Acta Chim. Slov. 2005, 52, 168-170 Short Communication (Sub)Micron CaF2 Cubes and Hollow Rods From Ionic Liquid Emulsions Andreas Taubert Department of Chemistrj, University of Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland. Fax: +41 61 267 3855; Tel: +41 61 267 3832; E-mail: andreas.taubert@unibas.ch Received 31-01-2005 Abstract I Uniform CaF2 cubes and rods with a well-defined crystal habit have been grown from emulsions of an ionic liquid (IL) in aqueous CaCl2 solutions via the hydrolysis of the PF6~ counterion of the IL. Intriguingly, at high IL contents, hollow CaF2 rods form, which has not been observed before. The rods can reach lengths up to over ten micrometers and widths of up to ca. 2.5 um. Furthermore, the surface of the IL droplet has a strong influence on the nucleation efficiency of the precipitating mineral. Key words: ionic liquids, emulsions, calcium fluoride, hollow rods Introduction Results and discussion Ionic liquids (ILs) have attracted interest as reac-tion media for organic reactions, for extraction, and for electrochemistrv.1 Recentlv, the self-assembly of ILs and the self-assembly of organic molecules in ILs,2'3 inorganic svnthesis in ILs,3"10 and emulsions with ILs11 have also been studied. It has been demonstrated that ILs can act as solvents for reactants and as precipita-tion templates for inorganic matter at the same time.3'4 If the IL is not consumed during the reaction, it can be retrieved after svnthesis and thus provides an economi-cal and potentialh/ ecological approach to inorganic materials. However, in some cases the IL does react with the environment: for example, PF6" and BF4" hy-drolvse in water over time and release fluoride F".12 Fluorides are widely used as a matrix material for rare earth ions and the controlled synthesis of CaF2 nanoparticles has recenth/ attracted some interest.13 This communication introduces a new pathway towards CaF2 crystals from emulsions of the IL l-methyl-3-octyl imidazolium hexafluorophosphate 1 (Scheme 1) in aqueous CaCl2 solutions. The system takes advantage of the slow hydrolysis of the PF6" anion and the subsequent release of F". Interestingly, high concentrations of the IL lead to hollow CaF2 rods with a square or rectangular cross-section, which has not been described before. Scheme 1 IL/water emulsions are stable for about two hours at room temperature and the droplet diameters range from ca. 0.3 to ca. 4 um. Figure 1 shows that replacing water with an aqueous 1 M CaCl2 solu-tion increases the emulsion stability to ca. 9 hours, after which the smaller droplets start to disappear. Larger droplets have been observed up to ca. 22 hours. Figure 1. Optical micrographs showing the transformation of a 10% (v/v) emulsion of 1 in 1 M CaCl2. The small droplets disappear after ca. 8 to 10 hours; larger droplets persist up to 20 to 22 hours. The disappearance of the droplets is not only associated with a demixing of the emulsion (though this is also observed), but can also be explained as follows: PF6" hydrolyses and releases F".12 At some point the solution becomes supersaturated and CaF2 precipitates. The Cl" introduced via the CaCl2 replaces PF6" as the counterion for the imidazolium cation, leading to a more water-soluble compound. The slight size increase of the IL droplets with increasing time suggests that with increasing PF6" hydrolysis level, wa-ter penetrates into the IL droplets, further hydrolyzes PF6", and ultimateh/ dissolves the IL into the aqueous phase. In this context it would also be interesting to Taubert CaF Cubes and Hollow Rods Acta Chim. Slov. 2005, 52, 168–170 169 monitor the stability of the emulsions obtained by Binks et al.11 Some of them are possibly also not stable over a longer period of time due to counterion hydrolysis. CaF2 crystals precipitated from a 1 M CaCl2 solu-tion (Figure 2) have a relatively uniform shape but a size distribution that is broader than that of CaF2 na-noparticles obtained via hydrothermal synthesis.13 SEM images show particle sizes from ca. 300 nm to over 4 |im; typical sizes are ca. 1 um. High magnification SEM images show that the surface of the particles is relativeh/ rough. It appears as if the particles were composed of even smaller nanoparticles that aggregate during the reaction similar to earlier findings on CaC03,14 Ti02,15 ZnO,16 and iron oxides.17 Figure 2d shows that only a negligible fraction of organic material is adsorbed on or incorporated into the crystal because intense Ca and F signals but only a weak C line are observed in the energy dispersive X-ray spectra of the particles. Figure 2. a) Optical micrograph of CaF2 crystals precipitated from a 10% (v/v) emulsion of 1 in 1 M CaCl2. b) SEM image of typical CaF2 crystals on a 0.22 (j.m filter. c) High magnification SEM image of a CaF2 crvstal surface showing the nm-sized subunits. d) X-ray spectrum of a CaF2 crystal showing that the crvstals only contain Ca and F and little impurities. The Au lines are from sample sputtering for SEM. To explore the tunability of the system with respect to control of particle shape, size, and size distribution, the reaction conditions were modified: rather than standing at room temperature, the reaction murture was vigorously stirred for 48 hours and the concentra-tion of the CaCl2 solutions was varied from 0.01 to 10 M. At 0.01 M no precipitation was observed but Figure 3 shows that the CaCl2 concentration and the stirring both influence CaF2 precipitation: at ali concentrations above 0.01 M, cubic crystals form but stirring leads to a more uniform crystal habit with well-defined crystal faces. Betvveen CaCl2 concentrations of 0.1 and 10 M the crystal size slightly increases from ca. 220-450 nm at 0.1 M CaCl2 to over 1 um at 10 M CaCl2. The particles are not isolated like similar particles obtained via hydrothermal synthesis,13 but form larger entities. Figure 3. The effect of CaCl2 concentration on CaF2 crystalliza-tion: a) 0.1 M CaCl2, b) 1 M CaCl2, c) 10 M CaCl2. Compare Figure 3b to 2b for the effect of stirring. As the hydrolysis of the PF6" counterion causes the precipitation, control experiments were made with NH4PF6 to elucidate whether the imidazolium ion had an effect on the crystallization or not. SEM shows no effect on the morphology, as also with-out the IL cubes are observed (data not shown). While no effect on the morphology is observed on replacing the imidazolium by NH4+, it is interest-ing to note that the IL 1 appears to promote CaF2 nucleation: in the control experiment with NH4PF6 (i.e. no IL) only ca. 1 mg of CaF2 was obtained, but ali precipitations with the IL yield ca. 10 to 15 mg of precipitate, which indicates a strong influence of the IL/water interface on nucleation. The author suspects that the ionic nature of the IL droplet (or droplet surface) is a key parameter to control nucleation and crystallization; this phenomenon is interesting and should be investigated further because it is possible that the use of another IL generates different precipitation conditions favoring other crystal morphologies or sizes. To further investigate the effect of the IL 1 on CaF2 precipitation, the ratio of IL to a 1 M CaCl2 solu-tion was varied. Neither a 30%, 50%, nor a 70% (v/v) mixture changed the appearance or size of the crystals significantly except for some roughening of the cube surfaces. However, a 90% (v/v) mixture leads to a mixture of rods and cubes, Figure 4. The rods are con-siderably larger than the previously described particles and reach lengths up to over ten microns and a width of up to ca. 2.5 um. Moreover, SEM images show that the rods are actually hollow structures with a square or rectangular cross section and a wall thickness of ca. 200 nm. The walls resemble the rough surfaces shown above. The formation of hollow CaF2 rods is intriguing and to the best knowledge of the author, they have not been described before. The reason for nanorod formation is not clear yet, but as the author has not observed liquid crystalline behavior of the 90% IL/10% CaCl2 reaction mixture, liquid cystal (LC) templating is rather unlikeh/. This is even more so because the mixture was stirred during the whole reaction time, which would greath/ disturb the order in any presumed LC template. Taubert CaF Cubes and Hollow Rods 170 Acta Chim. Slov. 2005, 52, 168–170 Figure 4. SEM image of CaF2 nanocrystals obtained from a 90% (v/v) emulsion with a 1 M CaCl2 solution: a) overview showing a typical mixture of cubes and rods; b) and c) are typical rods that have broken up during SEM sample preparation. Experimental In a typical precipitation, 2 mL of a 1 M aqueous CaCl2 solution were thoroughly mixed with 200 uL of 1 (Solvent Innovation, Germany) by shaking and stirring the mixture for ca. 20 minutes. The resulting liquid is slightly opaque and exhibits droplets of 1 surrounded by the aqueous phase under the optical microscope (Leica DM-RP), Figure 1. This mixture was allowed to stand at room temperature for 48 hours, during which time the CaF2 crystals form. The crystals were isolated by filtration through an 0.22 um Millipore filter and washed with 10 mL of water. SEM images and X-ray spectra were obtained with a Philips XL-30 ESEM/Noran X-ray detector operated at 20 kV and a working distance of 10 mm. Samples were sputtered with gold prior to imaging. Conclusions This paper demonstrates that ILs provide a flexible means for the controlled growth of CaF2 cubes and rods. It also shows that the IL does af-fect nucleation efficiency. While the rods are rather poorh/ defined and accompanied by regular (solid and cubic) particles, the approach presented here opens a new pathway towards CaF2 particles with unexpected shapes. Potentially, other fluorides can be grown using the approach presented here or doped structures can be generated; it is thus interesting to pursue further research into these processes and materials. Of par-ticular interest is the investigation of the effect of the IL droplet surface on the nucleation efficiency and particle shape selection; furthermore, the use of micro-waves, which heat up ILs quickly, could further enable the synthesis of novel compounds from IL emulsions. Acknowledgements A.T thanks Prof. W. Meier (U. Basel) for his support and for valuable discussions, Prof. H.-F. Eicke (U. Basel) and Dr. M. Wagner (Solvent Innovation) for insightful discussions, and D. Mathys and M. Diiggelin (ZMB Basel) for help with the SEM. References 1. Ionic liquids in synthesis; Eds: P. Wasserscheid, T. Welton, Wiley-VCH: Weinheim, 2002. 2. J. Dupont, /. Braz. Chem. Soc. 2004, 15, 341-350. 3. A. Taubert, Angew. Chem. Int. Ed. 2004, 43, 5380-5382. 4. T. Nakashima, N. Kimizuka, /. Am. Chem. Soc. 2003, 125, 6386-6387. 5. Y. Zhou, M. Antonietti, Adv. Mater. 2003, 15, 1452-1455. 6. Y. Zhou, M. Antonietti, Chem. Commun. 2003, 2564—2564. 7. Y. Zhou, M. Antonietti, /. Am. Chem. Soc. 2003, 125, 14960-14961. 8. S. Dai, Y. H. Ju, H. J. Gao, J. S. Lin, S. J. Pennycook, C. E. Barnes, Chem. Commun. 2000, 243-244. 9. J. Dupont, G. S. Fonseca, A. P. Umpierre, P. F. P Fichtner, S. R. Teixeira, /. Am. Chem. Soc. 2002, 124, 4228-4229. 10. R. R. Deshmukh, R. Rajagopal, K. V. Srinivasan, Chem. Commun. 2001, 1544-1545. 11. B. P. Binks, A. K. F. Dyab, P. D. I. Fletcher, Chem. Commun. 2003, 2540-2541. 12. R. P. Swatloski, J. D. Holbrey, R. D. Rogers, Green Chem. 2003, 5, 361-363. 13. X. Sun, Y. Li, Chem. Commun. 2003, 1768-1769. 14. H. Cölfen, Q. Limin, Chem. Eur. J. 2000, 7, 106-116. 15. R. L. Penn, J. F. Banfield, Science 1998, 281, 969-971. 16. A. Taubert, G. Wegner, /. Mater. Chem. 2002, 12, 805-807. 17. J. F. Banfield, S. A. Welch, H. Zhang, T. Thomsen Ebert, R. L. Penn, Science 2000, 289, 751-754. Povzetek Iz emulzije ionskega topila in vodne raztopine CaCl2 smo s hidrolizo PF6~ aniona ionskega topila. dobili pravilne I kristale CaF2 v obliki kock in valjev. Pri visoki vsebnosti ionskega topila v suspenziji smo kot prvi opazili votle valje CaF2. Valji imajo dolžino tudi več kot deset mikrometrov in premer do 2.5 \im. Površina kapljic ionskega I topila vpliva na učinkovitost nukleacije kalcijevega fluoride. Taubert CaF Cubes and Hollow Rods