Experimental Study on the Fine Iron Ore Tailing Containing Gypsum as Backfill Material

)e strength of the filling body is largely affected by the properties of the binder, mineral composition, fineness, and slurry concentration of tailing. In this paper, the rheological test was conducted to determine the slurry concentration of iron ore tailing containing gypsum. )en, the samples made from slurry and three binders, Portland cement, filling plant binder, and Huazhong binder, were tested, respectively. )e effects of curing time, binder-tailing ratio by mass (b/t), and slurry concentration on compression strength were investigated. )e sample made from Huazhong binder and iron ore tailing presented the largest compression strength.


Introduction
Tailings are produced as a lot of by-product in the mining industry. Large amount of open area is needed for the waste, and it has an important effect on the environment. However, voids occur after mined. Landslide may happen in the goaf zone. It is necessary to develop paste backfill technology to reuse waste and increase stability of land. About 60% of the waste will be reduced by the backfill. e tailing is used as the raw material of the filling body. e physical and chemical properties of tailing, as a kind of artificial sandy soil, mainly depend on mineral composition and grade. Generally, due to quartz of high content, tailing particle of hard ore rock after mineral processing is hard. Instead, due to the ore of high content, the tailings of soft and fractured rock or oxidized ore usually are tiny and viscous. While sometimes ore is mixed with much soil, the average particle size also decreases. In the mine-tailing-filled engineering, the physical and chemical properties of different tailings are complicated and changeable. When using different binder materials, the mechanical property of filling body is diverse. Huang et al. [1] had found that the bind type had an observable influence on the mode I fracture toughness of backfill structure. e backfill material is always pumped in a pipe into the site, so an appropriate rheological property is needed. Zhang et al. [2] had found that the mineral admixture type and dosage affect the rheological and mechanical properties of cemented foam backfill. e strength of the filling body not only depends on the properties of tail but also depends on the type, content, and fineness of the binder [3,4]. Jabulani et al. [5] investigated the effect of fly ash on the behavior of paste backfill produced by brine. e fly ash seems play a significant role than brine. Fall and Benzaazoua [6] studied the influence of Portland cement, fly ash cement, and slag cement on high sulfur tailings' paste filling strength. It turned out that Portland cement had a better effect than fly ash cement and slag cement on the paste filling body. Furthermore, the chemical composition of the mixed water played a certain role in the strength properties of the cemented backfill. Other researchers [7][8][9] studied the strength characteristics of sulfur tailings' backfill under different conditions. Cihangir et al. [10] and Yilmaz et al. [11] investigated the microstructure of the paste filling body and discussed the influence of the microstructure on mechanical properties and strength development. Some researchers [12][13][14][15][16][17][18] analyzed the coupling effect of temperature, water, and force on the strength of the filling body. ey also evaluated various mechanical properties of backfill under other similar complex environments. Fall et al. [19] and Abdul-Hussain et al. [20,21] analyzed the water pressure characteristics of cemented backfill. Moreover, they investigated the permeability of cemented backfill and cemented paste backfill under saturated and unsaturated conditions.
In this paper, the rheological behaviors of slurry were investigated to determine the concentration of tailing. Because there is some gypsum in iron ore tailing of chenchao, the usually used binder materials are unbefitting. en, three binders, Portland cement 325# (PC), filling plant binder (FP), and Huazhong binder (HZ), were, respectively, used as the binder to make backfill samples. e compressive strength of backfill samples with different binder-tailing ratio by mass (b/t), slurry concentration, and curing time was compared.

Preparation of Raw Materials
e iron ore tailing of chenchao and three binders, PC, FP, and HZ, were used. e physical properties of three binders are shown in Table 1.
In order to obtain the representative and uniform tailings, a rectangular material-receiving tank was dug at a tailing dam. Antiseepage color cloth was laid in the tank, and then, the tailing pumped from the mineral processing plant was drained into it by using a 4-inch white plastic tube, as shown in Figure 1. e tail paste will be settled naturally and water above the tailing mortar will be discharged. e physical properties of tailing were measured and shown in Table 1.
e particle size distribution of tailings was measured by a MASTERSIZER 2000 Laser diffraction particle size analyzer and shown in Figure 2. e chemical composition of the tailing is shown in Table 2. It contains about 22% CaSO 4 .

Rheological Test of Slurry.
e filling process is downwards, so the viscosity of slurry should not be too small to prevent separation. e filling distance is long, and the viscosity should be not too large to guarantee flow under self-gravity. e viscosity is largely dependent on the slurry concentration of the tailing. e slump test is usually conducted to describe the flow performance of slurry [22,23]. erefore, slurries with 7 different concentrations were prepared, and their mix proportions are shown in Table 3. A slump test were conducted, as shown in Figure 3.
e slump values are shown in Figure 4. With the advance of concentration, the slump increases.
Meanwhile, a rheometer was used to measure the yield strength and plastic viscosity of slurry, and the results are shown in Table 4. With the decrease of slurry concentration, the yield strength and plastic viscosity decrease. When the concentration is 74∼72%, the flow characteristics of slurry is bad and the slump is 20.0∼23.5 cm, making it difficult to flow under self-gravity. When the concentration is 70%, the slurry has a certain fluidity and the viscosity is large. When the concentration decreases to 68∼64%, the slump is 25.5∼27 cm. At this time, the fluidity of slurry is obviously improved. When the concentration is below 62%, the waterretaining property of slurry declines, and the bleeding phenomenon is obvious. Combining with rheological parameters and slump value, we determined the slurry concentration is 64%-68%.

3.2.
Production of the Filling Sample. 36 series filling samples with 3 different b/t, 4 types of tailing concentrations, and 3 types of binders were prepared. e mix proportion is shown in Table 5.
e tailing and binder were firstly poured into the mixer and mixed for 3 min. en, water was added and mixed for another 3 min. After the filling paste was prepared, it was poured into the mold. After one day, we removed the molds and put them into the curing room. e cubic filling sample with an edge length of 70.7 mm was fabricated.

Experimental Results and Discussion
A uniaxial compression test was conducted to measure the uniaxial compression strength of every series after curing time of 3, 7, 28, and 60 days. e load was applied at a constant loading rate of 0.5 MPa/s until the sample was broken. For every series, the average UCS of 3 samples was calculated.

Effect of Curing Time on UCS.
e effect of curing time on UCS of the sample with 66% tailing concentration is shown in Figure 5. Although the UCS of all samples of three binders increases with curing time, the increase rate is different.
When the curing time is 3 days and b/t is 1 : 4 ( Figure  .5(a)), the UCS of HZ is slightly larger than others, but for the other 3 series, their UCS is almost the same. When the curing time is 7 days and the dosage of binder is surplus (Figures 5(a) and 5(b)), the UCS of HZ is the largest and the UCS of PC is the lowest, but when the dosage of the binder decreases (Figures 5(c) and 5(d)), the UCS of FP is larger than others. However, after 60 days curing time, regardless of b/t, the UCS of HZ used sample is largely higher than others.

Effect of b/t on UCS.
e effect of b/t on UCS of the sample with 66% tailing concentration is shown in Figure 6. Regardless of curing time, the UCS of HZ increases with the increase of b/t. However, for PC and FP, the b/t has no obvious effect. e UCS of PC and FP at all curing ages is less than 2 MPa. ere are two reasons: On one hand, the size of iron tailing of chenchao is small, so its specific surface area (SSA) is large. e SSA of HZ is larger than the others, so the increase of contact area leads to the increase of reaction rate. On the other hand, the iron tailing of chenchao has only 27.8% SiO 2 , but the contents of CaSO 4 and MgO exceeds 24% and 7%. e flocculation of Ca+ has a negative effect on the strength of cementitious material.

Conclusions
36 series samples were conducted. e effects of curing time, b/t, and slurry concentration were investigated. e following conclusions were obtained: (1) Regardless of curing time, b/t, and slurry concentration, the UCS of HZ is larger than the others (2) e UCS of all samples increases with the curing time (3) e UCS of HZ increases with the increase of b/t and slurry concentration, but the UCS of PC and FP has no obvious variation HZ is a new kind of mine tailing filling material, which is made of blast furnace slag powder and some additives. Compared with PC and HZ, it has the advantages of early strength increasing quickly, late strength stability and reliability, and small dosage. At the same level of filling body strength, HZ dosage is just about 30∼60% of PC. Hence, for iron tailing with gypsum of high content, the HZ is a more suitable binder.
Data Availability e [DATA TYPE] data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.