ISSN 1070-4272, Russian Journal of Applied Chemistry, 2013, Vol. 86, No. 7, pp. 965−970. © Pleiades Publishing, Ltd., 2013. Original Russian Text © B.I. Gurevich, A.M. Kalinkin, E.V. Kalinkina, V.V. Tyukavkina, 2013, published in Zhurnal Prikladnoi Khimii, 2013, Vol. 86, No. 7, pp. 1030−1035.
INORGANIC SYNTHESIS AND INDUSTRIAL INORGANIC CHEMISTRY
Effect of the Mechanical Activation of Nepheline Concentrate on Its Binding Properties in Mixed Cements B. I. Gurevich, A. M. Kalinkin, E. V. Kalinkina, and V. V. Tyukavkina Tananaev Institute of Chemistry and Technology of Rare Elements and Mineral Resources, Kola Scientific Center, Russian Academy of Sciences, Apatity, Murmansk oblast, Russia e-mail:
[email protected] Received July 12, 2013
Abstract—Binding properties of a Portland cement–nepheline–water formulation were studied in relation to its nepheline content by using a preliminary mechanical activation. A thermal analysis was used to estimate the hydration rate of cement phases in the system under study. The accelerating role of nepheline in hardening of mechanically activated Portland cement–nepheline formulations was revealed and found to be more pronounced in early stages. The gain in the strength of the cement stone was analyzed in relation to the formulation composition and hardening duration. DOI: 10.1134/S1070427213070045
By its nature, nepheline belongs to the class of skeleton silicates and has a chemical formula close to (Na,K)2O·Al2O3·2SiO2 [2]. The crystal structure of nepheline is similar to that of β-tridymite in which half of Si4+ ions is substituted by Al3+. Na+ cations neutralizing the residual negative charge are situated in voids of the skeleton formed by (Si,Al)O4 tetrahedra. Nepheline crystals have a prismatic, short-columnar, or thick-tabular appearance and frequently form twin aggregates [3]. Highly dispersed nepheline additive are involved, together with cement, in the formation of the cement stone structure. The effect of these additives on the properties of the cement stone (the so-called “filler effect” [4]) depends on the dispersity of nepheline particles, degree of their amorphization, degree of filling of the cement binder matrix with nepheline, and on its chemical activity. These factors affect to a lesser or greater extent the hydration rate of cement phases and are important for prognosticating characteristics of the cement stone. According to our data, crystalline nepheline hardly has any binder properties and inertly behaves in hardening processes. One of ways to improve the activity
To obtain composite binders that are based on Portland cement and used as highly dispersed mineral additives to cement, it is advisable to use technogenic raw materials. With an adequate choice of microfillers, it is possible no only not to impair the physicomechanical properties, but also to obtain materials with better strength-related and other characteristics. As components “diluting” portland cement without any significant deterioration of the binder properties can serve tails from ore-dressing plants, overburden rock, wastes metallurgy and fueland-power sector (ashes, slugs), and other industrial wastes. Huge amounts of nepheline-containing wastes, which are tails from dressing of apatite-nepheline ores, have accumulated on the territory of Russia. These tails, 55–65% composed of nepheline, are regarded as a component of binder formulations, which make it possible to obtain low-and medium-grade cements contained in hardening mixtures for erection of filling masses. The content of Portland cement in formulations of this kind is about 50% and in some cases it can be reduced to 10–20%. In joint or separate grinding of apatite flotation tails with Portland cement in a ball mill to a specific surface area of 2500–400 cm2 g–1, a binder with a strength of up to 30 MPa in a 28-day age has been obtained [1]. 965
966
GUREVICH et al.
of components in hardening of binder composites and raise their chemical affinity to each other is by using the mechanical activation (MA). The MA can change the dispersity and the nature of the deformation state of the surface layers of low-active particles and thereby affect the hydration hardening of their dispersions [5]. In the present study, we used a preliminary joint MA of Portland cement and nepheline in order to obtain the largest specific surface area (Ssp) of the mixture, which would provide the best manifestation of hydraulic properties. There exist various methods for studying the hydration of cement in early stages (isothermal calorimetry, X-ray phase analysis with the Ritweld’s method, NMR spectroscopy, and electron microscopy [4]), which are employed to interpret experimental data. The goal of our study was to demonstrate how thermal analysis data for cement with nepheline additive can reveal factors affecting the hydration rate of cement phases and evaluate the extent of their influence.
of 40 g. As milling bodies served steel balls 8 mm in diameter, with a 6 : 1 ratio between the mass of the balls and the charge mass. The MA duration was chosen on the basis of an experimentally determined dependence of the Ssp of the clinker on its MA duration (not presented) to be 150 s. With increasing MA duration, the specific surface of the activated clinker passes through a maximum and starts to decrease because particle aggregates are formed. The specific surface area was measured by the air permeability method. To evaluate the binding properties of mechanically activated mixtures of PC and nepheline, we measured the compression strength of a hardened cement stone. For this purpose, 1.41 × 1.41 × 1.41 cm cubes were fabricated from a paste with plastic consistency and hardened under humid conditions at a temperature of 20–22°C. A thermal analysis of the hardened samples was made with a Paulik–Paulik–Erdey derivatograph (Hungary). The weighed portion of a substance was 0.5 g, and the heating rate, 10 deg min–1. The TG sensitivity was 200 mg, and the sensitivities of DTA and DTG, 1/5 and 1/10, respectively. An X-ray phase analysis was performed with a DRON-2 diffractometer (CuK radiation). Figure 1 shows thermograms of samples of the starting NG and that “diluted” with the nepheline concentrate in amounts of up to 80 wt %. The DTA curves show endothermic effects in the range 100–180°C, associated with the removal of adsorbed water and water released in dehydration of the gel of hydrosilicates and aluminosilicates. As the amount of NC in the starting mixtures increases, the peaks of the endothermic effects in the hardened cement paste shift to lower temperatures. In the temperature range 450– 490°C, water is removed mostly due to the dehydration of Ca(OH)2, which is formed in dehydration of the tricalcium silicate, and unstable hydrate compounds of calcium. The DTA curves of samples hardened during a
EXPERIMENTAL As starting components of binder formulations served portland-cement clinker from Savinskii plant with addition of 5% natural gypsum (PC) and the nepheline concentrate (NC) manufactured by Apatit OAO from flotation tails of apatite-nepheline ores from Khibiny massif. The chemical compositions of the components are listed in Table 1. The mineral composition of NC includes (wt %): nepheline 75–80, feldspar 8–16, minerals secondary to nepheline 1.5–10, aegirite 1.5–5, titanomagnetite 0.4– 0.6, apatite 0.2–0.8, sphene 0.5–1. To study hardening processes, we used PC–NC formulation with NC content varied from 1 to 80 wt %. The MA of the mixtures was performed in an AGO2 centrifugal-planetary mill [6] at a centrifugal factor
Table 1. Chemical composition of Portland cement clinker, nepheline concentrate, and gypsum, wt % Calcination The rest loss
Component
SiO2
Al2O3
Fe2O3
CaO
MgO
Na2O
K2O
TiO2
P2O5
SO3
Clinker
22.65
4.96
3.27
63.35
2.12
1.0
0.67
–
0.30
1.14
1.65
–
NC
43.37
29.48
2.90
0.84
0.27
12.7
9.01
0.27
0.03
–
1.13
–
–
–
–
25.25
–
–
–
–
–
36.41
15.69
22.65
Gypsum
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 86 No. 7 2013
EFFECT OF THE MECHANICAL ACTIVATION OF NEPHELINE CONCENTRATE (a)
967
(b)
T, K
T, K
Fig. 1. DTA curves for samples based on mechanically activated PC+NC mixtures hardened for: (a) 7 and (b) 360 days. Mixture compositions (%): (1) 100 PC, (2) 95 PC + 5 NC, (3) 70 PC + 30 NC, (4) 50 PC + 50 NC, and (5) 20 PC + 80 NC.
year (Fig. 1b) have the same form as that for 7-day-old samples (Fig. 1a), although the endothermic peaks are shifted to higher temperatures. The results of processing of the thermal analysis data are listed in Table 2. For each sample, the loss of mass upon heating in the range 450–490°C was calculated {Δm[Ca(OH)2]}. This quantity is conditionally attributed to the dehydration of Ca(OH)2 and recalculated to the content of a free calcium oxide (CaOfree by TG) in a fully calcined substance. According to XPA data, nepheline is partly amorphized after a MA for 150 s. The X-ray diffraction pattern of the activated sample retains nearly all the main reflections of nepheline, although their intensity decreases (Fig. 2). As already noted, the crystalline nepheline is hydrated under the standard conditions very slowly. If we consider that nepheline is an inert additive in our experiments, then the loss of mass in the temperature range 450–490°C must be proportional to the content of PC in the starting mixtures. This loss of mass, calculated with consideration for the composition of the mixtures and recalculated to the content of free calcium oxide in a fully calcined substance (CaOfree by calc.), are listed in Table 2. It can be seen that the calculated and found from the TG data contents of free calcium oxide do not coincide. If we assume that nepheline in hardening formulations has no effect on the hydration process of Portland cement, then Δm± must be close to zero. If Δm± is negative, the course of the hardening process presumably deviates for some reason toward prevalence
of the NG hydration over the hydration of PC in contact with nepheline grains. If Δm± is positive, this points to the existence of factors accelerating the hardening processes of the PC+NC mixture (i.e., the processes occurring near the surface of nepheline grains). The data in Table 2 show that Δm± is positive for all compositions of mixtures hardened for a year, which points to the acceleration of the PC hydration in the presence of the highly dispersed nepheline additive. The possible factors accelerating the hydration are defectiveness of the structure, mineralogical composition of a formulation, dispersity, etc. The increased defectiveness of PC in a MA in a mixture with NC is unlikely. Probably, the key role in that the strength increases is played by the heterogeneous nucleation of hydrates, which occurs on
20
40
60
2θ, deg
Fig. 2. X-ray diffraction patterns of (1) starting NC and (2) NC upon a MA for 150 s in an AGO-2 mill.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 86 No. 7 2013
968
GUREVICH et al.
Table 2. Results of processing of the thermal analysis data for samples with various hardening times, based on mechanically activated PC–NC mixtures Formulation no.
CaOfree CaOfree NC additive, Δmtot, ∆m ∆mtot, ∆m ∆m± wt % Ca(OH)2 TG data calculation wt% Ca(OH)2 TG data calculation wt % 7 days
98
–
∆m±
28 days
21.7
2.8
11.1
11.1
–
23.3
2.6
8.3
8.3
–
99
1
20.3
2.9
11.1
10.9
+0.1
22.1
2.3
9.3
8.2
+1.1
104
3
19.9
2.6
11.1
10.8
+0.3
20.6
2.6
10.4
8.1
+2.3
100
5
20.8
2.9
11.5
10.5
+1.0
21.6
2.8
11.2
7.9
+3.4
101
10
18.9
2.9
11.0
10.0
+1.0
22.0
2.9
11.5
7.4
+4.1
102
20
18.0
2.6
10.0
8.9
+1.1
20.1
3.1
12.0
6.6
+5.4
103
30
18.7
2.2
8.4
7.8
+0.6
20.2
2.0
7.7
5.8
+1.9
105
50
17.6
1.7
6.3
5.16
+0.7
17.6
1.5
5.5
4.1
+1.4
106
70
15.6
0.9
3.4
3.3
+0.1
16.6
1.2
4.3
2.5
+1.8
107
80
11.2
1.2
4.2
2.2
+2.0
13.2
Trace amounts
1.7
–
11.8
–
270 days 98
–
360 days
25.7
3.1
13.2
–
–
23.3
2.8
11.8
99
1
26.1
3.2
13.5
13.0
+0.5
104
3
24.5
3.4
13.8
12.7
+1.1
24.5
3.8
15.5
11.4
+4.1
100
5
25.4
3.8
15.7
12.5
+3.2
23.9
3.5
14.2
11.2
+3.0
101
10
24.4
3.4
13.8
11.8
+2.0
24.5
3.5
14.3
10.6
+3.7
102
20
24.7
3.3
13.5
10.5
+3.0
23.4
3.1
12.7
9.2
+3.5
103
30
24.8
2.6
10.8
9.2
+1.6
23.2
2.4
10.0
8.3
+1.7
105
50
22.3
1.7
6.9
6.6
+0.3
20.9
1.9
7.4
5.9
+1.5
106
70
16.9
1.6
5.9
4.0
+1.9
19.3
3.3
–
107
80
16.1
Trace amounts
2.6
–
12.9
Trace amounts Trace amounts
2.3
–
the surface of nepheline grains and is due to the specific crystal-chemical features of its structure. Possibly, the mechanical activation of a mixture leads to an increase in the mutual chemical affinity of the components, which leads to their closer contact and, as a result, to aggregation of PC hydration products with nepheline. A similar phenomenon of the increasing hydration rate and strength has been noted in hardening of formulations based on Portland cement with addition of badeleyite [7] and anatase [4].
Not determined
During the first 28 days, addition of NC in an amount of 1 to 30% improves the compression strength up to 130% as compared with Portland cement fabricated with the same clinker, but without nepheline additive. Further addition of NC (50 to 80%) impairs the strength; if, however, nepheline is regarded as an inert material, dilution of the mixture with nepheline would have yielded a lower sample strength than that actually determined (Table 3). For all the times specified, nepheline accelerates hardening processes. At the 7-
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 86 No. 7 2013
EFFECT OF THE MECHANICAL ACTIVATION OF NEPHELINE CONCENTRATE
969
Table 3. Effect of NC additives to PC on the mixture strength Formulation no. No number
Composition, wt % NC
PC
Ssp, m2 kg–1
W/C
Rcompr, MPA, at indicated hardening time, days
Rcompr, %
7
28
270
360
7
28
270
360
100
–
1062
0.31
1.1
1.1
0.9
1.2
–
–
–
–
98
–
100
710
0.31
48.3
58.3
69.5
70.6
100
100
100
100
99
1
99
742
0.31
63.9
68.0
72.2
–
139
117
104
–
104
3
97
825
0.31
57.9
72.4
73.9
74.8
120
124
105
106
100
5
95
844
0.31
55.3
71.3
72.5
73.9
114
122
104
105
101
10
90
834
0.31
52.1
73.0
75.7
76.3
108
125
109
108
102
20
80
848
0.32
54.3
76.6
80.4
80.9
112
131
116
115
103
30
70
919
0.33
59.5
70.8
77.2
77.8
123
121
111
110
105
50
50
832
0.33
39.6
44.4
53.1
54.0
82
76
76
76
106
70
30
761
0.34
24.6
26.0
31.0
31.8
51
44
45
45
107
80
20
780
0.33
21.0
24.7
27.9
28.7
43
42
40
41
and 28-day age, the gain in strength is larger, but it is also observed at ages of 270 and 360 days. The largest gain in the strength of PC–NC formulations, compared with pure PC, is observed for compositions containing 20–30% NC and this behavior is observed for all the four hardening durations. The fact that the hardening of PC-clinker is accelerated by addition of nepheline is correlated with the positive values of Δm± for all the mixture compositions (Table 2) and also with the XPA data (not presented). According to the latter, the intensities of the main peak of Ca(OH)2 for hardened samples, which can be used to indirectly judge about the degree of hydration, grow faster than the content of PC-clinker in the samples, rather than in proportion to this content. CONCLUSIONS (1) The accelerating role of nepheline in hydration processes of mechanically activated formulations
composed of Portland cement and nepheline concentrate was revealed. This effect is more pronounced in the initial stage. (2) It was found that addition of the nepheline concentrate to portland cement in amounts of 1 to 30 wt % improve, in early hardening stages, the compression strength up to 130%, compared with Portland cement fabricated with the same clinker, but without nepheline additives. The hardening processes are probably accelerated as a result of the heterogeneous nucleation of hydrates, which occurs on the surface of nepheline grains and is due to the crystal-chemical specific features of its structure. REFERENCES 1. Bessonov, I.I., Leont’ev, A.A., Konokhov, V.P., et al., Zakladochnye materialy iz otkhodov proizvodstva (Fill Materials Produced from Industrial Wastes), Apatity: Kol’skii Filial Akad. Nauk SSSR, 1988.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 86 No. 7 2013
970
GUREVICH et al.
2. Ravich, B.M., Okladnikov, V.P., Lygach, V.N., and Menkovskii, M.A., Kompleksnoe ispol’zovanie syr’ya i otkhodov (Integrated Use of Raw Materials and Wastes), Moscow: Khimiya, 1988. 3. Betekhtin, A.G., Mineralogiya (Mineralogy), Moscow: Gos. Izd. Geol. Lit., 1950. 4. Le Su, G. and Ben Khakha, M., Tsement Ego Primenenie, 2012, no. 4, pp. 46–51. 5. Vagner, G.R., Fiziko-khimiya protsessov aktivatsii tsementnykh dispersii (Physical Chemistry of Cement
Dispersion Activation Processes), Kiev: Naukova Dumka, 1980. 6. Avvakumov, E.G. and Gusev, A.A., Mekhanicheskie metody aktivatsii v pererabotke prirodnogo i tekhnogennogo syr’ya (Mechanical Activation Methods in Processing of Natural and Technogenic Raw Materials), Novosibirsk: Geo, 2009. 7. Gurevich, B.I., Kalinkin, A.M., Tyukavkina, V.V., and Kalinkina, E.V., Zh. Prikl. Khim., 2011, vol. 84, no. 5, pp. 736–742.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 86 No. 7 2013
