemcrossing allow for the build-up of population in either the singletor triplet excited state, depending on the duration of the laserpulse. C60 also possesses excited state absorption cross sectionslarger than those of the ground state over the complete visi-ble spectrum.[20] Fullerene derivatives, such as methanofullerenesand fulleropyrrolidines, exhibit different electronic properties. Theground state absorptions are very different in the UV–vis region, asmethanofullerenes show a major peak at 500 nm, whilst fulleropy-rrolidines absorb much less. However, the triplet–triplet absorp-tion is stronger for methanofullerenes than fulleropyrrolidines,thus making them equally useful for optical limiting purposes.[21]

Fullerene solutions exhibit excellent optical limitingproperties.[16] One of the major drawbacks of C60 is its lowsolubility; however, derivitisation of C60 significantly improvessolubility.[22] The use of solid devices is preferred for practicalapplications, thus crystalline films of C60 have previously beenstudied,[2] but were found to be inefficient for pulses longer thantens of picoseconds. This is ascribed to the fast de-excitation of theexcited state due to interactions of neighboring C60 molecules inthe solid phase. Studies have shown that C60 does, however, retainoptical limiting properties after inclusion of in solid matrices,such as sol–gel glasses,[19,23 – 27] PMMA[20,28] and glass-polymercomposites;[21] however the optical responses of C60 in PMMA aremuch weaker than those displayed in solution. This difference inperformance has been attributed to the fact that different opticallimiting mechanisms occur in the solution and solid phases.[3]

A modified optical limiting mechanism has been proposed,detailing the contribution of bimolecular processes, such asself-quenching of the fullerene excited triplet state by groundstate fullerene molecules and triplet–triplet annihilation.[3]

Blending in, or alternatively incorporating covalently, smallfunctional molecules in polymers is a powerful approach, whichallows the development of new materials and devices. The idealsituation is to increase the concentration of optical limitingmolecules per unit volume without affecting the responsiveness ofthe molecules. A promising approach towards such responsive or‘smart materials’ is the integration of an addressable function intothe desirable building blocks – in our case polyhedral oligomericsilsesquioxanes (POSS).[29 – 34] Our aim was to prepare molecular

∗ Correspondence to: David J. Clarke, Industrial Research Limited, Photonics, POBox 31-310, Lower Hutt 5040, New Zealand. E-mail: d.clarke@irl.cri.nz

a School of Chemistry, Physics & Earth Sciences, Flinders University, Australia

b Department of Materials Engineering, Monash University, Clayton, Victoria,Australia 3800

c Laser Physics Centre, Australian National University, 0200 ACT Australia

Appl. Organometal. Chem. 2010, 24, 184–188 Copyright c© 2009 John Wiley & Sons, Ltd.

18

5

Polyhedral oligomeric silsesquioxane-bound iminofullerene

Figure 1. Synthesis of POSS iminofullerene.

POSS systems with an optical limiting ability incorporated intothe extended, organofunctional arm.

We have previously reported the covalent attachment of POSSto C60 through the formation of POSS fulleropyrrolidines.[35]

The lengthy synthetic pathway dictated the requirement for asimpler, higher yielding synthesis. This was achieved throughthe formation of monosubstituted POSS azides and subsequentcycloaddition to C60 to yield the desired POSS iminofullerene, thesynthesis and characterization of which is the subject of this report.

Results and Discussion

The usefulness of the octasilsesquioxane (POSS) cube as adendritic scaffold, its nanoscale size, efficient cellular uptake,low toxicity and cytocompatibility offer new applications forthese nanoparticles.[36,37] Such applications require the designof new POSS structures in which specifically selected functionalgroups are placed along the periphery of the radially extendedorganofunctional arms to confer the resultant materials with newproperties, i.e. POSS molecules with inbuilt functionality in whichthe systems exhibit a responsive behavior toward an external forcewithout altering the composition of the material.

The synthetic pathway is depicted in Fig. 1, and describedherein. Monosubstituted POSS azides have previously been syn-thesized through the ring opening of epoxides by azides andnucleophilic substitution via azide/halide exchange.[38] Incom-pletely condensed POSS [R7Si7O9(OH)3, R = iBu] was reactedwith trichloro[(4-chloromethyl)phenyl]trichlorosilane to form themonosubstituted POSS chloride, which was subsequently re-acted with sodium azide to form the desired mosubsituted POSSazide.[38] The resultant POSS azide underwent 1,3-dipolar cylcoad-dition with C60, followed by thermal extrusion of nitrogen,[39] toyield the desired POSS iminofullerene.

The presence of the azide group was confirmed by FTIR, withthe azido moiety exhibiting a characteristic stretching band at2099 cm−1,[38,40] which disappeared upon reaction with C60. Anintense Si–O stretching band was observed in all POSS compoundsat approximately 1105 cm−1.

1H NMR exhibited a downfield shift in the benzyl CH2 resonanceof 0.62 ppm, attributed to the increased deshielding associatedwith the presence of the carbon sphere.[41] An analogousdownfield shift was also apparent in the 13C NMR spectrum

Figure 2. Open [5, 6] and closed [6, 6] iminofullerene isomers.

Figure 3. UV–vis spectra of C60 and POSS iminofullerene.

(13.09 ppm), with the benzyl CH2 resonance evident at δ 68.12.Depending on the mode of addition, one of four isomers can beformed, the open [5,6], closed [5,6], open [6,6] or closed [6,6].[39]

The iminofullerene structure can be quantified through 13C NMR,as closed [6,6] fullerenes possess two sp3 type carbons, whilst open[5,6] fullerenes contain only sp2 type carbons (Fig. 2). Therefore,closed [6,6] fullerenes exhibit a resonance at approximatelyδ 83 in the sp3 region, whereas open [5,6] fullerenes exhibit nopeaks in this region of the spectra, with all carbons apparentin the sp2 range.[41,42] All fullerene resonances were apparentfrom δ 128.25–152.74, indicating that the iminofullerene was ofthe open [5,6] type structure. 29Si NMR resonances of the POSSiminofullerene exhibited multiple peaks (−67.97, −68.25) for theT-type silicon atoms.

The UV–vis spectrum of the POSS iminofullerene is shown inFig. 3, obtained at concentrations of 5 × 10−6 M (iminofullerene)and 1×10−6 M (C60), with magnifications (10×) included for clarity.The reaction of a carbon–carbon bond, located on a [5,6] junctionimplied that the 60π electron nature of the fullerene core waslargely conserved in both open and closed isomers. The UV–visspectra of the open [5,6] iminofullerenes thus reveal strong resem-blance to the isoelectronic core of C60. The most noticeable devia-tion of the iminofullerene spectra compared with that of C60 residesin the low-intensity maximum at approximately 698 nm, with a mo-lar extinction coefficient of 138, which refers to the S0 → ∗S1 tran-sition that characterised the energy of the singlet excited state.[15]

The fluorescence spectrum of the POSS iminofullerene (Fig. 4)exhibited an emission band at 726 nm. This broad, weak emissionband is characteristic of the [5,6] open structure, whereas closed[6,6] iminofullerenes exhibited emission bands at approximately

Appl. Organometal. Chem. 2010, 24, 184–188 Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/aoc

18

6

D. J. Clarke et al.

Figure 4. Fluorescence spectra of C60 and POSS iminofullerene.

680 nm, thus providing further confirmation of the [5,6] openstructure. The weakness of the emission, similar in intensity tothat of C60, is related to the combination of short singlet lifetime,quantitative intersystem crossing and the symmetry-forbiddennature of the lowest-energy transition.[45]

Figure 5 exhibits the power limiting results obtained for C60 andthe iminofullerene compound, with the power limiting propertiesof the POSS iminofullerene observed essentially identical to thoseof C60. A solution of POSS iminofullerene displayed distinct powerlimiting of the transmission with the onset at approximately200 mJ cm−2. Strong thermal effects were seen at fluences aboveca 1 J cm−2, evident from increased scattering. While the resultsfor the iminofullerene were obtained at higher concentrations byweight, the molar concentrations of C60 and the iminofullerenewere similar, thus leading to comparable number densities of theC60 moieties.

Experimental

Commercially available chemicals were used without purificationunless otherwise stated. R7Si7O9(OH)3 was purchased from Hy-brid Plastics (Hattiesburg, MS, USA). C60 was purchased from SESLaboratories (Houston, TX, USA). Solvents were purified and driedaccording to literature procedures.[46] Chlorobenzene was dis-tilled from calcium hydride, degassed via the freeze–pump–thawmethod and distilled via standard Schlenk techniques. Monoben-zyl chloride and azide POSS were synthesized according to theprocedures detailed in the literature.[38] 1H, 13C and 29Si NMRwere recorded on a 300 MHz Varian Gemini FT-NMR. Externaltetramethysilane was used as a reference for 29Si NMR spectra,with solvent peaks used as references for 1H and 13C NMR spec-tra. Fourier transform infrared (FTIR) spectra were obtained on aNicolet Nexus 8700 FTIR spectrometer. Accurate mass electrosprayionisation (ESI) analysis was performed on a Bruker BioApex II 47eFourier transform mass spectrometer (FT-MS). Ultraviolet–visible(UV–vis) spectra were recorded on a Cary 50 Scan UV–vis spec-trophotometer and fluorescence spectra were recorded on a CaryEclipse spectrophotometer.

Power limiting measurements were performed using a diode-Q-switched Nd : YLF laser which after frequency doubling of theoutput provided ∼25 ns 523 nm pulses of microjoule energies. The

experimental setup was of a similar type to the standard f/5 test bedused in the literature[47] and toluene solutions of the investigatedcompounds with concentrations adjusted to provide ∼70%transmission at 523 nm were examined in 1 mm glass cells. Thetransmission vs fluence curves were each constructed from severalruns in which the incident pulse energy was manipulated and thefluence was additionally varied by scanning the sample positionalong the z-axis. The shapes of open and closed aperture Z-scansobtained in such a way were used to calculate the fluence values.

1-(4-benzyl chloride)-3,5,7,9,11,13,15-heptakis(isobutyl)pentacyclo[9.5.1.15,9.15,15.17,13.]octasiloxane (Mono-benzylChloride POSS)

Mono-benzyl chloride POSS was synthesized accordingto the method outlined by Wei et al.[38,48] Trichloro[4-(chloromethyl)phenyl]silane (2.7 ml, 14 mmol) was addeddropwise to a solution of iBu7Si7O9(OH)3 (10 g, 12.6 mmol) andtriethylamine (6 ml, 43 mmol) in THF (30 ml). The reaction mixturewas stirred overnight and then filtered through celite. The filtratewas added to a stirred solution of acetonitrile and the resultantprecipitate was isolated by filtration and dried in vacuo.

Yield = 4.00 g (34%); FTIR (KBr, cm−1): 2954m, 2869m, 1465m,1400 w, 1366 w, 1332 w, 1230m, 1110 vs, 1039m, 838m, 741m,694 w; 1H NMR (CDCl3): δ 0.58–0.62 (m, 16H, iBu CH2), 0.93–0.97(m, 42H, iBu CH3), 1.85–1.91 (m, 7H, iBu CH), 4.59 (s, 2H, CH2Cl),7.39 (d, 3JH−H = 7.50 Hz, 2H, CH), 7.64 (d, 3JH−H = 7.50 Hz, 2H,CH); 13C NMR (CDCl3): δ 22.37, 22.46 (iBu CH2), 23.85 (iBu CH), 25.69(iBu CH3), 46.09 (Bz CH2), 127.68 (CH), 132.13 (Cq), 134.44 (CH),139.28 (Cq)29Si NMR (CDCl3): δ −67.97, −68.26, −68.29, −68.49(RSiO3); [M + Na]+ 963.2536 (963.2529 theory).

1-(4-benzyl azide)-3,5,7,9,11,13,15-heptakis(isobutyl)pentacyclo[9.5.1.1.5,9.5,15.1.7,13.]octasiloxane (Mono-benzylAzide POSS)

Mono-benzyl azide POSS was prepared according to the modifiedliterature procedure.[38,49] Sodium azide (1.10 g, 16.9 mmol) wasadded to a solution of mono-benzyl chloride POSS (1.62 g,1.69 mmol) in DMF (40 ml) and the solution was heated at 80 ◦Covernight. The solution was cooled to room temperature, dilutedwith chloroform (150 ml), and washed with a NaHCO3 solution(1 M, 2 × 100 ml) and water (2 × 100 ml). The organic layer wasdried (Na2SO4) and the solvent removed in vacuo to yield thedesired POSS azide as a white solid.

Yield = 0.33 g (33%); FTIR (KBr, cm−1): 2954s, 2869m, 2099m,1465m, 1399 w, 1383 w, 1366 w, 1332 w, 1230s, 1106 vs, 1038m,838m, 803 w, 744m; 1H NMR (CDCl3): δ 0.57–0.67 (m, 14H, iBu CH2),0.95–0.99 (m, 42H, iBu CH3), 1.85–1.93 (m, 7H, iBu CH), 4.36 (s, 2H,CH2N3), 7.33 (d, 3JH−H = 7.8 Hz, 2H, CH), 7.69 (d, 3JH−H = 7.8 Hz,2H, CH); 13C NMR (CDCl3): δ 22.59, 22.71 (iBu CH2), 23.28, 23.37(iBu CH), 24.05, 24.12 (iBu CH3), 55.93 (Bz CH2), 127.48 (CH), 132.72(Cq), 134.74 (CH), 137.55 (Cq); 29Si NMR (CDCl3): δ −67.58, −68.24(RSiO3); [M + Na]+ 970.2937 (970.2933 theory).

1-(4-benzyl iminofullerene)-3,5,7,9,11,13,15-heptakis(isobutyl)pentacyclo[9.5.1.1.5,9.5,15.1.7,13.]octasiloxane (POSSIminofullerene)

POSS azide (266 mg, 0.28 mmol) was added to a solution of C60

(0.2 g, 0.28 mmol) in chlorobenzene (100 ml) and the solution wasrefluxed overnight. The solvent was removed in vacuo and theresidue was purified by flash chromatography (eluant hexane).

www.interscience.wiley.com/journal/aoc Copyright c© 2009 John Wiley & Sons, Ltd. Appl. Organometal. Chem. 2010, 24, 184–188

18

7

Polyhedral oligomeric silsesquioxane-bound iminofullerene

(b)

(a)

Figure 5. Power limiting of (a) C60 and (b) POSS iminofullerene.

Appl. Organometal. Chem. 2010, 24, 184–188 Copyright c© 2009 John Wiley & Sons, Ltd. www.interscience.wiley.com/journal/aoc

18

8

D. J. Clarke et al.

Yield = 82 mg (36%); FTIR (KBr, cm−1): 2955 m, 2926 m, 2906 m,2870 m, 1606 w, 1464 w, 1400 w, 1382 w, 1365 w, 1332 w, 1262m,1229 w, 1168 m, 1105 vs, 1022 m, 801 s, 742 m, 691 w; 1H NMR(CDCl3): δ 0.63–0.68 (m, 16H, iBu CH2), 0.91–0.99 (m, 42H, iBuCH3), 1.85–1.93 (m, 7H, iBu CH), 4.98 (s, 2H, CH2N), 7.78–7.80(m, 4H, CH); 13C NMR (CDCl3): δ 22.68, 22.74 (iBu CH), 24.05 (iBuCH2), 25.88 (iBu CH3), 68.12 (Bz CH2), 128.25 (Bz CH), 131.75 (BzCq), 134.69 (CH), 136.53, 137.31, 138.45, 138.76, 139.54, 139.63,141.05, 141.71, 142.37, 142.92, 143.31, 143.36, 143.63, 144.05,144.24, 144.51, 144.76, 144.92, 145.37, 146.52, 147.98, 152.74 (Cq);29Si NMR (CDCl3): δ −67.97, −68.25 (RSiO3); [M]+ 1779.0801(1779.0783 theory); elemental analysis: theoretical (%) C 70.78, H3.84, N 0.74; experimental C 69.29, H 4.14, N 0.75; UV–vis λmax

(nm) [ε (M−1 cm−1)] (toluene) 334 (23163), 698 (138); fluorescence(λexc = 335 nm, toluene) 733 nm.

Conclusions

POSS-bound iminofullerene was synthesized and characterizedby FTIR, NMR, ESI UV–vis and fluorescence, which confirmedthat the desired iminofullerene structure was present. The opticalproperties of the compound in solution were investigated withpower limiting measurements, indicating that the power limitingof the POSS iminofullerene was essentially identical to that ofC60.

Acknowledgments

This work was largely funded by the Australian Research CouncilDiscovery Scheme (DP0449692).

Supporting information

Supporting information may be found in the online version of thisarticle.

References

[1] A. W. Allaf, M. D. Zidan, Opt. Lasers Eng. 2005, 43, 57.[2] R. C. Hollins, Curr. Opin. Solid St. Mater. Sci. 1999, 4, 189.[3] J. E. Riggs, Y.-P. Sun, J. Phys. Chem. A 1999, 103, 485.[4] S. M. O’Flaherty, S. V. Hold, M. J. Cook, T. Torres, Y. Chan, M. Hanack,

W. J. Blau, Adv. Mater. 2003, 15, 19.[5] J. W. Perry, K. Mansour, I.-Y. S. Lee, X.-L. Wu, P. V. Bedworth,

C.-T. Chen, D. Ng, S. Marder, Science 1996, 273, 1553.[6] J. W. Perry, K. Mansour, S. R. Marder, K. J. Perry, D. Alvarez Jr,

I. Choong, Opt. Lett. 1994, 19, 625.[7] J. S. Shirk, R. G. S. Pong, S. R. Flom, H. Heckman, M. J. Hanack, Eur.

J. Org. Chem. 2001, 3759.[8] D. Dini, M. Barthel, M. Hanack, Eur. J. Org. Chem. 2001, 3759.[9] S. Shi, W. Ji, S. H. Tang, J. Am. Chem. Soc. 1994, 116, 3615.

[10] S. Banerjee, G. R. Kumar, P. Mathur, P. Sekar, Chem. Commun. 1997,299.

[11] T. Xia, A. Dogarui, K. Mansour, D. J. Hagan, A. A. Said, E. W. VanStryland, S. Shi, J. Opt. Soc. Am. B 1998, 17, 1497.

[12] X. Zhong, Y. Geng, S.-L. Ong, J. Hu, W.-J. Ng, Z. Wang, Chem.Commun. 2003, 1882.

[13] M. Pittman, P. Plaza, M. M. Martin, Y. H. Meyer, Opt. Commun. 1998,158, 201.

[14] Y.-P. Sun, J. E. Riggs, K. B. Henbest, R. B. Martin, J. Nonlinear Opt. Phys.Mater. 2000, 9, 481.

[15] Y.-P. Sun, J. Chem. Phys. 2000, 112, 4221.[16] L. Tutt, A. Kost, Nature 1992, 356, 225.[17] D. G. McLean, R. L. Sutherland, M. C. Bryant, D. M. Brandelik, Opt.

Lett. 1993, 18, 858.[18] Y.-P. Sun, J. E. Riggs, Chem. Mater. 1997, 9, 1268.[19] M. P. Dirk M. Guldi, Acc. Chem. Res. 2000, 33, 695.[20] A. Kost, L. Tutt, M. B. Klein, T. Kirk Dougherty, W. Elias, Opt. Lett.

1993, 18, 334.[21] R. Gvishi, J. D. Bhawalker, N. D. Kumar, G. Ruland, U. Narang,

P. N. Prasad, B. A. Reinhardt, Chem. Mater. 1995, 7, 2199.[22] A. Hirsch, Top. Curr. Chem. 1999, 199, 1.[23] C. De Favari, M. Maggini, G. Scorrano, M. Prato, G. Brusatin,

M. Guglielmi, M. Meneghetti, R. Signorini, R. Bozio, Chem. Eur. J.1999, 5, 2501.

[24] M. Prato, K. Kordatos, E. Menna, G. Scorrano, M. Maggini, J. Sol–GelSci. Technol. 2001, 22, 237.

[25] P. Innocenzi, B. Lebeau, J. Mater. Chem. 2005, 15, 3821.[26] A. Kraus, M. Schneider, A. Gugel, K. Mullen, J. Mater. Chem. 1997, 7,

763.[27] R. Signorini, M. Zerbetto, M. Meneghetti, R. Bozio, M. Maggini, C. De

Faveri, M. Prato, G. Scorrano, Chem. Commun. 1996, 1891.[28] M. Gevaert, P. V. Kamat, J. Phys. Chem. 1992, 96, 9883.[29] A. Provatas, J. G. Matisons, Trends Polym. Sci. 1997, 5, 327.[30] P. G. Harrison, J. Organomet. Chem. 1997, 542, 141.[31] G.-Z. Li, H. Ni, C.U. Pittman Jr, J. Inorg. Organomet. Polym. 2001, 11,

123.[32] T. S. H. Shawn H. Phillips, Sandra J. Tomczak, Curr. Opinion Solid St.

Mater. Sci. 2004, 8, 21.[33] X. Cao, R. H. Baney, Silicon Containing Polymers: The Science and

Technology of their Synthesis. Kluwer Academic: Dordrecht, 2000,157.

[34] M. G. Voronkov, V. I. Larentyev, Top. Curr. Chem. 1982, 102, 199.[35] D. J. Clarke, J. G. Matisons, G. P. Simon, M. Samoc, A. Samoc, Appl.

Organometal. Chem. 2008, 22, 460.[36] R. Y. Kannan, H. J. Salacinski, M. Odlyha, P. E. Butler, A. M. Seifalian,

Biomaterials 2006, 27, 1971.[37] C. Musker, J. B. Carroll, V. M. Rotello, Chem. Commun. 2005, 996.[38] L. Petraru, W. H. Binder, Polym. Prepr. 2005, 46, 841.[39] M. Cases, M. Duran, J. Mestres, N. Martin, M. Sola, J. Org. Chem. 2001,

66, 433.[40] Y. Ju, D. Kumar, R. S. Varma, J. Org. Chem. 2006, 71, 6697.[41] Q. C. Li, M. Prato, F. Wudl, V. Lucchini, J. Am. Chem. Soc. 1993, 115,

1148.[42] S. Xiao, Y. Li, H. Fang, H. Li, H. Liu, L. Jiang, D. Zhu, Synth. Met. 2003,

135, 839.[43] E. G. Shockey, A. G. Bolf, P. F. Jones, J. J. Schwab, K. P. Chaffee,

T. S. Haddad, J. D. Lichtenhan, Appl. Organometal. Chem. 1999, 13,311.

[44] G. C. Claudia Marcolli, Appl. Organometal. Chem. 1999, 13, 213.[45] D. M. Guldi, H. Hungerbuhler, I. Carmichael, K.-D. Asmus,

M. Maggini, J. Phys. Chem. A 2000, 104, 8601.[46] W. L. F. Armarego, C. L. L. Chai, Purification of Laboratory Chemicals,

5th edn. Butterworth-Heinemann: Oxford, 2003.[47] Y. Morel, P. Najechalski, N. Sanz, A. Ibanez, C. Nguefack, C. Andraud,

P. L. Baldeck, Synth. Met. 2000, 109, 215.[48] C. M. Leu, T. Y. Chang, K. H. Wei, Macro. 2003, 36, 9122.[49] V. Ervithayasuporn, X. Wang, Y. Kawakami, Chem. Commun. 2009,

5130.

www.interscience.wiley.com/journal/aoc Copyright c© 2009 John Wiley & Sons, Ltd. Appl. Organometal. Chem. 2010, 24, 184–188