Introduction

Porcelains are one of the most important glazed or unglazed vitrified whitewares which contain ternary compound of kaolin, feldspar and quartz [1]. Kaolin is an aluminosilicate mineral that includes [Si2O5]2− and [Al2(OH)4]2+ layers which provide plasticity and green strength during shaping process. Feldspar is used as flux in porcelain industries to reduce firing temperature and make glass phase and quartz is used as filler which is effective on the reduction of pyroplastic deformation through firing process [2]. During firing process, free quartz and mullite are formed and distributed in glassy matrix and fired body reach to maximum density between 1250 and 1300 °C [1,2,3,4]. Most properties of hard porcelain bodies depend on glassy matrix homogeneity [3], pores (size, shape and distribution) [2], and type of crystals. Difference of density, refractive index and thermal expansion coefficient of crystals and glass phase were caused to change the whiteware’s strength and translucency of fired body. Because of high crystalline to glassy phase ratio, this whiteware body has higher mechanical strength (~ 100 MPa) and in spite of this, compare to hard porcelain, bone china has better whiteness and translucency [5,6,7]. Bone china composition includes tri-calcium phosphate, anorthite, quartz and glassy phases [8]. Similar reflective index with glassy phase causes more translucencies. Furthermore, thermal expansion coefficient (TEC) of anorthite (~ 4.3 × 10−6 K−1) closed to glassy phase (~ 3–4.5 × 10−6 K−1), which makes it resistance to thermal shock but TEC of tri-calcium phosphate (~ 12 × 10−6 K−1) [9], increases TEC of bone china (~ 8.5 × 10−6 K−1) [8] and makes it suitable for low-temperature glaze [9]. These two appropriate properties of anorthite encouraged the producer to improve physical properties of hard porcelain bodies by creating anorthite phase in body [10]. Capoghlu investigated to produce translucent whiteware with anorthite, mullite and a small amount of glassy phase [11]. Furthermore, he increased the whiteness of porcelain body using new composition includes wollastonite, alumina, ball clay and quartz. Taskiran and colleagues reported that high ratio of crystalline phase which included 50% anorthite to glassy phase (70/30) is caused by high flexural strength [12]. Anorthite is one of the main members of plagioclase feldspar family [13]. According to limited natural anorthite sources [14], scientists strive to synthesize it by difference methods such as sol–gel [15], sintering the mixture of raw materials [16] and mechanochemical method [17]. Due to the high melting temperature of anorthite, several studies were carried out to decrease the crystallization of anorthite using different fluxing agents such as boron, Na2CO3 and some other additives [17,18,19]. Supplementary scientists investigate about various calcium sources to reach anorthite phase in porcelain bodies. They used Ca(OH)2, calcite, marble powder and gypsum mold waste [20], dolomite, wollastonite and calcite [21]. In this research, white cement was used as a new CaO source for anorthite base porcelains and its influence on sintering behavior, M.O.R, TEC, bulk density, and water adsorption were investigated.

Materials and methods

The raw materials used in this investigation were kaolin (Super Standard Porcelain, IMERYS Ceramics, UK), quartz (Morvarid–Iran), potassium feldspar (Quantum AP200F, Cibelco–India) and white cement. The chemical compositions of raw materials were determined by X-ray fluorescence (XRF) analyzer (PANalytical, Axios model) (Table 1). A standard composition for hard porcelain was prepared by addition of 50 wt% kaolin, 25 wt% potassium feldspar and 25 wt% quartz.

Table 1 The chemical composition of raw materials
Full size table

Six additional body mixtures were formulated partially by replacing 0, 3, 4.5, 6, 7.5 and 9 wt% of potassium feldspar by white cement. In Table 2, the composition of specimens has been listed.

Table 2 Composition of specimens
Full size table

The slips of S0, S3, S4.5, S6, S7.5 and S9 were prepared by milling 2000 g of the mixtures with 3000 g water for 8 h in a jar mill. The resulting watery slurries were passed through a 100-µm sieve, deironing by a permanent magnet, and dewatered on a plaster plate. Finally, the resulted mud was extruded into 2 cm × 1 cm × 10 cm bars. The bars were first air-dried at room temperature for 24 h and then oven-dried at 100 ± 5 °C for 2 h. After drying, the samples were fired at 1020, 1100, 1195, 1250, and 1340 °C in electric kiln (Exciton, EX-1700 model). The linear shrinkage, water adsorption (ASTM C373-88) and bulk density of fired body were calculated by densitometer (Sartorious, LA230S model; Germany). Phase characterization was accomplished by a DX-27 mini X-ray diffractometer (XRD) using Cu Kα radiation and operating at 40 kV and 30 mA. For this purpose, bulk sample were scanned in the angle range of 2θ: 4–70°. Modulus of rupture of the samples was measured by NETZSCH-Geratebau bending strength tester 40. Finally, Cambridge S360 analytical scanning electron microscopes run at 19.9 kV to study the microstructure of these two samples. Gold coating of the samples was carried out using a Sputter-coating instrument.

Results and discussion

In all firing temperature, linear shrinkage of samples decreased by cement content (see Fig. 1). Besides, increasing the firing temperature increases the amount of liquid phase which fills the pores and causes to increase the linear shrinkage.

Fig. 1
figure 1

Linear shrinkage of fired samples at different temperatures (°C)

Full size image

Water adsorption and bulk density as function of cement content at various temperatures were showed in Figs. 2 and 3, respectively. According to results of linear shrinkage, as one expect water adsorption increases by addition of white cement. Whereas the reduction of firing temperature was cause to increase water adsorption.

Fig. 2
figure 2

Water adsorption of fired samples at different temperatures (°C)

Full size image
Fig. 3
figure 3

Bulk density of fired samples at different temperatures (°C)

Full size image

The water adsorption values of fired specimens at 1250 and 1340 °C for all samples were zero which was due to complete sintering. At the other temperatures, water adsorption values decreased by white cement content. According to high viscosity of glass phase due to presence of white cement, water adsorption was increased at 1195 °C by increasing white cement content. Water adsorption has direct relevance of porosity and the porosity has inverse relationship with the bulk density; therefore, it is obvious that bulk density increases by increasing the firing temperature for S0 body. In other bodies, the bulk density decreases after 1250 °C, and also it decreases by adding white cement in the composition, this suggests decreased density of the body as a result of development of glass phase or development of a bubble in the fired body [22]. Furthermore, the specimens that fired at 1250 °C reached the maximum value of bulk density that implies full sintering at this temperature. Bulk density of fired specimens at 1020, 1100 and 1195 °C reduces corresponding increasing water adsorption. Bulk density of samples with various amounts of cement which were fired at 1250 °C is similar to bulk density of fired sample at 1340 °C without any cement. According to these results, CaO play a main role during firing process. In fact, CaO increase the surface tension of glassy phase which could be due to reduction of density. Figure 4 shows modulus of rupture of specimens (M.O.R) as function of the amount of cement at different firing temperatures. Behavior of specimens which were fired at 1020 °C, 1100 and 1195 °C fits perfectly with the bulk density values. At these temperatures, as one would expect, modulus of rupture of samples decreases with decreasing firing temperature and also decreases with adding white cement. It is to be noted that severe loss of modulus of rupture of the samples with more than 6% white cement can be seen. It might be because of sudden increasing of glassy phase. White cement provides more CaO beside SiO2 in glassy phase.

Fig. 4
figure 4

Modulus of rupture of fired samples at different temperatures (°C)

Full size image

Modulus of rupture values for fired samples at 1250 °C with 3, 4.5 and 6 percent white cement is more than the fired samples at 1340 °C, whereas before adding white cement fired sample at 1340 °C has higher modulus of rupture. Also adding more than 6% white cement leads to decrease modulus of rupture of both. Reduction of modulus of rupture of fired sample at 1340 °C in the presence of white cement might be because of over firing, bloating and deformation of specimens. As previously mentioned, more than 6% white cement might be leads to increase glassy phase. Figures 5 and 6 show XRD patterns of S0, S4.5 and S9 that fired at 1340 and 1250 °C.

Fig. 5
figure 5

XRD patterns of fired S0, S4.5 and S9 at 1340 °C

Full size image
Fig. 6
figure 6

XRD patterns of fired S0, S4.5 and S9 at 1250 °C

Full size image

No anorthite phase exists in the structure of fired samples at 1340 °C except the sample which contained 9 wt% white cement. However, XRD patterns of fired samples at 1250 °C show anorthite peaks. Previous studies [16] reported that maximum amount of anorthite phase was formed between 1170 and 1245 °C. It seems that anorthite phase was melted and dissolved in glass phase by increasing temperature from 1250 to 1340 °C. These results were proved in SEM picture (Fig. 7). The thermal expansion coefficient of porcelain body is a key factor when considering the thermal match between body and glaze. According to thermal expansion coefficients, the average TEC of the samples between room temperature and 600 °C is calculated and listed in Table 3. Anorthite phase formation with low thermal expansion coefficient would be cause to decrease TEC of body. According to Table 3, TEC decrease by addition of white cement. This low TEC confirms that anorthite base porcelain would be resistant to thermal shock.

Fig. 7
figure 7

SEM images of fired bodies (a S9 fired at 1340 °C; b S0 fired at 1340 °C; c S9 fired at 1250 °C; d S0 fired at 1250 °C)

Full size image
Table 3 Average TEC values of samples measured in the range from 25 to 600 °C
Full size table

White cement increased CaO, which decreased the linear shrinkage rate. It is because of the amount of glassy phase. In these circumstances, the water adsorption was directly related to temperature rising and also it was inversely correlated with increasing the amount of CaO. The bulk density had the same behavior. CaO caused increasing surface tension of glassy phase so the bulk density decreased. It should be noted that CaO addition more than 6% especially at higher temperatures was leading to sudden increase the amount of glassy phase.

Conclusions

By replacing potassium feldspar with white cement, anorthite appeared in fired body at 1250 °C. This crystal can be seen in fired body at 1340 °C replacing 9% potassium feldspar with white cement. Replacing white cement and constitution of anorthite caused decreasing linear shrinkage. This replacing decreases bulk density, linear shrinkage and modulus of rapture. Although water adsorption and consequently firing temperature increase by replacing white cement, 1250 °C was a good temperature for firing replaced bodies.