**2.4.2.4 Reaction time**

By observing brightness and reflectance versus reaction time based on the Taguchi method, it can be observed that the maximum of lightness and reflectance are related to reaction time of about level 2 (Fig. 2.10).

Fig. 2.10. Effect of time parameter on the a) lightness and b) reflection, at different levels

As SEM analysis show in figures 2.11-a and 2.11-b, when reaction time is low a few coating of SnO2 are formed on the mica flakes, thus lightness and reflectance are decreased (figure 2.11-a). While increase of reaction time, until level of 2, a uniform SnO2 coating is formed on the mica flakes and thus, lightness and reflectance increase as shown in figure 2.13-b. However, increasing time more than level 2 causes decreasing of lightness and reflectance, because the SnO2 coatings formed on the mica flakes are separated gradually.

Ceramic Coatings for Pigments 253

Figure 2.13 shows the SEM images of optimized sample with surfactant and without it. As figure 2.13-a shows in the sample containing surfactant, SnO2 particles coat mica flakes uniformly without any agglomeration. In addition, these particles agglomerate on mica flakes only and are not formed freely among mica flakes, but in the sample without surfactant, SnO2 particles formed irregularly and are agglomerated on mica flakes and also

(a) (b)

Surfactant molecules are able to surround a small volume of suspension including seed and due to repulsive forces between electrical charged boundaries of mica and SnOH particles, prevent the over sticking of particles and growth of agglomerates [2.13, 2.18, 2.19]. In addition the mica boundary and hydrated metal oxide formed by hydrolysis is activated with increasing amount of surfactant making for a smooth surface and stable adsorption [2.20].

After performing experiments designed by Taguchi method and analysing results with Winrobust software the optimized sample was provided using the following condition:

The XRD pattern of this sample shows the hydroxide phase before calcination and Casiterite

According to SEM results, SnO2 particles are coated on mica flakes almost regularly and uniformly (figure 2.13-a). According to the laser beam diffraction technique the mean particle size of pigments is about 65 micron. The density of this pigment is 3.3 g/cm3 and the lightness and reflective percent are 97% and 89% respectively in which case these amounts of l\* and R are equal to the Winborust software results. Consequently choosing to

Fig. 2.13. SEM images of coated mica flakes in optimized sample a) with surfactant in

magnification of 1 KX and b) without surfactant in magnification of 500 X

pH = 2.5, T = 65°C, C = 40 g/lit, t = 7 h, R = 200 r.p.m .

use the Taguchi method and L16 algorithm is correct for this work.

**2.4.2.6 Surfactant effect** 

**2.4.3 Optimized sample** 

SnO2 phase after that.

freely among them (Figure 2.13-b).

Fig. 2.11. SEM images of pigments synthesized at different levels of reaction time

## **2.4.2.5 Stirring rate**

By observing brightness and reflectance versus stirring time based on (Figures 2.12-a and 2.12-b) Taguchi method, it can be seen that the maximum of lightness and reflectance are related to a stirring rate of about level 2.

Fig. 2.12. Effect of stirring rate parameter R on the a) lightness and b) reflection at different levels

The stirring rate affects the densification of the membrane layer directly. If the stirring rate is too low, the reaction solution can not form sufficient turbulence [2.17], then the microscopic mixture is uneven, and the generated crystal particles of SnO2 are of various sizes. On the other hand, if the stirring rate is too high, it will affect the rate of growing of the crystal nucleus, with the result that some tiny colloidal micro-particles enter the solution through the filter paper rather than depositing on the surface of mica flakes, which will cause the light to scatter. Therefore, the stirring rate in this reaction should be of level 2.
