Methods Comparison

Samples which developed by solid state reaction were denoted as ZSpV series, the other trials denoted as ZSpI, ZSpIII and ZSpIV will be described. Comparing with the solid state reaction, which is a very common, easy, and cheap process, and the results of the other three methods, the Zn2SiO4:Mn (ZSpII) by colloidal method shows an extremely good PL property.

The XRD results of ZSpV (Zn2SiO4:Mn made of solid state reaction) included not only cristobalite and Zn2SiO4 phases, but also ZnO phase. Three diffraction peaks of ZnO located at 2θ = 31.7^o, 34.4^o, and 36.2^o with the corresponding (hkl) = (100), (002), and (101), respectively, as shown in Fig. 5- 1. The Zn2SiO4 phase has appeared by calcined at 1000^oC, but not much changes in the intensity if calcined to higher temperatures. The effect of re-grinding/re-calcination (the treatment cycle) of ZSpV is shown in Fig. 5- 2, which shows slight decrease of ZnO as the treatment cycle increases from 1 to 3 times. The reaction of crystalline ZnO with amorphous SiO2 is sluggish.

The difference between ZSpII and ZSpV could be easily distinguished in Fig. 5- 3.

If ZSpV is ground and re-annealed for 1 to 3 cycles, the Zn2SiO4 phase is apparent, but the ZnO phase still exists (Fig. 5- 2). The un-desirable phase (residue ZnO) existing in

ZSpV series does reduce the light emission property, as shown in Fig. 5- 3 in much poor contrast to that of ZSpII. Also, the PL spectra of those samples are shown in Fig. 5- 4.

Increasing the homogeneity by grinding/annealing for three cycles does enhance the emission intensity, but slightly. In fact, the emission intensity for ZSpII is much stronger than that of ZSpV-3. It can be speculated that the second phase existing in PL material may reduce the emission intensity due to:

(1) Lattice deformation or distortion that affected Mn²⁺ ions diffused or substituted the Zn²⁺ sites. The occurrence of this may reduce the efficiency of Mn²⁺ activation.

(2) The second phase provides another media with different reflection index that can adsorb or scatter the emission light. The reflection index of the matrix silica is 1.39.

In addition, the purpose of this study is to construct a PBG crystal based on Zn2SiO4:Mn particles that required PL particles in a uniform size (and also shape) arranged in a periodic way. Solid state reaction synthesizes a powder normally in irregular shape (as shown in Fig. 5- 5), and does not satisfy this requirement.

Besides, the methods of ZSpI, ZSpIII, and ZSpIV were also tried in this study.

ZSpI also started from mono-dispersed SiO2 suspension in basic solution (the pH value is about 11 in order to provide a repulsive force between SiO2 particles). Unfortunately, the Zn(NO3)2 and Mn(NO3)2 solution that added into the suspension precipitated out in

precipitation behavior of Zn and Mn species has been mentioned in section 4.1. This result did not satisfy our requirement. Therefore, a modified method (ZSpII) was proposed, using a dried SiO2 suspension as indicated in section 3.2.

ZSpIII was an idea from Y. J. Chen [62]. The method started from ZnO-B2O3-SiO2

ternary system. This idea was to grow Zn2SiO4 crystals from melted Zn-B-Si oxide liquid at elevated temperature. The powder reactants were turbo mixed and then heated to 1300^oC to get Zn-B-Si glass. Boron oxide (B2O3) in this system acted as a flux which was used to lower the glass forming temperature. The higher the concentration of B2O3, the lower the glass forming temperature of the system. The sample did totally transform to glass (amorphous) phase, as shown in Fig. 5- 7. Annealing was required to get Zn2SiO4 crystalline phase. Therefore, the ground glass powder was annealed at 800^oC for several hr. However, there was an additional un-expected phase appearing in the system, zinc borate (Zn4O(BO2)6). The morphology of the thermally treated ZSpIII is shown in Fig. 5- 8. Three different contrasts were seen in the BSE (back-scattering-electronic) images, brightest, gray, and dark features. Compared with the XRD patterns, the brightest features (marked as X) should be the Zn2SiO4; the gray one (marked as Y) should be the Zn4O(BO2)6 due to the boron element, which shows a smaller atomic number and be darker under BSE mode; the dark feature is porosity. The porosity could be resulted from the crystallization of Zn4O(BO2)6 phase competing with

sintering [72]. If crystallization occurs before sintering is finished, the viscosity of glass matrix increases sharply and the sintering stops. Therefore, a porous glass-ceramic was produced, resulting in a higher porosity in Fig. 5- 8(b) than in (a). The Zn2SiO4 grains prepared by ZSpIII method were fixed in the Zn-B-Si matrix, and were hardly purified by acid etching. Therefore, this process was given up.

ZSpIV, a hydro-thermal method, started with the dissolution of ZnCl2 (ZnO dissolved in HCl) and Na2SiO3(aq) in aqueous solution. Two chemicals were mixed in Teflon tubes and heated by a micro-wave sample preparation system with a pressure of 60 bar, a temperature of 240^oC by an input power of 1200 W holding for 4 hr. Generally, samples prepared by hydrothermal method should be crystallized after reaction, but ZSpIV sample is still amorphous. Therefore, the product was dried and annealed at about 400^oC for 1 hr to see if there’s anything occurred. The annealed sample did crystallize. However, a wrong product (see Fig. 5- 9) was made, to be sodium zinc chloride hydrate (Na2ZnCl4(H2O)3, #731757). Na⁺ and Cl^- ions reacted in this system instead of SiO32-. Therefore, this process did not synthesized Zn2SiO4 successfully and was given up.

Base on those experimental results, we conclude that the ZSpII colloidal process is a proper method to synthesize Zn2SiO4:Mn particles.

15 20 25 30 35 40 45 50 55 60 65 70 75

15 20 25 30 35 40 45 50 55 60 65 70 75

15 20 25 30 35 40 45 50 55 60 65 70 75

ZnO-653411 Zn₂SiO₄ 37-1485 ZSpV1200

Intensity

2 Theta

ZSpV900 ZSpV1000 ZSpV1100 (101)

(002) (100)

Fig. 5- 1 XRD patterns illustrating the Zn2SiO4:Mn particles prepared by solid state reaction. The patterns were arranged by different annealing temperatures from 900-1200^oC for 2 hr (bottom to top).

15 20 25 30 35 40 45 50 55 60 65 70 75

Fig. 5- 2 XRD patterns illustrating the Zn2SiO4:Mn particles prepared by solid state reaction, which samples were ground and re-annealed for one to three times at 1100^oC for 2 hr.

Fig. 5- 3 Photos showing the Zn2SiO4:Mn disks (1100^oC for 2 hr) exposed (a) without, and (b) with applied 254 nm UV light, where the left piece and right ones were synthesized via solid state reaction and colloidal process, respectively.

(b) (a)

ZSpV-3 ZSpII

450 500 550 600 650

Fig. 5- 4 PL spectrum of the emission intensity of Zn2SiO4:Mn disks prepared by solid state reaction (ZSpV series) and solution process (ZSpII). All of these samples were annealed at 1100^oC for 2 hr. The number following the sample notation (ZSpV) is the cycles of grinding and re-annealing for one (square), two (circle) and three (triangle)

Fig. 5- 5 SEM image showing irregular shape of Zn2SiO4:Mn powder (ZSpV) synthesized via solid state reaction at 1000^oC for 2 h.

Fig. 5- 6 SEM image showing ZSpI samples annealed at 1000^oC for 2 hr. The Zn and Mn ions precipitated randomly, instead of uniformly coating on SiO2 particles surface.

ZSpIII-annealing for 5 hr

Fig. 5- 7 XRD patterns showing crystalline phase of ZSpIII powder series after quenching (top) and 800^oC annealing (the other two patterns). The product contained not only Zn2SiO4, but also a Zn4O(BO2)6 phase.

Fig. 5- 8 SEM images showing the morphology of ZSpIII series after calcination at 800^oC for (a) 5 hr and (b) 10 hr, where (a2) and (b2) were imaged under BSE mode. X area is Zn rich region; while Y area is B rich region. The black area is pores that are resulted from crystallization in the glass.

a1

b1 b2

a2 X Y

15 20 25 30 35 40 45 50 55 60 65 70 75

Intensity, a.u.

2 Theta

ZSpIV-400^oC

Na₂ZnCl₄(H₂O)₃ #73-1757

Fig. 5- 9 XRD patterns showing the crystalline phase of ZSpIV annealed at 400^oC for 2 hr. The crystalline phase is indexed as Na2ZnCl4(H2O)3.