Soft solution processing of cerium hydroxysulfate powders with different morphologies

Abstract

Polycrystalline cerium hydroxysulfate powders have been prepared by soft solution processing using various basic solvents. The crystals prepared have varying morphologies, spherical and flaky, depending on the solvent used. The crystals obtained from distilled water–pyridine and aqueous ammonia solvent mixtures are spherical, whereas those obtained from mixtures of distilled water and ethylenediamine or hydrazine hydrate are flaky. All the crystalline cerium hydroxysulfate samples display luminescence properties. It was found that the flaky crystals generally show a much stronger luminescence than their spherical counterparts.

---

Introduction

Rare earth metal hydroxysulfates, including LaOHSO₄, PrOHSO₄, and NdOHSO₄, were first obtained from rare earth oxide–hydroxide–sulfate systems above 400 °C.¹ The thermal decomposition curve for LaOHSO₄ shows that it does not decompose in a nitrogen atmosphere up to 550 °C, which indicates good thermal stability for this material and other rare earth metal hydroxysulfates. Cerium hydroxysulfate (CeOHSO₄) was first prepared via the thiourea–cerium(III) nitrate system below 250 °C,² a much lower temperature than that required to prepare similar hydroxysulfate compounds, as noted above. Furthermore, the hydroxysulfate compounds were studied, for the first time, as potential optical materials. CeOHSO₄ powders are flaky, showing a preferential (0k0) orientation and a broad-band emission at ca. 388 nm of considerable intensity;² La_1 − xCe_xOHSO₄ powders have diverse luminescence intensities and average particle sizes which are dependent on the chemical composition of the solid solution.³ The luminescence curve of CeOHSO₄ powders corresponds well to the spectral sensitivity curve for common insects in the near-ultraviolet region,⁴ which suggests that one of the possible uses of the hydroxysulfate material is, or will be, as a black light radiation material, e.g. a bug killer, owing to its luring power to general insects.

In the thiourea–cerium(III) nitrate system, a basic medium was found to be favorable for the formation of CeOHSO₄ under hydrothermal conditions. The initial solution is acidic (pH = 5), nevertheless, a basic environment (pH = 8.5) is achieved by thermal decomposition of thiourea. The sulfate anion, which is essential to the reaction, can only be produced in a closed system via the oxidation of H₂S at temperatures above 160 °C.⁵ The preparation of solid CeOHSO₄ of good purity was realized over 200 °C. At lower temperatures, the system produces cerium hydroxycarbonates displaying much weaker luminescence.² It was supposed that CeOHSO₄ would be produced at a lower temperature, by directly using a sulfate salt as a precursor material and employing an appropriate solvent to make the starting solution basic. In addition, it is known that phosphor powders with different morphologies can have different luminescence performances, owing to their specific microstuctures.⁶ Therefore, a soft solution process for the preparation of phosphor powders with different morphologies as well as different luminescence performances is of interest in the fields of materials chemistry and optical materials. This paper reports the synthesis of CeOHSO₄ powders with different morphologies under even milder condition than those employed in our previous work.

Experimental

Analytical purity Ce(NO₃)₃·6H₂O and (NH₄)₂SO₄ were used as starting materials. The main solvent used in the experiments was distilled water, a basic solvent was added solely for the purpose of adjusting the pH of the source solution. Analytically pure pyridine, aqueous ammonia solution, ethylenediamine, and hydrazine hydrate were specifically selected in order to avoid introducing metal ions to contaminate the products. The reaction was performed in a 30 ml Teflon-lined autoclave. In the preparation, 1 mmol Ce(NO₃)₃·6H₂O and 1.1 mmol (NH₄)₂SO₄ were introduced into the autoclave, then 20 ml distilled water was poured into the reactor to dissolve the reactants. Afterwards, a basic solvent (not pyridine) was added dropwise and further distilled water added while stirring. Finally, a basic solution (pH = 8.5) with a total volume of 25 ml (the autoclave was 85% filled) was obtained. Pyridine is a Lewis base and is the least basic of all the solvents tested, so it was mixed with distilled water in a volume ratio of 1∶3 to serve as the reaction medium. The autoclave was kept at 160–200 °C and the solvothermal process was allowed to proceed for 6 h. Solid products produced using different basic solvents were collected after the reaction, washed several times with distilled water, and dried in vacuo at less than 60 °C for 0.5 h. In order to examine the influence of the starting material on the product, CeCl₃·7H₂O was substituted for Ce(NO₃)₃·6H₂O as an alternative cerium source for the reaction.

FT-IR spectra were recorded at room temperature on a Magna IR 750 spectrometer over the frequency range 4000 to 400 cm⁻¹ with a resolution of 2 cm⁻¹, in order to identify any ionic groups in the compounds. A thin disc was made by compacting the uniformly mixed powders of a 2 mg sample and 198 mg potassium bromide. X-Ray photoelectron spectra (XPS) were obtained with an MK II electron spectrometer, in order to identify the valence states of cerium and sulfur. X-Ray diffraction (XRD) of the powders was conducted on a rotating anode X-ray diffractometer with graphite-monochromatized Cu-Kα radiation, in order to characterize the crystallinity of the powders prepared in various solvents. Scanning electron microscopy (SEM) was performed using an X-650 scanning electron microanalyzer with a 200 kV accelerating voltage.

Luminescence spectra were recorded at room temperature with a Hitachi 850 fluorescence spectrophotometer. In particular, the emission intensity was used to compare the luminescence performances of the samples obtained from the different basic solutions.

Results and discussion

The powders produced from Ce(NO₃)₃·6H₂O and (NH₄)₂SO₄ in various solvents at 175 °C were all hydroxysulfates, and their FT-IR spectra are shown in Fig. 1. The sharp peak at 3490 cm⁻¹ is due to the O–H stretching vibration of the OH⁻ group. The SO₄²⁻ group has four fundamental vibrations: the non-degenerate symmetric stretching mode, ν₁, the doubly degenerate bending mode, ν₂, the triply asymmetric stretching mode, ν₃, and the triply degenerate asymmetric bending mode, ν₄. In a crystal, all the vibration modes could be active, and the absorption bands around 1142, 1133, 1108, 1104, 981, 629, 625, 607, 614, and 451 cm⁻¹ may be a result of splitting of the various vibration modes.^7,8 Therefore, the peaks in the range 1214–599 cm⁻¹ can be ascribed to the SO₄²⁻ group; the deviation of the peak positions from the reference data results from the distinctive chemical environments of different compounds. An analytically pure sample of Ce(SO₄)₂, which is close to CeOHSO₄ in chemical composition, was used as a reference to examine the infrared absorption in the range 1400–580 cm⁻¹. It was found that there are multiple SO₄²⁻ bands around 1240, 1106, 1048, 638, 610, and 590 cm⁻¹, close to those marked in Fig. 1.

Fig. 1 FT-IR spectra of CeOHSO₄ powders prepared using (a) pyridine, (b) ammonia, (c) ethylenediamine, and (d) hydrazine hydrate.

The XPS of the hydroxysulfates are also similar to each other, no obvious impurities were found on the surfaces of the samples. The spectra indicate that the sulfur exists completely in the form of SO₄²⁻ and the cerium exists as Ce(III), with no Ce(IV) all of the samples except for that prepared using ethylenediamine. As shown in Fig. 2, the binding energy for the S 2p orbital of SO₄²⁻ is 169 eV and that for the Ce 3d of Ce(III) is associated with two bands at 883.5–886 eV and 901.5–905.5 eV. Referring to the data collected in literature,⁹ the peak at 917 eV is attributed to Ce(IV). However, the peak is so small that it suggests that the Ce(IV) content is insignificant compared with Ce(III). The small amount of Ce(IV) detected could have resulted from the nitrate, since only a very small amount of ethylenediamine was used to adjust the pH and its reducing ability is not sufficient to counteract the oxidizing nitrate. This is supported by experimental observations that only Ce(III) existed in the hydroxysulfates prepared from the thiourea–cerium(III) nitrate system, in which the excess thiourea guarantees a reducing environment that is unfavorable to Ce(IV).^2,3

Fig. 2 XPS of CeOHSO₄ powder prepared using ethylenediamine. The insets show enlargements of the S 2p and Ce 3d regions.

The XRD patterns of the hydroxysulfates synthesized from Ce(NO₃)₃·6H₂O and (NH₄)₂SO₄ at 175 °C are shown in Fig. 3. The characteristic diffraction patterns demonstrate that the compounds are pure polycrystalline monoclinic CeOHSO₄. Lower reaction temperatures (e.g. 160 °C) resulted in poorly crystallized powders sometimes containing impurities. On the contrary, higher temperatures (200 °C, for example) afforded CeOHSO₄ powders with much better crystallinity. The Ce(IV) in CeOHSO₄ crystals prepared using ethylenediamine brings no substantial change in composition, as demonstrated by both the FT-IR spectrum and XRD pattern of the sample. Alternatively, the Ce(IV) could lead to defect formation in the crystal lattice. The charge could be balanced by excess hydroxyl anions, which are abundant in the basic system in which the compound was synthesized. As can be seen from the diffraction patterns, the samples produced from distilled water containing pyridine or ammonia are similar to each other, but different from those synthesized using ethylenediamine or hydrazine hydrate, both of which show a preferred (0k0) diffraction. Empirically, for crystals with the same chemical composition and symmetry of structure, different XRD patterns usually imply different crystal morphologies.^10,11 Therefore, the powders obtained using pyridine and ammonia [Fig. 3(a) and (b), respectively] should have a similar morphology, but one which may be very different from that of the products prepared using ethylenediamine and hydrazine hydrate [Fig. 3(c) and (d), respectively], which also should have a similar morphology.

Fig. 3 XRD patterns of CeOHSO₄ powders prepared using (a) pyridine, (b) ammonia, (c) ethylenediamine, and (d) hydrazine hydrate.

Fig. 4 shows SEM images of CeOHSO₄ powders prepared from Ce(NO₃)₃·6H₂O and (NH₄)₂SO₄. The product obtained with pyridine contains largely spherical particles [Fig. 4(a)], whereas using ammonia appears to provide crystals which are also spherical, but comparatively smaller and more compact [Fig. 4(b)]. In contrast, the crystals obtained with ethylenediamine are flaky and have a layered structure [Fig. 4(c)], and those produced with hydrazine hydrate are also flaky, but relatively smaller and thinner [Fig. 4(d)]. These results not only demonstrate that the crystal morphology is closely related to the diffraction features, but also confirm that the crystal morphology can be controlled by adding the appropriate basic solvent.

Fig. 4 SEM images of CeOHSO₄ powders prepared using (a) pyridine, (b) ammonia, (c) ethylenediamine, and (d) hydrazine hydrate.

Under the same operating conditions, CeCl₃·7H₂O was used as a reactant, resulting in products similar to those obtained from Ce(NO₃)₃·6H₂O. The crystal morphology of CeOHSO₄ powders produced from CeCl₃·7H₂O is also dependent on the solvent. For example, the product prepared using pyridine contains spherical particles [Fig. 5(a)] and that synthesized with ethylenediamine contains flaky crystals [Fig. 5(b)].

Fig. 5 SEM images of CeOHSO₄ powders prepared from CeCl₃·7H₂O using (a) pyridine and (b) ethylenediamine.

It is clear that, in this case, the solvent, rather than the reactant, plays a more important role in determining the crystal morphology. Solvents with different physical and chemical properties can influence the solubility, reactivity, and diffusion behavior of the reactants,¹² in particular, the polarity and coordinating ability of the solvent can influence the crystal morphology of the final product. Some selected physical and chemical parameters of the solvents used in the experiments are listed in Table 1. From these data, however, it is difficult to definitely elucidate the relationship between the solvent and the crystal morphology, except that it appears that the solvents having only one coordinating site seem favor the formation of spherical crystals, whereas those with two coordinating sites lead to flaky crystals.

Table 1 Selected parameters of the solvents used in experiments

Name of solvent	Dielectric constant, ε (20 °C)	Dipole moment, μ/D	Coordinating sites	pKa (25 °C)	Boiling point/°C
Pyridine	13.26	2.21	1	5.23	115
Ammonia	16.61	1.4718	1	9.25	−33.33
Ethylenediamine	13.82	1.99	2	9.92 (step 1), 6.86 (step 2)	117
Hydrazine	51.7	1.75	2	8.1	113.55, 119 (hydrate)

---

The excitation spectra of CeOHSO₄ samples revealed that the optimal excitation wavelengths were 253 and 280 nm at room temperature. Fig 6 shows the emission spectra (λ_ex = 253 nm) of the CeOHSO₄ powders produced from Ce(NO₃)₃·6H₂O and (NH₄)₂SO₄ in various solvents. Except for the sample prepared using ethylenediamine, all the powders display a broadened peak at ca. 337 nm, besides the peak at ca. 387 nm, of which the latter is stronger. The well-separated peak at 337 nm is due to the extrinsic emission of Ce(III),¹³ yet it was not observed in the CeOHSO₄ powders prepared from thiourea and cerium(III) nitrate at a higher temperature.² Thus, the soft solution process described herein affords products with an additional extrinsic emission. The luminescence of the CeOHSO₄ powders at 387 nm was not observed for bromoborate,¹⁴ floride,¹⁵ oxide, or oxysulfide materials.^16–18 Nevertheless, it is fairly close to that at 420 nm for the oxyorthosilicate, which has been assigned to 4f–5d transitions.¹⁹ Ce(IV) has a 4f⁰ electronic structure, lacking the free electron in the outer shell that is essential to luminescence, hence, Ce(IV) itself normally does not show luminescence. The defect (cation vacancy) which results from the presence of Ce(IV) could associate with Ce(III), producing a broad-band luminescence centered on 373 nm,¹⁴ but no such peak is evident in the luminescence spectrum of the CeOHSO₄ material prepared using ethylenediamine. The luminescence intensities of the CeOHSO₄ powders differ from each other depending on the crystal morphology, or, in other words, on the solvent used for the preparation. Under the same measurement conditions, the spherical crystals prepared using pyridine show a weak luminescence, whereas the flaky crystals prepared with hydrazine hydrate display the strongest luminescence of all the samples. The emission spectra of the polycrystalline CeOHSO₄ powders produced from CeCl₃·7H₂O were recorded for comparison. Likewise, the flaky crystals showed a much stronger luminescence, similar to that shown in Fig. 6.

Fig. 6 Emission spectra of CeOHSO₄ powders (λ_ex = 253 nm) prepared using (a) pyridine, (b) ammonia, (c) ethylenediamine, and (d) hydrazine hydrate.

Numerous experimental results have confirmed that the luminescence properties of phosphor materials are closely related to the morphology, nevertheless the reasons why crystals with different morphologies show different luminescence properties are often explained ambiguously, if at all. A rough but universal explanation is that changes in the sizes and microstructures of crystals with different morphologies cause modification of the electronic structures, influencing the carriers excited from the valence band to the conduction band which then relax their energy on the crystal surface, leading to variations in luminescence. Being a newly discovered material, the hydroxysulfate has been understood merely at a preliminary level. A better understanding of the luminescence properties of CeOHSO₄ powders is needed and further investigations on this topic are under way.

Conclusion

A soft solution processing method for polycrystalline CeOHSO₄ powders with different morphologies has been developed. Both cerium(III) nitrate and cerium(III) chloride were used as cerium sources, reacting with ammonium sulfate in various basic solutions. Spherical crystals were prepared in a solvent made up of distilled water and pyridine or aqueous ammonia, whereas flaky crystals resulted from the use of a solvent composed of distilled water and ethylenediamine or hydrazine hydrate. The powders showed emissions at ca. 337 and 387 nm, of which the luminescence intensity of the latter was the stronger. Moreover, the luminescence of the flaky powders was generally much stronger than that of the spherical powders. Although various basic solvents can be used, hydrazine hydrate proved especially good for the preparation, which afforded the material with the strongest luminescence. In terms of the luminescence intensity, the flaky CeOHSO₄ appears to be a better optical material than the spherical CeOHSO₄.

Acknowledgements

Financial support from the National Natural Science Foundation of China and the Australian Research Council is gratefully acknowledged. The authors thank Ms Ning Chen, Mr Neng Guo and Mr Can Xue for their help in the experiments.

References

1. J. M. Haschke, J. Solid State Chem., 1988, 73, 71.
2. Z. H. Han, Y. Qian, S. Yu, K. Tang, H. Zhao and N. Guo, Inorg. Chem., 2000, 39, 4380.
3. Z. H. Han, C. Zhong, J. Cao, S. Yu, K. Tang, H. Zhao and Y. Qian, J. Cryst. Growth, 2001, 222, 528.
4. Luminescence of Solid State Materials (in Chinese), Jilin Institute of Physics, Changchun, Chinese Academy of Sciences, Beijing, and University of Science and Technology of China, Anhui, 1976.
5. Z. H. Han, Y. Qian, K. Tang, G. Q. Lu, S. Yu and N. Guo, unpublished results.
6. Z. H. Han, N. Guo, K. Tang, S. Yu, H. Zhao and Y. Qian, J. Cryst. Growth, 2000, 219, 315.
7. A. Chtioui, T. Guerfel, L. Benhamada and A. Jouini, Solid State Sci., 2001, 3, 859.
8. G. Hertzberg, Infrared and Raman Spectra of Polyatomic Molecules, Van Nostrand, New York, 1966.
9. C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder and G. E. Muilenberg, Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer Corporation, Norwalk, CT, 1979.
10. Z. H. Han, N. Guo, F. Li, W. Zhang, H. Zhao and Y. Qian, Mater. Lett., 2001, 48, 99.
11. S. Yu, Y. Wu, J. Yang, Z. H. Han, Y. Xie, Y. Qian and X. Liu, Chem. Mater., 1998, 10, 2309.
12. W. S. Sheldrick and M. Wachhold, Angew. Chem., Int. Ed. Engl., 1997, 36, 206.
13. A. J. Wojtowicz, M. Balcerzyk, E. Berman and A. Lempicki, Phys. Rev. B, 1994, 49, 14880.
14. V. Dotsenko, J. Mater. Chem., 2000, 10, 561.
15. W. W. Moses, S. E. Derenzo, M. J. Weber, A. K. Ray-Chaudhuri and F. Cerrina, J. Lumin., 1994, 59, 89.
16. G. Blasse, W. Schipper and J. J. Hamelink, Inorg. Chim. Acta, 1991, 189, 77.
17. S. Yokono, T. Abe and T. Hoshina, J. Lumin., 1981, 24/25, 309.
18. G. A. Slack, S. L. Dole, V. Tsoukala and G. S. Nolas, J. Opt. Soc. Am. B, 1994, 11, 961.
19. W. M. Yen, M. Raukas, S. A. Basun, W. van Schaik and U. Happek, J. Lumin., 1996, 69, 287.

---

Footnote

† Present address: Nanomaterials Centre and Department of Chemical Engineering, The University of Queensland.

---