Journal of Colloid and Interface Science 302 (2006) 509–515www.elsevier.com/locate/jcis

Synthesis and characterization of ultrafine CeF3 nanoparticles modified bycatanionic surfactant via a reverse micelles route

Hui Zhang, Hongfei Li, Deqian Li ∗, Shulan Meng

Key Laboratory of Rare Earth Chemistry and Physics, Changchun Institute of Applied Chemistry, Graduate School of the Chinese Academy of Sciences,Chinese Academy of Sciences, Changchun 130022, China

Received 29 March 2006; accepted 29 June 2006

Available online 8 July 2006

Abstract

In this article, we firstly reported on the synthesis and characterization of ultrafine CeF3 nanoparticles (NPs) modified by catanionic surfactantvia a reverse micelles-based route. The catanionic surfactant PN was prepared by mixing the di(2-ethylhexyl) phosphoric acid (DEHPA) andprimary amine (N1923) with 1:1 molar ratio. It exhibited a high surface activity and formed much small reverse micelles in comparison withits individual component (DEHPA or N1923). The PN reverse micelles were then used as templates to prepare ultrafine CeF3 NPs. The nar-row distributed nanoparticles have an average diameter 1.8 nm. FTIR spectra indicated that there existed strong chemical interactions betweennanoparticles and the adsorbed surfactants. The modification resulted in the FTIR peak position of P=O shifting to lower energy. Due to the effectof modification and small size, the CeF3 NPs showed a remarkable red shift of 54 nm in the fluorescence emission in comparison with that of bulkmaterial and a red shift of 18 nm in contrast with that of the normal CeF3 NPs with an average diameter of 16 nm.© 2006 Elsevier Inc. All rights reserved.

Keywords: Catanionic surfactant; Reverse micelles; Cerium fluoride; Nanoparticles; Modification; Fluorescence

1. Introduction

Cerium fluoride is increasingly widely used in the field ofoptical material [1–3], catalyst [4], and solid lubricant [5,6].Many approaches on the synthesis of CeF3 particles and filmshave been reported, such as MOCVD [7], molecular beam epi-taxy [8], aqueous precipitation [9], polyol method [10], andreverse micelles or microemulsions methods [5,11,12]. For in-stance, Lian et al. [12] prepared lutetium-doped CeF3 NPs(15–20 nm) using microemulsions consist of CTAB, n-butanol,n-octane, Ce(NO3)3 and NH4F aqueous solution and found theemission peak of nanosized CeF3 red-shifted about 30 nm incontrast to that of CeF3 single crystals.

It has been reported that catanionic surfactants, which arethe mixture of cation and anion surfactants, have high surfaceactivity, good ion-endurance and ability of self-organization[13–15]. Many articles had reported the diversity of their ag-

* Corresponding author. Fax: +86 431 5698041.E-mail address: ldq@ciac.jl.cn (D. Li).

0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.jcis.2006.06.062

gregate microstructures (micelles, spontaneous vesicles, andlamellar phases), but most of the studies are limited to thesolubilization and phase behavior [16–23]. Generally, catan-ionic surfactants are applied to prepare mesoporous materialsby forming vesicles in aqueous solutions [24–27]. The prepara-tion of nanomaterials via catanionic reverse micelles process islimited. Shi and co-workers [28,29] reported the hierarchicalsuperstructure consisting of BaCrO4 nanobelts and polymer-directed penniform BaWO4 nanostructures in reverse micelles.But the influence of catanionic surfactants to materials proper-ties needs to be discussed.

In this article, a catanionic surfactant system PN (Scheme 1)that consisted of equal mole of di(2-ethylhexyl) phosphoric acid(DEHPA) and primary amine (N1923) was prepared, and thendispersed to form reverse micelles to prepare CeF3 NPs. Theproperties of PN and reverse micelles were characterized by theinterfacial tension measurement, DLS and FTIR. The propertiesof CeF3 NPs were characterized by TEM, XRD, FTIR, and flu-orescence spectroscopy. To the best of our knowledge, this isthe first report on ultrafine modified nanoparticles synthesis viacatanionic reverse micelles.

510 H. Zhang et al. / Journal of Colloid and Interface Science 302 (2006) 509–515

Scheme 1. The structures of DEHPA, N1923, and PN.

2. Experimental

2.1. Materials

DEHPA (purity > 98%) and primary amine N1923 (purity> 99%) were from Tianjin Chemical Reagents Company andused as received. The average molecular weight of DEHPA andN1923 is 299.4 and 259.8, respectively. The catanionic surfac-tant PN was prepared by mixing equal mole of DEHPA andN1923. All other reagents were of analytical reagent grade.

2.2. Preparation and characterization of catanionic surfactantand reverse micelles

The stock solution (0.5 mol/L) of the surfactant (PN,DEHPA, or N1923) was prepared by dissolving known weightof regents in n-heptane. Organic solutions with different con-centration were prepared by volume dilution method from thestock solution.

2.3. FTIR spectra measurements

DEHPA, N1923, and PN samples were performed using aBIO-RAD model FTS-40 IR spectrometer in the range 400–4000 cm−1. The samples were pressed KBr pellets for the spec-tral measurements.

2.4. Interface tension measurements

Organic solutions with different concentration equilibratedwith equal volumes of pure water. The oil/water interfacetension was measured by the Du Nouy ring method with aSigma701 tensiometer (KSV Instrument) at 25 ± 1 ◦C.

2.5. Light scattering experiments

10 mL of 0.15 mol/L PN, DEHPA and N1923 heptane so-lution mixed with 0.1 mL pure water and ultrasonicated for20 min, respectively. The diameters of reverse micelles in dif-ferent systems were evaluated by dynamic light scattering usinga Malvern Instruments Zetasizer 1000 HS.

2.6. Preparation and characterization of CeF3 nanoparticles

In a typical procedure, 8 mL of 0.15 mol/L PN heptane so-lution and 2 mL of 0.149 mol/L Ce(NO3)3 aqueous solution(pH 0.75) were mixed in a tube and shaken for 20 min. Themicroemulsion I (organic phase) was obtained by centrifugingthe mixture, and then water phase was collected. The final con-centration of Ce3+ in the microemulsion was obtained by theinitial amount of Ce3+ minus the amount of Ce3+ in the resid-ual aqueous phase.

The microemulsion II was prepared by mixing 50 mL of0.15 mol/L PN heptane solution, 45 mL of n-butanol and 3 mLof 2 mol/L NH4F aqueous solution, then ultrasonicated for20 min. The finial concentration of F− in the microemulsionis about 0.061 mol/L.

The two microemulsions were mixed together with Ce:Fmole ratio 1:3 under ultrasonication and kept for 10 min. Theprecipitation was washed thoroughly with ethanol and dried for3 h at 100 ◦C in the air.

In a control experiment, normal CeF3 NPs was prepared di-rectly by mixing 10 mL of 0.1 mol/L Ce(NO3)3 and 30 mLof 0.1 mol/L NH4F aqueous solutions. The mean diameter ofnormal CeF3 NPs is about 16 nm measured by TEM.

The diameters of reverse micelles in microemulsions I andII were evaluated by dynamic light scattering using a MalvernInstruments Zetasizer 1000 HS. The final product (cappedCeF3 NPs) was also dispersed in ethanol and ultrasonicatedfor 20 min to measure the size of NPs. The phase purity ofthe CeF3 particles was examined by X-ray diffraction (XRD)measurement performed on a Rigaku D/max-II B X-ray diffrac-tometer with monochromatic CuKa radiation (λ = 1.5418 nm).The step-scan covered the angular range from 20◦ to 80◦ in stepof 0.02◦ and the scanning rate of 4.0◦/min using silicon as in-ternal standard. The morphology of the final products was char-acterized by TEM (JEM-2010, JEOL). Powdered samples weredispersed in absolute ethanol by ultrasonication, dripped anddried on a copper grid with a supported thin carbon film. Theresults of FTIR spectroscopy was obtained via a FTS-40 Fouriertransform infrared spectrometer in the range 400–4000 cm−1.The samples were pressed KBr pellets for the spectral measure-ments. The fluorescence spectra of samples were measured bya Hitachi F-4500 FL spectrophotometer.

All experiments were carried out at room temperature (25 ±1 ◦C).

H. Zhang et al. / Journal of Colloid and Interface Science 302 (2006) 509–515 511

Fig. 1. The FTIR spectra of DEHPA, N1923, and PN heptane solutions.

Fig. 2. The dependence of the tension (γ ) on the concentration (log c) of differ-ent surfactants.

3. Results and discussion

3.1. Characterization of the catanionic surfactant

Fig. 1 shows the FTIR spectrum of DEHPA, N1923, andPN. The peaks at 2335.8 cm−1 (νP–O–H) and 1617 cm−1 (δNH2 ),which are the characteristic peak of DEHPA and N1923, re-spectively, are not found in the PN spectrum whereas new peaksat 1557 cm−1 and 1645 cm−1 (δNH3+ ) appear. It indicates thatthe DEHPA and N1923 have completely reacted and formed acatanionic surfactant (PN) [30,31].

As the tension measurement shown in Fig. 2, the catanionicsurfactant PN exhibits a lower oil/water interface tension thanthe individual components, DEHPA or N1923, in the concentra-tion region. As the R–NH2 and (RO)2POOH become R–NH+

3and (RO)2POO−, the catanionic surfactant have more polar andbe more hydrophilic than its components. It tends to be con-

centrated at the interface, whereas the DEHPA and N1923 arehydrophobic and tends to stay in the organic phase. Over therange of catanionic surfactant concentration from log c = −3.5to −1.5, the interfacial tension decreases dramatically with theincrease of surfactant concentration due to the amount of thecatanionic surfactant adsorbed at the interface rapidly increas-ing. In the region log c > −1.5 of the catanionic surfactantconcentration, the interfacial tension is nearly constant with anincrease in the surfactant concentration. In this situation, the in-terface has been fully occupied by the surfactant molecules andthe adsorption of the surfactant at the interface is almost satu-rated. Further increasing the surfactant concentration will causealmost all the additional surfactant molecules to dissolve in theorganic phase and form reverse micelles. Due to the electrosta-tic interactions among the polar groups, the polar headgroupsof the catanionic surfactant will automatically approach eachother and extend into the inner water pool, whereas the hy-drophobic hydrocarbon chains will extend into the nonpolarsolvent. This mechanism is similar to a simple surfactant, asRef. [32] represented, but the difference is that the oppositelycharged surfactant heads play an important role in minimizingrepulsions between the same charged surfactant heads and de-creasing the interface tension [33]. These results are in accordwith the reference [34] that the mixed surfactants are more ef-fective than a single surfactant in forming microemulsions withlow surfactant content. Due to the higher surface activity of acatanionic surfactant, water is more easily dispersed and formssmaller reverse micelles in organic phase. It is conformed by thelight scattering experiments that the reverse micelles (22 nm) inPN microemulsion are smaller than that in DEHPA (223.9 nm)or N1923 (149.6 nm) emulsion.

Solvated ions in reverse micelles formed from ionic surfac-tants cause a contraction of RM (radius of reverse micelles),due to interactions with the charged surfactant heads, and tendto make the micelles more spherical. Moreover, the influenceof solvated ions on RM tends to increase with increasing con-centration and charge [35]. As Ce3+ has three charges, it is rea-sonable that the reverse micelles in microemulsion I are muchsmaller than that in the original microemulsion (mixed withpure water). Similar process occurs in microemulsion II. TheDLS measurements (Fig. 3) accord with the above conclusion.

3.2. TEM photograph and XRD pattern of CeF3 NPs

Fig. 4 shows the TEM image of CeF3 NPs. It can be seenthat the particles are nearly monodispersed with an averagediameter of 1.8 nm. The size of products is much smallerthan the results of Refs. [36,37], in which other nanoparticleswere prepared using HDEHPA or N1923 microemulsions andRefs. [11,12] in which CeF3 NPs prepared with other surfac-tant (CTAB or T154) microemulsions approach. In this work,due to the strong electrostatic attraction between catanionic sur-factant molecules, the reverse micelles are very firm and act asconfining microreactors. Furthermore, the surfactant moleculesadsorbed onto the surface of the ultrafine particles due to thestrong interactions also prevent particles growing.

512 H. Zhang et al. / Journal of Colloid and Interface Science 302 (2006) 509–515

Fig. 3. DLS results of different samples: (1) microemulsion I (average size 1.77 nm); (2) microemulsion II (average size 5.03 nm); (3) the modified CeF3 NPsdispersed in ethanol (average size 2.87 nm).

Fig. 4. TEM photographs of modified CeF3 nanoparticles.

The XRD pattern of the products dried at 100 ◦C (Fig. 5b)gives a very wide peak (noncrystal) containing some crystalpeaks of hexagonal CeF3, such as the diffraction peaks of (110),(111), (300) and (113) crystal plane (marked with stars), whichcan be explained by a mixture of CeF3 crystallite and noncrys-tal organic surfactant. After particles are calcined at 800 ◦C for30 min in air, the XRD pattern of residue (Fig. 5c) agrees wellwith the hexagonal CeF3 phase (PDF08-0045). As estimated byScherrer equation, the size of particles has grown up to 93.2 nmafter the process of removing the organic surfactant by calci-nation. This result indicates the original small particles havegrown up.

Scheme 2 shows the model illustrating the formation processof CeF3 nanoparticles in reverse micelles. In w/o microemul-sions, the aqueous phase is dispersed as microdroplets sur-rounded by a layer of surfactant molecules in the continuousorganic phase. When particles are formed, this layer still coversthem and confines their final size.

3.3. The FTIR spectroscopy of modified CeF3 NPs

Fig. 6 shows the FTIR spectra of different CeF3 samples. Ascompared to normal CeF3 NPs, all groups of the catanionic sur-factant, such as long carbon chain, phosphate and amine salt

H. Zhang et al. / Journal of Colloid and Interface Science 302 (2006) 509–515 513

Scheme 2. Model illustrating the formation process of CeF3 NPs in reverse micelles.

Fig. 5. XRD patterns of different samples: (a) standard peaks of CeF3(PDF08-0045); (b) modified CeF3 NPs dried at 100 ◦C; (c) the residue afterthe modified CeF3 NPs are calcined at 800 ◦C.

Fig. 6. The FTIR spectra of modified CeF3 NPs and normal CeF3 NPs.

Table 1The νP=O peak in different samples

Samples DEHPA PN PN/C4H9OH/heptane/H2Omicroemulsion

NaDEHP/C5H11OH/Kero/H2Omicroemulsion [31]

ModifiedCeF3 NPs

Peak(cm−1)

1230 1224 1211 1209.5 1203

groups are found in the spectrum of modified CeF3 NPs. Thepeaks at 2931 and 2858 cm−1 are assigned to carbon chain vi-bration ν(CH2)n while the peaks at 1063, 1203, and 1010 cm−1

are due to νP–O–C, νP=O, and νC–N, respectively. Furthermore,a strong shoulder peak at 3250 cm−1 can be attributed to νNH3+ .It is speculated that the surfactant molecules are adsorbed ontothe surface of the particles to form organic modification layers.

The surfactant cannot be removed simply by washing withalcohol. This is attributed to the following two facts: One is thatthe catanionic surfactant molecules assembled on the nanopar-ticles surface due to the strong electrostatic attraction betweenthe oppositely charged headgroups of P–O− and NH+

3 . Ob-viously, this kind of passivating layer hardly exists for a sin-gle surfactant system due to the repulsion of headgroups. Theother is chemical interaction between surfactant molecules andsurface atoms. The atoms on nanoparticles surface are bond-unsaturated and the phosphate groups of PN molecules havethe ability to coordinate with cerium [38]. PN molecules canbe chemically bonded to the surface of CeF3 nanoparticles.In those similar studies [39–41], phosphates (phosphate headgroups) and amines as protecting agents and the chemical bondsbetween particles and surfactants are also found.

Such strong interactions have been reflected in their FTIRspectra as shown in Table 1. The peak of νP=O in differentconditions shift to lower energy by the order: H < R–NH+

3 <

R–NH+3 (hydrated) < Na+ < Ce3+ (on CeF3 NPs surface). Due

to higher polarizing power of the Ce3+ cation, the P=O peak ofsurfactant molecules on the CeF3 NPs surface shifts from 1224to 1203 cm−1 in comparison with free PN molecules.

514 H. Zhang et al. / Journal of Colloid and Interface Science 302 (2006) 509–515

Fig. 7. Fluorescence spectra of different CeF3 samples: (a) excitation spectrumof modified CeF3 NPs measured at 344 nm; (b) emission spectrum of modi-fied CeF3 NPs excited at 254 nm; (c) excitation spectrum of normal CeF3 NPsmeasured at 327 nm; (d) emission spectrum of normal CeF3 NPs excited at280 nm.

3.4. Fluorescence spectroscopy of modified CeF3 NPs

Fig. 7 illustrates an emission spectrum of modified CeF3NPs (curve b) ranging from 300 to 400 nm with a peak at 344nm, which can be attributed to the emission of perturbed Ce3+.In contrast to normal CeF3 NPs (curves c, d) with a diameterof about 16 nm, the modified CeF3 NPs have a broader exci-tation range (curve a) and the peak position of excitation andemission has a red shift of 17 and 18 nm, respectively. More-over, the red-shift in peak position is 19 nm relative to that ofCeF3 NPs (size 15–20 nm) prepared by CTAB microemulsions[12] and 54 nm as compared to that of CeF3 bulk materials [42].Such great shift in the emission spectra can be ascribed to theeffects of small size as well as chemical modification. Becausecerium(III) has single 4f electron, easily affectable 5d orbit andbroad and tunable 5d–4f emission, its emission spectrum is verysensitive to the environment. In our case, owing to the coordi-nation between Ce(III) and PN, the energy gap between 5d and4f orbit decreases, making the peaks of excitation and emissionspectra red-shift. As Ref. [43] reported, the surface state satura-tion is modified by particle–surfactant interactions, which maydiffer between the single surfactant case and the mixed surfac-tant case.

4. Conclusion

A catanionic surfactant PN was used to form reverse mi-celles to prepare CeF3 NPs. It had higher surface activity andcould form smaller reverse micelles than that of the individualcomponents. Furthermore, we synthesized the ultrafine CeF3NPs modified by the catanionic surfactant via a reverse micellesroute. The product was narrow distributed with average sizeof 1.8 nm. FTIR spectra demonstrated that there was a strongchemical attraction between surfactant and nanoparticles sur-face. The modification as well as small size of the CeF3 NPs

could result in a huge red shift (18 nm in contrast to normalCeF3 NPs and 54 nm in contrast to bulk materials) of the emis-sion peak in fluorescence spectroscopy.

Although CeF3 NPs and catanionic surfactant were reportedpreviously, but generally, catanionic surfactants are applied toprepare mesoporous materials by forming vesicles in aqueoussolutions. The preparation of nanosized materials via catanionicreverse micelles process is very limited and then needs furtherresearches. Here we prepared ultrafine CeF3 NPs via a reversemicelles route and found the product was modified by the catan-ionic surfactant. The modified ultrafine CeF3 NPs showed someworthy properties in the fluorescence emission due to the effectsof small size and chemical modification. As cerium fluoride hasmany valuable properties, our paper gives a possible way toa better material by one-step method consisting of preparationand modification in a mild condition.

Acknowledgments

This project is supported by the National Program enti-tled “Basic Research on high-efficiency utilization of rareearths elements in the field of environmental modification”(2004CB719506) and National Natural Science Foundationof China (20371046) and National “863” project (2002AA-647070).

Supplementary material

The online version of this article contains additional supple-mentary material.

Please visit DOI: 10.1016/j.jcis.2006.06.062.

References

[1] K. Tanaka, K. Fujita, N. Soga, J. Qiu, K. Hirao, J. Appl. Phys. 82 (1997)840.

[2] K. Shimamura, E.G. Víllora, S. Nakakita, M. Nikl, N. Ichinose, J. Cryst.Growth 264 (2004) 208.

[3] R.A. Glukhov, A.N. Belsky, C. Pedrini, A.N. Vasilev, J. Alloys Compd.275–277 (1998) 488.

[4] K.V. Narayana, B. David Raju, S. Khaja Masthan, V. Venkat Rao, P. KantaRao, A. Martin, J. Mol. Catal. A 223 (2004) 321.

[5] S. Qiu, J. Dong, G. Chen, Wear 230 (1999) 35.[6] H.E. Aldorf, Rare Earth Halide Grease Compositions, U.S. patent No.

4507214, 1995.[7] R. Lo Nigro, G. Malandrino, I.L. Fragalà, M. Bettinelli, A. Speghini,

J. Mater. Chem. 12 (2002) 2816.[8] C. Dujardin, C. Perdrini, N. Garnier, A.N. Belsky, K. Lebbou, J.M. Ko,

T. Fukuda, Opt. Mater. 16 (2001) 69.[9] Z. Wu, X. Tao, J. Zhou, H. Dang, Chin. Rare Earths 22 (2001) 1.

[10] S. Eiden-Assmann, G. Maret, Mater. Res. Bull. 39 (2004) 21.[11] S. Qiu, J. Dong, G. Chen, Powder Technol. 113 (2000) 9.[12] H. Lian, M. Zhang, J. Liu, Z. Ye, J. Yan, C. Shi, Chem. Phys. Lett. 395

(2004) 362.[13] B. Zhu, H. Shi, J. Huang, X. He, Acta Chim. Sin. 59 (2001) 913.[14] S. Karlsson, R. Friman, M. Björkqvist, B. Lindström, S. Backlund, Lang-

muir 17 (2001) 3573.[15] B. Joensson, P. Jokela, A. Khan, B. Lindman, A. Sadaghiani, Langmuir 7

(1991) 889.[16] F. Al-Ali, A. Brun, F. Rodrigues, G. Etemad-Moghadam, I. Rico-Lattes,

Langmuir 19 (2003) 6678.[17] S. Segota, S. Heimer, D. Tezak, Colloids Surf. A 274 (2006) 91.

H. Zhang et al. / Journal of Colloid and Interface Science 302 (2006) 509–515 515

[18] B. Yue, C. Huang, M. Nieh, C.J. Glinka, J. Katsaras, J. Phys. Chem. B 109(2005) 609.

[19] W. Yu, Y. Yang, C. Chang, Langmuir 21 (2005) 6185.[20] S. Yeh, Y. Yang, C. Chang, Langmuir 21 (2005) 6179.[21] X. Li, H. Kunieda, Curr. Opin. Colloid Interface Sci. 8 (2003) 327.[22] H. Chakraborty, M. Sarkar, Langmuir 20 (2004) 3551.[23] R.M. Minardi, P.C. Schulz, B. Vuano, Colloids Surf. A 197 (2002) 167.[24] B. Prelot, T. Zemb, Mater. Sci. Eng. C 25 (2005) 553.[25] S. Pevzner, O. Regev, A. Lind, M. Linden, J. Am. Chem. Soc. 125 (2002)

652.[26] H. Hentze, S.R. Raghavan, C.A. McKelvey, E.W. Kaler, Langmuir 19

(2003) 1069.[27] A. Lind, B. Spliethoff, M. Linden, Chem. Mater. 15 (2003) 813.[28] H. Shi, L. Qi, J. Ma, H. Cheng, J. Am. Chem. Soc. 125 (2003) 3450.[29] H. Shi, L. Qi, J. Ma, H. Cheng, B. Zhu, Adv. Mater. 15 (2003) 1647.[30] D. Li, Z. Wang, G. Zeng, He Huaxue Yu Fangshe Huaxue 6 (1984) 153.[31] X. Fu, Y. Pan, H. Xin, Z. Hui, H. Yu, Acta Chim. Sin. 53 (1995) 178.[32] X. Song, S. Sun, Z. Yin, W. Zhang, Y. Yang, Colloids Surf. A 209 (2002)

57.

[33] K. Yang, X. Shi, S. Duan, Surf. Industry 9 (1996) 18.[34] X. Li, E. Lin, G. Zhao, T. Xiao, J. Colloid Interface Sci. 184 (1996) 20.[35] B.L. Cushing, V.L. Kolesnichenko, C.J. O’Connor, Chem. Rev. 104 (2004)

3893.[36] X. Song, S. Sun, W. Zhang, Z. Yin, J. Colloid Interface Sci. 273 (2004)

463.[37] X. Song, S. Sun, W. Fan, Z. Yin, Mater. Lett. 57 (2003) 3026.[38] D.F. Peppard, in: F.H. Spedding, A.H. Daane (Eds.), The Rare Earth, Wi-

ley & Sons, New York, 1961, pp. 38–54.[39] C. Yee, G. Kataby, A. Ulman, T. Prozorov, H. White, A. King, M. Rafai-

lovich, J. Sokolov, A. Gedanken, Langmuir 15 (1999) 7111.[40] M. Textor, L. Ruiz, R. Hofer, A. Rossi, K. Feldman, G. Hahner, N.D.

Spencer, Langmuir 16 (2000) 3257.[41] D. Shieh, F. Cheng, C. Su, C. Yeh, M. Wu, Y. Wu, C. Tsai, C. Wu, D. Chen,

C. Chou, Biomaterials 26 (2005) 7183.[42] W.W. Moses, S.E. Derenzo, M.J. Weber, A.K. Ray-Chaudhuri, F. Cerrina,

J. Lumin. 59 (1994) 89.[43] B.A. Simmons, S. Li, V.T. John, G.L. McPherson, A. Bose, W. Zhou,

J. He, Nano Lett. 2 (2002) 263.