Plume dispersion and the continuous process of settling, deposition and bed formation of mine tailings depend on the tailings characteristics. Phase 1 activities were done to understand the sediment characteristics, to narrow down uncertainties in the predicted behaviour of the mine tailings and are the basis of the project to select potential flocculants to be used in Phase 2 and 3. Phase 1 consists of three main activities:

1. Collation of typical tailings
2. Accurate determination of settling velocities
3. Settling and consolidation of the tailings

1) Collation of typical tailing samples for further laboratory analyses

The settlement of the tailings will be tested in settling column experiments. The design of the settling column experiments requires information about the water content, particle size distribution and density. The selection of the floculant requires an understanding of the material properties; chemical properties and settling behaviour. 

Chemical composition

The project area is wide spread, 50 x 30 square kilometers (Figure 1). From this side 50 different samples were available, obtained at different locations within the project zone. This research required a selection of three representative samples of the total batch.  As part of phase 1, the chemical composition was determined based on mineralogical data defined for the coarser grain sizes of each sample (0.8 mm – 80 mm). A Principal Component Analysis (PCA) was done to:

• summarize the chemical composition of the different samples,
• compare the composition of all different size fractions within the samples, and;
• compare the samples to find out the range of composition.

The results showed that:

• the mineralogical composition changes with size fraction, and
• the spatial variation in sediment characteristics, i.e. the mineralogy and granulometry of coarser fractions, is limited. Samples 28 and 32 differ most from the other samples, presumably because of the higher mud content.

Based on these conclusions and on the practical limitations to retrieve the sediment from the stock, samples of three locations were used for further analyses, named 28, 46, and 47.After a second round of mineralogical analyses on the fine fraction (< 0.8 mm, see Table 3.1), the conclusions above were confirmed.

Figure 1: Project area Chatham Rise. The sample locations are marked by the red dots.

Particle Size Distribution (PSD)

The particle size determines the fall velocity of the particles. In a sample there are multiple grains with different sizes. The particle size distribution (PSD) has been determined with the Malvern device, which is an optical instrument based on the principle of laser diffraction. The pitfall of this optical measuring technique is that the fraction of very fine particles (<10 μm) may be underestimated as they may be overshadowed by larger particles. Therefore, first the sample is diluted in water and measured, then the sample is left to let the larger particles settle for 1 hour. As a second step, the supernatant water with sediment is measured with the Malvern device. The PSD is then computed based on those two measurements. Figure 2 shows the PSD of the 3 selected samples. The variation between the samples is not big, and the median grain size D₅₀ ranges between 4-7 μm (Table 1). The settling velocity of the samples may however still deviated from each other to a larger extent, as the settling velocity is not only determined by the grain size, but also by the mineralogy, influencing floc formation and the amount of water encapsulated in those flocs. However, Table 3.1 shows that also those differences are relatively small.

Table 1: Selected samples with D50, percentage of mud and clay. 

Figure 2: PSD of selected samples

Water content and densities

Tailings contain water and sediment particles. The ratio is necessary to design the settling column experiments and has influence on the type and amount of flocculent. The specific water content, dry mass, bulk density and concentration of clay in the samples of the three locations were determined. Of each location subsamples were weighted, dried in an oven at 120 °C (24 hrs) and weighted again to obtain the wet and dry weight. From the water content and bulk density, the clay content can be estimated. The results are presented in Table 3.3. Sample 47 is less consolidated than the other 2 samples, as can be concluded from the bulk density and the high water content. To obtain a certain sediment mass, more tailings of sample 47 are required. Thereafter the compaction of sample 47 will be lower compared to the other two.

Table 2: water content, density and sediment concentrations of the selected samples

2) Accurate determination of sediment settling velocities

During the mining  process a plume of material will be released above the bed. In time the particles within the plume will settle. This hindered settling process is related to time. The higher the settling velocity, the faster the particles reach the bed and chances of spreading of the plume leading to environmental issues is smaller. When using flocculant, the settling time and dispersion can be influenced. To understand the effect of the flocculant, first the hindered settling of the material without flocculant is tested in 1 m tall settling columns. For each tailings type, two different concentrations are used to resolve the two unknowns; gelling concentration en the maximum settling velocity. These two material properties determine the formation of a bed and settling of an individual particle in still water. The material is placed into the columns (well mixed). Thereafter the hindered settling process is recorded with camera's. From this data set the hindered settling velocity can be determined as shown in the graph below. The higher the concentration, the lower the settling velocity because more particles hinder each other. For concentrations larger than 150 g/l, the settling velocity reaches zero. The particles are close enough to each other to form a (gelled) bed. The maximum fall velocity is determined by the weight of the particle and the surrounding concentration. However it should be taken into account that these are controlled laboratory conditions. In the field the plume be exposed to currents or storm conditions enforcing uplift conditions.

Figure 3: hindered settling regime of the three samples for given concentration.

3) Settling and consolidation columns

To (partly) mimic the conditions at sea, the tailings are placed in 1m high columns. For every sample an expected minimum and maximum concentration is used. The material is mixed and for 3 weeks the experiment was recorded by an automated camera (Figure 4).

Figure 4: from left to right 1Set-up consolidation experiments - 2 After 10 minutes of hindered settling phase - 3 After 9 hours: consolidation phase 1

After settling the material will start to consolidate and the strength of the bed will increase. The strength of the bed determines the sensitivity to re-suspension and possible dispersion of the tailings . At three moments in time the strength of the bed is measured by vane tests. A vane is a rotating rod with 4 blades at the end. The soil resistance to rotation is a measure for the strength. Figure 5 shows an illustration of the device and an results of sample 47. 

Figure 5: Left: Vane test, Right: Result of measures peak shear strength

The strength of the bed was measured after 20 hrs, 1 week and 3 weeks of settlement. The results are shown in Figure 7. As can be seen, there is an rapid increase in strength in the first week. After this week the rate of increase decreases. This is according to theory. Thereafter there are some samples with a lower strength in week 3 compared to week 1. Probably this is due to artifacts.