When performing these studies, both raw materials were studied, and after firing, the physical-mechanical properties of the developed ceramics and the processes of formation of their structures, explaining the properties obtained.
3.1 Raw materials characterization
Both raw materials were characterized in terms of their granulometric, chemical and mineral compositions and morphological microstructure, because these properties are the most effective on the structure and properties of the developed ceramics.
3.1.1 Chemical composition of the raw materials
Both industrial waste materials used as raw materials in the study were obtained from local factories in Brazil. The red mud (RM) was supplied by a refinery in Sao Paulo’s state, and blast
furnace slag (BFS) from a metallurgical plant in Parana’s state. The sample of natural clay mixed with 10% sand was obtained on the local ceramics production plant and was used with this work only as a generally accepted industrial standard.
Table 1
– Chemical composition of the raw materials.
Oxides
|
RM
|
BFS
|
CSM
|
Oxides
|
RM
|
BFS
|
CSM
|
Fe2O3
|
29.9
|
62.1
|
6.1
|
Cr2O3
|
< 0.1
|
0.3
|
< 0.1
|
SiO2
|
15.5
|
13.8
|
53.3
|
V2O5
|
0.1
|
0.0
|
< 0.1
|
SO3
|
0.6
|
10.0
|
0.1
|
BaO
|
0.0
|
0.2
|
0.1
|
Al2O3
|
21.2
|
2.6
|
24.7
|
P2O5
|
0.0
|
0.1
|
0.0
|
CaO
|
4.2
|
2.5
|
0.4
|
SnO2
|
0.0
|
0.1
|
0.0
|
MgO
|
0.1
|
0.3
|
0.7
|
ZnO
|
< 0.1
|
0.1
|
< 0.1
|
Na2O
|
10.3
|
0.9
|
0.1
|
CuO
|
0.0
|
0.1
|
0.0
|
K2O
|
0.4
|
0.4
|
1.0
|
ZrO2
|
0.2
|
< 0.1
|
0.1
|
MnO
|
0.2
|
0.6
|
0.1
|
MoO3
|
0.0
|
< 0.1
|
0.0
|
TiO2
|
2.4
|
0.4
|
1.4
|
I.L.
|
14.4
|
1.7
|
11.5
|
P2O5
|
0.6
|
3.5
|
0.1
|
Total
|
100.0
|
100.0
|
100.0
|
The main component of this research, red mud consisted mainly of Fe2O3 (29.9 %), followed by Al2O3 (21.2%), SiO2 (15.5%), and Na2O (10.3%), determined by the XRF method (Table 1). The rather high loss on ignition value of the red mud (14.36%) might be explained by the carbonates content, by the hydroxide OH-group and the water content after undergoing the thermochemical Bayer process for bauxite ore treatment. The total Na2O can be used as a good flux material, taking into account the low melting point of most of its forms. The presence in red mud of Zn (0.72%), Ni (1.26%), Ba (0.79%), Sn (1.18%) and Cr (0.54%) was also measured via the AAS method. These values far exceeded the relevant Brazilian norms [24]. The high pH value (13.5), with the high heavy metals content, forced to classify red mud as a hazardous waste. BFS contained extremely high amounts of ferrous oxides (63.6%) in addition to SiO2 (13.8%) and SO3 (10.0%), Al2O3 (2.6%), and CaO (2.5%). Natural clay with sand addiction (NC), manufactured in local brick plants, was the only traditional raw material in the present study, used as a reference for industrial ceramics.
3.1.2 Particles size classification
The grain size classification (Table 2) showed that RM had a sum of particles from 0.3 to ≥ 1.2 mm equal to 85.46%. The majority of BFS particles (85.37%) on the contrary, presented sizes between 0 and 0.29mm. The RM showed high humidity (32.2%); DFS manifested no tendency to absorb water.
Table 2 - Grain size distribution in weight %, tapped density in g/cm3, and humidity (%) of the raw materials under study
Grain size distribution
|
Size (mm)
|
Red mud
|
Furnace slag
|
Clay-sand mix
|
0–0.074
|
0.32
|
1.35
|
89.76
|
0.075–0.149
|
1.08
|
17.63
|
5.68
|
0.15–0.29
|
13.14
|
66.39
|
3.17
|
0.3–0.59
|
46.13
|
14.63
|
1.39
|
0.6–1.19
|
39.33
|
0.00
|
0
|
≥ 1.2
|
0.00
|
0.00
|
0
|
|
Tapped density (g/cm3)
Humidity (%)
|
0.86
32.2
|
2.75
3.1
|
1.56
9.34
|
*Tapped density refers to the bulk density of the powder after a specified compacting process, involving vessel vibration. |
3.1.3 Mineral compositions of the raw materials under study
The mineral compositions of the raw materials studied by the XRD method (Fig. 1) were also somewhat different: RM held magnetite Fe3O4, hematite Fe2O3, bauxite Al2O3·nH2O, and quartz SiO2, while BFS held fayalite Fe2·2SiO4 and troilite Fe S.
All raw materials showed very weak crystal peaks and very robust X-Ray backgrounds due to the meager number of crystalline structures and the predominance of amorphous structures.
3.1.4 Morphological structures of the raw materials
Analysis of the particles' morphological structures (Fig. 3-a and b) suggested a wide variety of shapes, sizes, and forms of every material’s grain, which meant there was no uniform structure. Such particle morphology indicated a predominantly amorphous structure of the materials, established by the XRD method. All of them were disconnected and had no chemical bonds between them. According to Table 1, Figs. 1 and 2, there was a significant difference between RM and BFS regarding chemical and mineralogical compositions and the structure of the particles.
3.2 Physical properties of the developed ceramics
The developed ceramics flexural resistance strength changes, linear shrinkage, water absorption, and apparent density after sintering at different temperatures were looked into to characterize their physical properties.
3.2.1 Flexural resistance strength of the developed ceramics
Red mud was the principal component of this study. It contained the remains of the alkaline
reagent from bauxite ore decomposition (hydroxide NaOH), having a melting point (Tm) of 323°C. This occurrence led to the formation of chemical compounds, such as NaНСО3 (Tm = 270°С), Na2СО3 (Tm = 852°) and Na2SО4 (Tm = 883°С). Therefore, these chemical species might serve as flux components during the ceramics firing process. The high content of these substances was a beneficial factor to reduce the firing temperature of the developed ceramics. For standard industrial glasses, the softening starts in the temperature range of 400–600°C and ends between 700 and 750°C.
Comparing the flexural strength results (Table 3) obtained at 1,200°C (12.17 MPa) and 1,225°C (12.32 MPa), it may be stated that, at about 1,225°C, the RM was very close to the beginning of the excessive melting. An increase in BFS content drove to a decrease in the flux of RM content and an increment in the temperature of excessive melting of ceramics.
Table 3
– Flexural resistance strength of ceramics after sintering at different temperatures (°С).
№
|
Compositions, wt.%
|
Flexural strength (MPa) of ceramics sintered at T°C
|
CSM
|
RM
|
BFS
|
800
|
900
|
1,000
|
1,050
|
1,100
|
1,150
|
1,200
|
1,225
|
1
|
100
|
0
|
0
|
5.36
|
5.68
|
7.09
|
8.72
|
10.08
|
14.65
|
13.36
|
6.05
|
2
|
0
|
100
|
0
|
0
|
0
|
0.45
|
0.93
|
4.34
|
6.06
|
12.17
|
12.32
|
3
|
0
|
90
|
10
|
2.21
|
2.66
|
3.13
|
4.24
|
5.51
|
7.42
|
12.27
|
13.10
|
4
|
0
|
80
|
20
|
2.27
|
2.44
|
2.64
|
3.62
|
4.65
|
7.81
|
14.28
|
15.96
|
5
|
0
|
70
|
30
|
1.26
|
2.08
|
2.34
|
3.15
|
4.72
|
8.49
|
10.65
|
16.36
|
6
|
0
|
60
|
40
|
1.21
|
1.67
|
2.26
|
4.30
|
5.01
|
8.10
|
10.11
|
18.05
|
7
|
0
|
50
|
50
|
1.59
|
2.17
|
3.34
|
5.39
|
8.12
|
13.05
|
15.55
|
19.78
|
Therefore, ceramics 7, with the highest (50%) slag content, had the highest (19.78 MPa) strength after firing at 1,225°C, and ceramics 2 (without slag) and 3 (with 10% slag content) presented the lowest resistance (12.32 MPa and 13.10 MPa, respectively). According to the Brazilian standard [25], the flexural strength of solid bricks is classified as follows: Class A < 2.5 MPa; Class B from 2.5 to 4.0 MPa; and Class C > 4.0 MPa. That means, practically all developed ceramics corresponded to de demands of Class A (< 2.5 MPa) after firing at 800°C; to de demands of Class B (2.5–4.0 MPa) corresponded after 900°C the ceramics 3, and the ceramics 4 and 7 after 1,000°C and all other ceramics after 1,050°C; a class C corresponded ceramics 3, 6 and 7; after firing at 1,100°C all ceramics significantly exceeded the demands of class C (> 4.0 MPa).
The values of standard deviations of flexural resistance strength of the tested samples varied between 0.5 and 1.2 MPa.
3.2.2 Linear shrinkage, water absorption and apparent density of the developed ceramics
The changes in linear shrinkage values were studied (Table 4) by measuring the largest linear dimensions – samples’ length - after sintering the samples at different temperatures (Fig. 3).
Table 4
Linear shrinkage (%) of ceramics at different temperatures (°C)
№
|
Composites, wt.%
|
Linear shrinkage (%) of ceramics sintered at T°C
|
NC
|
RM
|
BFS
|
800
|
900
|
1,000
|
1,050
|
1,100
|
1,150
|
1,200
|
1,225
|
1
|
100
|
0
|
0
|
5.36
|
5.68
|
7.76
|
9.09
|
10.47
|
11.52
|
11.95
|
11.86
|
2
|
0
|
100
|
0
|
1.89
|
2.27
|
3.42
|
4.33
|
5.29
|
7.24
|
11.67
|
12.45
|
3
|
0
|
90
|
10
|
0.96
|
1.08
|
1.92
|
2.98
|
4.30
|
5.64
|
9.43
|
9.56
|
4
|
0
|
80
|
20
|
0.54
|
0.63
|
0.81
|
1.51
|
2.64
|
4.41
|
7.40
|
11.07
|
5
|
0
|
70
|
30
|
0.07
|
0.24
|
0.31
|
0.83
|
2.46
|
4.38
|
11.37
|
12.54
|
6
|
0
|
60
|
40
|
0.04
|
0.11
|
0.18
|
0.39
|
1.69
|
5.26
|
12.32
|
12.35
|
7
|
0
|
50
|
50
|
0.08
|
0.12
|
0.17
|
0.44
|
1.13
|
4.28
|
10.99
|
11.40
|
In many cases, sharp changes were strongly linked to more significant changes in flexura strength. The linear shrinkage values increased up to a temperature of maximum strength. When the melting point of a sample was exceeded, the shrinkage decreased because of the foaming effect. The total shrinkage values of all compositions at all temperatures ranged from 0.04 to 12.54%. The values of standard deviations were between 0.4 and 1.7%.
3.2.3 Water absorption of the developed ceramics
The water absorption value (Table 5) is an indirect indicator of the porosity of the samples, so the changes in water absorption were also strongly linked to changes in their flexural strength.
Table 5
The water absorption value of the developed ceramics
№
|
Composites, wt.%
|
Water absorption (%) of ceramics sintered at T°C
|
CSM
|
RM
|
BFS
|
800
|
900
|
1,000
|
1,050
|
1,100
|
1,150
|
1,200
|
1,225
|
1
|
100
|
0
|
0
|
20.40
|
20.00
|
18.49
|
15.26
|
12.76
|
10.53
|
8.78
|
8.10
|
2
|
0
|
100
|
0
|
33.78
|
31.36
|
28.68
|
25.79
|
24.62
|
19.98
|
11.79
|
9.31
|
3
|
0
|
90
|
10
|
27.65
|
27.28
|
27.86
|
22.86
|
20.51
|
17.03
|
9.67
|
6.14
|
4
|
0
|
80
|
20
|
28.65
|
26.35
|
24.90
|
21.67
|
19.49
|
16.12
|
8.41
|
4.56
|
5
|
0
|
70
|
30
|
22.66
|
24.94
|
23.19
|
22.48
|
18.92
|
14.23
|
7.29
|
2.29
|
6
|
0
|
60
|
40
|
26.94
|
25.14
|
20.29
|
19.22
|
17.17
|
11.04
|
5.64
|
1.45
|
7
|
0
|
50
|
50
|
23.37
|
19.63
|
17.09
|
16.73
|
14.68
|
9.16
|
2.86
|
2.09
|
The water absorption values decreased with the temperature rise due to a more intense process of closure of the samples’ open pores. The values decreased from and 35.22 to 1.35% with standard deviations values 0.07–2.1%.
3.2.4 Density of the developed ceramics
The nature of the changes in the samples’ apparent density was similar (Table 5) to the changes in linear shrinkage values: they grew with the increase in temperature from 1.26 to 2.54 g/cm3.
Table 6
Density of ceramic composites after firing at different T°C
N°
|
Compositions, wt.%
|
Density (g/cm3) of ceramics sintered at T°C
|
CSM
|
RM
|
BFS
|
800
|
900
|
1,000
|
1,050
|
1,100
|
1,150
|
1,200
|
1,225
|
1
|
100
|
0
|
0
|
1.56
|
1.58
|
1.61
|
1.70
|
1.77
|
1.84
|
1.87
|
1.80
|
2
|
0
|
100
|
0
|
1.28
|
1.31
|
1.35
|
1.42
|
1.44
|
1.55
|
1.89
|
1.96
|
3
|
0
|
90
|
10
|
1.38
|
1.42
|
1.47
|
1.50
|
1.56
|
1.62
|
1.82
|
1.87
|
4
|
0
|
80
|
20
|
1.47
|
1.52
|
1.55
|
1.58
|
1.65
|
1.72
|
1.99
|
2.16
|
5
|
0
|
70
|
30
|
1.53
|
1.59
|
1.64
|
1.62
|
1.72
|
1.81
|
2.26
|
2.47
|
6
|
0
|
60
|
40
|
1.71
|
1.75
|
1.77
|
1.77
|
1.82
|
1.97
|
2.24
|
2.54
|
7
|
0
|
50
|
50
|
1.85
|
1.88
|
1.90
|
1.90
|
1.96
|
2.14
|
2.40
|
2.51
|
Such behavior was highly expected, given the ongoing shrinkage of the samples and the reduction of water absorption (porosity) as the temperature increased. The values of standard deviations are within the limits of 0.09 and 0.4 g/cm3.
3.3 Physical-chemical processes of ceramics’ structure formation.
Composition 7 was considered the most promising for the study of the physical-chemical processes of the ceramics structure formation for the following reasons: 1. the high content of red mud (50%); 2. the simplicity of its composition - only two components in a 50:50 ratio; 3. the very high flexural strength (8.12, 13.05, 15.55 and 19.78 MPa), corresponding to the demands of class C [25] (> 4.0 MPa) at the temperatures of 1,100°, 1,150°, 1,200° and 1,225°C correspondingly.
3.3.1. Changes in mineral composition during ceramics sintering
Comparison of the XRD patterns of ceramics composition sintered at 1,000° and 1,225°C (Fig. 4 and Table 7) revealed the presence of magnetite Fe3O4, hematite Fe2O3, fayalite Fe2SiO4, albite NaAlSi3O8, and quartz SiO2, with a very high content of amorphous materials.
The vast majority of these peaks overlapped to one another. The only mineral peaks that were free from these simultaneities served as indicators of growth or reduction in the number of minerals formed during the temperature’s rise from 1,000° to 1,225°C. A significant decrease in the only free peaks of fayalite at 2Θ° = 34.2° and the almost complete disappearance of hematite at 2Θ°=24.2° were observed. It is well known that during sintering at high temperatures, hematite turns into magnetite. This transformation was not visible in Fig. 4 due to the absence of free magnetite peaks and their coincidence with hematite, fayalite, and quartz peaks, which suffered a decrease or even disappeared in Fig. 4-b.
Table 7
– Changes in position (d, Å) and intensities (I, %) of XRD peaks of Fig. 4, ceramic 7.
2Θ°
|
Comp. 7 − 1,000°C
|
Comp. 7 − 1,225°C
|
d, Å
|
I, %
|
Symbol
|
d, Å
|
I, %
|
Symbol
|
23.9
|
3.7
|
53.74
|
H; A
|
3.7
|
28.46
|
H; A
|
24.2
|
3.67
|
36.73
|
H
|
3.6
|
8.37
|
H
|
26.6
|
3.3
|
11.57
|
Q
|
|
|
|
33.2
|
2.69
|
100.0
|
H; A
|
2.69
|
100.00
|
H; A
|
34.2
|
2.62
|
15.53
|
F
|
2.62
|
5.17
|
F
|
35.7
|
2.51
|
68.89
|
M; H; F; A
|
2.51
|
56.76
|
M; H; F; A
|
40.9
|
2.20
|
21.34
|
F; A
|
2.20
|
17.64
|
F; A
|
42.2
|
2.14
|
3.84
|
A
|
2.16
|
8.86
|
A
|
49.6
|
1.837
|
28.13
|
H; F; A
|
1.838
|
23.33
|
H; F; A
|
54.2
|
1.692
|
33.51
|
H; F; A
|
1.691
|
28.12
|
H; F; A
|
57.6
|
1.598
|
6.61
|
H; F; A; Q
|
1.599
|
4.26
|
H; F; A
|
62.5
|
1.485
|
19.98
|
M; H; F; A
|
1.484
|
15.97
|
M; H; F; A
|
64.1
|
1.452
|
18.89
|
H; F; A; Q
|
1.452
|
13.68
|
H; F; A
|
The decrease in the intensity of almost all crystalline peaks of minerals was almost invisible to the naked eye, but it was neatly seen when comparing their intensity values in Table 7. The only mineral free of coincidence with other minerals’ peaks was quartz (2Θ°=26.6°), which disappeared after sintering at 1,225°C. Albite was the only mineral, which undoubtedly had its intensity increased (2Θ°= 42.2°). A reduction in the intensity scale from 1,500 to near 1,400 counts per second also indicated a decrease in the number of crystalline phases in the ceramic, with its transition to the amorphous state.
3.3.2 Changes in the morphological structure during ceramics sintering
The comparison of SEM micro-images of the raw materials (Fig. 2-a and b) and ceramic 7 after sintering at 1,000°C and 1,225° (Fig. 2-c and d), at the same magnification (5,000 times), showed an increase in the particles’ dimensions and their sphericity (Fig. 2-c) at 1,000°C, depicting the beginning of the initial mix’s melting process. There were many pores of different sizes between the particles. Nevertheless, the flexural resistance value of the composite 7 (Table 4) increased until 8.12 MPa. The increase in the heating temperature up to 1,225°C showed (Fig. 2-d) a complete transformation of the earlier separated particles into a nearly pore-free glass-like monolithic structure. It seems the only explanation for the morphological changes might be the melting of all particles and the chemical interaction between them. Flexural strength enhanced almost 2.5 times, up to 19.78 MPa (Table 3) with significant improvement in other mechanical characteristics. The samples’ edges of ceramics 7 (Fig. 3, the lowest) were slightly melted, and the surface turned black, apparently due to the transition of hematite to magnetite, which validated the XRD analysis’ results (Fig. 4).
3.3.3 Microchemical composition of the ceramics’ new formations.
The new formations were evaluated by the EDS method (Fig. 2d). Three different areas were analyzed, and all of them showed high heterogeneity (Table 8), such as Na content ranging from 0.00 to 7.16%, and Ca content from 0.00 to 7.00%.
Table 8
– The chemical composition of the points (Fig. 2d) by EDS method.
Spectrum
|
Na
|
Al
|
Si
|
Ca
|
Fe
|
Zr
|
Total
|
1
|
0.00
|
0.00
|
19.60
|
0.00
|
10.79
|
69.62
|
100.00
|
2
|
7.16
|
9.04
|
13.23
|
5.48
|
65.09
|
0.00
|
100.00
|
3
|
1.96
|
2.26
|
4.91
|
0.00
|
4.67
|
86.20
|
100.00
|
4
|
0.00
|
6.75
|
11.48
|
7.00
|
74.78
|
0.00
|
100.00
|
5
|
0.00
|
0.00
|
0.00
|
0.00
|
59.35
|
40.65
|
100.00
|
6
|
0.00
|
2.76
|
23.56
|
0.00
|
16.61
|
57.07
|
100.00
|
Similar results were obtained in isotopic compositions by the laser micro-mass analysis (LAMMA) method (Fig. 5).
The isotope sets (Fig. 5) and their percentage (peaks’ height) exhibited a significant heterogeneity in their compositions at the comparable points.
3.3.4 Environmental characteristics of the developed ceramics.
The study of the leaching and solubility of the hazardous red mud extract showed that the heavy metals content (Zn, Ni, Ba, Sn, and Cr) far exceeded Brazilian toxicity standards [25]. For this reason, the ceramics of composition 7 were studied for leaching and solubility by the AAS method (Table 9). The comparison of these values before and after the use of red mud for ceramics production revealed a strong chemical fixation of heavy metals in red mud, culminating in more environment-friendly materials when compared with the requirements of the Brazilian norms [25].
Table 9
– Results of leaching and solubility tests of composition 7, sintered at 1,225°C.
Elements
|
Leaching, mg/L
|
Solubility, mg/L
|
Red mud
|
Comp. 7
|
[25]
|
Red mud
|
Comp. 7
|
[25]
|
As
|
7.63
|
0.21
|
1.0
|
9.56
|
< 0.001
|
0.01
|
Ba
|
94.28
|
< 0.1
|
70.0
|
95.47
|
< 0.1
|
0.7
|
Cd
|
7.34
|
< 0.005
|
0.5
|
15.41
|
< 0.005
|
0.005
|
Pb
|
4.43
|
< 0.01
|
1.0
|
7.66
|
< 0.01
|
0.01
|
Cr total
|
18.46
|
< 0.05
|
5.0
|
22.67
|
< 0.05
|
0.05
|
Hg
|
1.47
|
< 0.001
|
0.1
|
3.81
|
< 0.0002
|
0.001
|
Se
|
2.75
|
-
|
1.0
|
3.35
|
-
|
0.01
|
Al
|
28.76
|
< 0.10
|
*
|
36.44
|
< 0.10
|
0.2
|
Cu
|
16.29
|
< 0.05
|
*
|
30.08
|
< 0.05
|
2.0
|
Fe
|
98.31
|
0.07
|
*
|
108.75
|
< 0.05
|
0.3
|
Mn
|
55.11
|
-
|
*
|
69.43
|
-
|
0.1
|
Zn
|
68.48
|
< 0.10
|
*
|
84.13
|
< 0.10
|
5.0
|
According to leaching and solubility tests, the developed material presented no propensity to react with the environment. For this reason, ceramics might be shredded and recycled as an aggregate material in concrete’s production at the end of their useful life.