Advertisement

Mining, Metallurgy & Exploration

, Volume 36, Issue 1, pp 111–116 | Cite as

Rheology of ore suspensions with fibrous minerals and its impact on flotation performance

  • P. SomasundaranEmail author
  • P. Patra
  • Tarun Bhambani
  • D. R. Nagaraj
Article
  • 65 Downloads

Abstract

Flotation separation and recovery of value minerals from ores containing fibrous silicate minerals are known to be quite challenging. Earlier studies have established that the metallurgical challenges (poor selectivity and recovery, high energy consumption, etc.) arise from the impact of the fibrous minerals (with high aspect ratio) on pulp rheology, which was found be rather complex. The complex rheological behavior of ore suspensions containing particles with high aspect ratio and its influence on flotation outcome are investigated in this study. Flotation tests and rheological measurements were carried out with both ores (containing fibrous minerals) and a model system comprising nylon fibers (chemically inert) added to a copper ore that had no fibrous mineral. These studies allowed the determination of contributions of physical properties of the fiber to pulp rheological behavior. Optical microscopy images of the fiber pulp indicated that the fibers entangled to form two levels of microstructures: two-dimensional (2D) aggregates under semi-dilute (< 20% solids) conditions and three-dimensional (3D) microstructures at concentrations > 30% solids. The yield stress (σ) values determined for the fiber pulp were logarithmically related to the volume fraction of the 2D/3D structures. The relationships developed here for different ARs showed that the yield stress of a flotation pulp changes with respect to shape, size, and the concentrations of the microstructures. Implications to the effects of such entangle networks on selectivity of flotation separation are discussed.

Keywords

Microstructure Planar aggregates Aspect ratio Flotation Froth phase 

1 Introduction

Rheological behavior of suspensions containing fibrous particulates is of importance in governing processes in the pulp and paper industries and in froth flotation beneficiation of ores containing fibrous minerals. For suspensions containing long and flexible fibers, such as chrysotile, experimental and theoretical studies of the rheological behavior are often complex due to the interdependence of the factors that contribute to the rheological behavior, such as fiber-wall interactions, fiber-solvent interactions, and most importantly inter-fiber interactions [1, 2]. Inter-fiber interactions are mostly the result of the suspension hydrodynamics that leads to the formation of regions of high concentrations of fibers, otherwise referred to as fiber flocs with entangled structures [3, 4]. With the increase in the fiber concentrations of suspensions, the probability of fiber-fiber contacts increases, leading to the formation of network structures that spread throughout the suspension [5]. Such network structures are inhomogeneous in terms of the shapes and sizes of the fiber flocs. The associative strengths of such flocs differ depending on the aspect ratios (ARs) of the fibers, where larger fibers make more contact points than the smaller fibers, and the net cohesive strengths become higher owing to the frictional forces at the contact points. Here, we emphasize development of a correlation between the aspect ratio (AR—typically represented as width (thickness)/length) of fibers and types of entangled structures that are formed with an increase in fiber concentrations.

In this investigation, froth flotation studies conducted with pulp containing fibrous chrysotile minerals and nylon fibers that were chemically inert. The fibrous chrysotile minerals pose challenges in the selective flotation separation of minerals [6], which has been attributed to poor collector adsorption due to slime coating (electrostatic attraction between positively charged fibrous chrysotile and negatively charged valuable minerals), a widely proposed cause for reduced mineral separation [7]. In this investigation, the focus was on assessing the contributions of morphological aspects of chrysotile fibers. Chrysotile belongs to the serpentine group of minerals with fibrous morphologies [8]. Our earlier work had indicated that suspensions containing fibrous chrysotile minerals are viscous and, therefore, selective floatability is significantly affected; a decrease in value recovery occurs with an increase in suspension viscosity due to an increase in pulp fiber content [9, 10]. However, the impact of fiber ARs and characteristics of the microstructures on the rheological behavior and the forth phase stability has not been fully explored. In this study, the rheological and flotation studies were carried out with three different types (different ARs) of fibrous ores in order to determine the morphological effects of fibrous particles on the flotation performance and the rheological behavior. In order to isolate physical effects from surface chemical effects, experiments were conducted with a model ore system in which nylon fibers with different ARs were added to a Cu ore that had no fibrous minerals. Nylon is chemically inert under conditions prevailing in flotation pulps; therefore, the surface chemical contributions from nylon fibers to flotation performance, such as slime coating, were deemed negligible. The emphasis of this work was to understand the type of the microstructures that are formed with an increase in both content and ARs of fibers and how such microstructures affect pulp rheological behavior.

2 Materials and Methods

Different Ni ore samples, designated as 1, 2, and 3, were obtained from Vale, Inco Corporation, and contained approximately 1 wt.% Ni sulfides and 60–65% of serpentine minerals. The non-fibrous copper ore from a North American mine contained 0.7% copper and had no fibrous minerals based on to SEM (scanning electron microscopy) analysis. Nylon fibers (AR of 1000) of 1 cm long and approximately 10 μm in thickness were obtained from Nyconn Industries. The fibers were subjected to size reduction to obtain an aspect ratio of value around 100.

Flotation tests were carried out at 25% solids in a 2.5-L Denver cell for 12 min at pH 4–5 with the three fibrous serpentine ore types. Ore was ground at 50% solids in a rod mill and flotation was conducted using a dithiophosphate-based collector (40 g/t) and MIBC (20 g/t) frother. The pH was adjusted using 50% H2SO4. Flotation tests in another set of experiments included a set of model ore systems where copper ore was spiked with 1 and 2% nylon fibers. The model ore mixture was ground at 60% solids in a rod mill and tests were done at 25% solids, at pH 10, for 7 min.

Rheological measurements were carried out using an Anton Paar DSR 301 rheometer calibrated with standard Cannon viscosity standards using a vane type probe. Rheological characterization was carried out on three types of suspension samples: (1) Ni ore suspension, (2) copper ore suspension to which nylon fibers in desired amounts were added before grinding, and (3) nylon fiber suspensions. The ore suspension obtained from the rod mill was transferred into a 100-mL sample holder, and a vane geometry sensor was used to measure the yield stress values. Solid concentrations of the ore suspensions were increased in steps by removing 10 ml of water from the suspension in each step using a syringe having a 0.2-μm filter. Yield stress values of the pulp were measured after each step. In a separate set of experiments, nylon fiber suspensions were prepared by adding the fibers to water at a concentration in the range of 1–5 wt.%. Details of the rheological characterization are similar to that described elsewhere (Bennington, 1990). In brief, the rheological measurement procedure using an Anton Paar DSR 301 rheometer involved three sequential steps: (1) homogenization of the suspension, (2) suspension stabilization for 5 s, and (3) measurement of yield stress. Homogenization of samples was carried out to uniformly disperse the particles in suspension and avoid settling of particles during rheological measurements in step 3. In step 1, homogenization shear rate and time were standardized by trial and error where the criteria were to prevent particle settling and repeatability of yield stress values. Step 2 was incorporated to eliminate stress inertia developed in the homogenization step. Thus, the parameters (shear rate, ramp values, and ramp speed) that were set in the three steps were standardized. A stress ramp (Δσ/Δt = 1 Pa/s) was applied until the deformation γ diverged and flow occurred which marked the end of the experiment. The speed of the stress ramp Δσ/Δt) did not affect the trend observed with the variations in yield values at different percentages of solids in suspension but only the shape of the γ divergence.

Optical images of the nylon fiber suspension were taken using an Olympus Optical microscope. For SEM studies, the ore suspension samples were acquired from the ball mill and were tumbled in 10-mL glass vials. Drops of the ore suspensions taken from the vials were mounted on an SEM stub. A technique known as acetone-replacement drying [11] (Fitzpatrick, 1993) was used instead of conventional drying to maintain the structural integrity of the sample. The samples after drying were carbon-coated and SEM micrographs were obtained using the Zeiss DSH 982 Gemini SEM system and a low electron beam voltage (3 KeV).

3 Results and Discussions

Relationships and correspondence between results obtained from flotation and rheological studies were analyzed to determine the role of the suspension microstructures in flotation outcome (both recovery and grade).

3.1 Flotation and Rheological Studies with Different Ore Types

Results from ore flotation tests and pulp rheology measurements are shown in Fig. 1 wherein Ni recovery and pulp yield stress are plotted against average aspect ratio of chrysotile mineral; SEM images (Fig. 1) of the pulp samples of the three different ore types are also shown.
Fig. 1

(Top) Ni recovery (empty circles) and yield stress values (filled circles) obtained with ore types having varying aspect ratios; (bottom) SEM of ore types (left to right) 1, 2, and 3

Flotation recovery of Ni was higher (~ 90%) for ore type 1 than that for ore type 2 (~ 81% Ni recovery). The yield stress value for the pulp of ore type 2 was ~ 50 Pa, whereas that for ore type 1 was < 5 Pa. Thus, the decrease in the Ni recovery for ore type 2 could be attributed to the increase in the yield stress of the ore pulp.

3.2 Impact of aspect ratio

As reported earlier [6], chemical factors such as slime coating or other factors owing to mineral surface chemistry could play a role in the reduced recovery of Ni minerals in froth flotation processes. In order to isolate such contributions due to chemical factors leading to poor recovery of Ni, flotation and rheological studies were carried out with a model ore system containing chemically inert nylon fibers of different ARs (described in the experimental section).

The yield stress values of suspensions were determined for fibers with ARs (length/diameter) of 100 and 1000. The Hershel Buckley/Casson model was used to assess the rheological behavior of suspensions. The yield stress values were 40 and 175 Pa for suspensions (1 wt.% fiber) with fibers of AR 100 and 1000, respectively (Fig. 2). The yield stress values at 2% (by wt.) of fibers were observed to be 160 and 409 Pa for suspensions with fibers of ARs of 100 and 1000. The pulp became increasingly viscous as the AR of the fibers increased. A correlation between the increase in the yield stress values and a decrease in Cu recovery was observed (Fig. 2). The copper recovery decreased from 82 to 65%, with 1% of nylon fibers of AR 100, and the recovery decreased to 54% with 1% of nylon fibers with AR 1000. At 2% of fibers with AR 100, the Cu recovery was only 28%; conducting reliable flotation at 2 wt.% of AR 1000 was difficult as the pulp was almost immovable. Even though the fiber content by weight was the same in the suspensions, the rheological behavior and flotation recoveries of the value minerals were quite different. In rationalizing this correlation, it is postulated that depending on the fiber ARs, characteristic (in terms of shape and volume) microstructures are formed with an increase in their concentrations. Size and shape of the microstructures are postulated to influence pulp rheological behavior and, therefore, flotation performance. To explore these concepts further, we examined the shape and size of microstructures as a function of the fiber AR.
Fig. 2

Copper recovery and yield stress values obtained on ground copper ore pulp spiked with nylon fibers of varying aspect ratios, dashed lines are for 100 AR nylon fibers and dotted lines are for 10,000 AR nylon fibers. Copper recovery value for 2% of 10,000 AR fibers are not shown as the pulp was immovable and froth was absent which is critical for value mineral recovery

The yield stress values of ore suspensions (30–60% solids) with chrysotile and nylon fibers were measured to determine the effect of microstructures in suspensions. Fig. 3a shows that the yield stress values measured were higher for ore 3 suspensions with aspect ratio of 750 (avg.) compared to that of ore 2 suspensions with AR 110 (avg.). At a particular solids concentration value, the yield stress for the ore suspensions increased in the order ore 3 > ore 2 > ore 1. Ores 2 and 3 exhibited different trends in terms of the slope of the yield stress versus concentration (30–60%) profiles. The differences in the slope S = Δσ/Δc (where σ is yield stress and c is concentration) are proposed to be due to the formation of AR-dependent unique microstructures in the suspensions. In the concentration regime of 30–40% solids, the S values were 1.1 and 1.5 for ores 2 and 3 respectively. The S values increased to 1.6 and 10 for the concentrations in the 40–50% range and to 1.7 and 20 in the 50–60% concentration regime. These S values indicated that three concentration regimes exist where the yield stress values increased with respect to fiber content exhibiting characteristic correlations.
Fig. 3

Yield stress values of ore pulp containing a chrysotile and b nylon fibers of varying aspect ratios. Yield stress values were taken upon increasing the % solids in the suspension

To study the observed trend in S values with fiber AR and concentration (three regions), yield stress values were determined for the ores (devoid of any fibers) spiked with nylon fibers of specific ARs. Figure 3b shows that yield stress values for ore containing nylon fibers exhibited trends similar to that of ores 2 and 3 in terms of slope in the range of 30–60% solids (Fig. 3a). In general, the yield stress values measured for the ore suspensions increased with an increase in the AR. The slope values, S, for ore containing nylon fiber of ARs 100 and 1000 are shown in Table 1. An increase in the slope values was observed with an increase in either solids concentration or AR of the fibers.
Table 1

S (S = Δσ/Δc) for ore containing nylon fibers of varying ARs and concentrations

Aspect ratio and concentration

30–40% solids

40–50% solids

50–60% solids

AR 100–1%

0.6

2.3

3.14

AR 100–2%

1.7

5.3

6.1

AR 1000–1%

3.4

4.2

13.5

AR 1000–2%

3.5

12.2

20.5

The data reported for the ARs are with 5% error

This suggested that the number of the entangled fiber structures increases with the increase in fiber concentration, and the structures that are formed by the longer fibers are possibly different in terms of volume fraction and the three-dimensional aspects of the entangled structure from that of the shorter fibers. Figures 4 b and c demonstrate the formation of the 2D microstructures in the fiber concentration range of 1.5–3% (wt.). At concentrations > 3% (wt.), 3D microstructures were observed to be formed (Fig. 4 d–f).
Fig. 4

Optical microscopic pictures of nylon fibers in suspension. a < 0.5%, b ~ 1.5%, c ~ 2%, d ~ 3%, e ~ 4%, and f indicated higher magnification micrograph of dotted box shown

3.3 Conceptual relation between entangled fiber structures and rheological behavior

From Fig. 5, it can be observed that for nylon fibers of AR 1000, three different regimes were observed in terms of the S (Δσ/Δc) values. Three distinct rheological behavior (in terms of yield stress) regimes were observed with the increase in the suspension fiber content (Fig. 5). This correlation suggests that 2D aggregates formed in the concentration range of 2–4% and 3D aggregates formed at concentrations > 4% (wt.), which in turn caused an increase in the yield stress values. The occurrence of similar regimes has been reported for the rheological behavior of polymer solutions, which was based on theoretical calculations indicating the formation of polymer-entangled structures [12]. For example, in a very dilute solution where concentration of fibers is less than 1/L3 (L, length of polymers), the Brownian motion of the polymers is independent. With an increase in concentration [c > 1/L3], the rotational, end-over-end, motion of each polymer chain is restricted and the translational motion perpendicular to the rod axis leads to macroscopic effect such as entanglement of fibers [13].
Fig. 5

Yield stress values (square points) of pulp containing nylon fibers (AR 1000) in varying concentrations. The spherical points indicate yield stress for the suspensions containing 100 μm silica particles of equal weight of nylon fiber. The dotted lines are shown to demonstrate change in the slope S (Δσ/Δc)

3.4 Effect of suspension microstructures on the behavior of flotation froth

It was clear from the flotation tests that the froth generated was dramatically different for the three ore types (with different ARs) and the model ore system with nylon fibers. This suggests a link between suspension microstructures and froth properties. Figure 6a shows a rich mineralized froth that forms when fibrous minerals were absent in the pulp. The froth phase was less mineralized when 2% fibers of AR 100 were present (Fig. 6b). Additionally, large networks were observed to be transporting to the froth phase. With 2% fibers of AR 1000 (Fig. 6c) the froth phase was absent and the pulp was almost immovable. Thus, the structures formed by the large fibers cause a drastic increase in the flotation pulp viscosity leading to poor froth formation and flotation performance.
Fig. 6

Snapshots of froth phase a in absence of any fibers and b in presence of 2% (wt.) of ~ 1 mm nylon fibers (AR 100) and c 2% of 1 cm nylon fibers (AR 1000)

4 Summary and Conclusions

Ore suspensions become more viscous and have higher yield stress when high aspect ratio (AR) fibrous minerals (or fibers of other solids such as nylon) are present. There is also a concomitant adverse impact on the flotation performance from such ore suspensions. With an increase in fiber concentrations in suspensions, two types of macro-structures of sizes as large as 1–2 cm are formed: planar (2D) aggregates and three-dimensional flocs. Contributions from these 2D and 3D aggregates to the rheological behavior is more than the contribution expected from simply accounting for the volume fractions of the fibrous particles in a suspension. This study reveals that reducing the ARs of the fibrous minerals in froth flotation pulp is a plausible approach to improve froth flotation recovery of valuable minerals from ores containing fibrous minerals.

Notes

Acknowledgements

The authors acknowledge the support of Manqiu Xu, Zongfu Dai, Andrew Lee, and Ken Scholey at the Vale Technical Services in Mississauga, Canada. The authors acknowledge the support received for this research work from the National Science Foundation Industry/University Collaborative Research Center of Particulate and Surfactant Systems (IIP 1362078).

Compliance with ethical standards

Conflict of Interest

The authors declare that they have no conflict of interest.

References

  1. 1.
    Schmid CF, Switzer LH, Klingenberg DJ (2000) Simulations of fiber flocculation: effects of fiber properties and interfiber friction. J Rheol 44:781–809.  https://doi.org/10.1122/1.551116 CrossRefGoogle Scholar
  2. 2.
    Bennington CPJ, Azevedo G, John DA, Birt SM, Wolgast BH (1995) The yield stress of medium- and high-consistency mechanical pulp suspensions at high gas contents. J Pulp Paper Sci 21(4):111–118Google Scholar
  3. 3.
    Mason SG (1950) The flocculation of pulp suspensions and the formation of paper. TAPPI J 33:440–444Google Scholar
  4. 4.
    Schmid CF, Klingenberg DJ (2000a) Mechanical flocculation of flowing fiber suspensions. Phys Rev Lett 84:290–293.  https://doi.org/10.1103/PhysRevLett.84.290 CrossRefGoogle Scholar
  5. 5.
    Kerekes RJ, Schell CJ (1992) Characterization of fiber flocculation regimes by a crowding factor. J Pulp Paper Sci 18:J32–J38Google Scholar
  6. 6.
    Eltham JA, Tilyard PA, (1973) An approach to the flotation of Western Australian Ni ores, Proceedings, Australasian Institute of Minerals and Metallurgy Annual Conference, Western Australia, 417–429Google Scholar
  7. 7.
    Edwards CR, Kipkie WB, Agar GE (1980) The effect of slime coatings of the serpentine minerals, chrysotile and lizardite, on pentlandite flotation. Int J Miner Process 7:33–42.  https://doi.org/10.1016/0301-7516(80)90035-6 CrossRefGoogle Scholar
  8. 8.
    Rinaudo C, Gastaldi D, Belluso E (2003) Characterization of chrysotile, antigorite and lizardite by FT-Raman spectroscopy. Can Mineral 41(4):883–890.  https://doi.org/10.2113/gscanmin.41.4.883 CrossRefGoogle Scholar
  9. 9.
    Patra P, Bhambhani T, Vasudevan M, Nagaraj DR, Somasundaran P (2012) Transport of fibrous gangue mineral networks to froth by bubbles in flotation separation. Int J Miner Process 104-105:45–48.  https://doi.org/10.1016/j.minpro.2011.11.007 CrossRefGoogle Scholar
  10. 10.
    Patra P, Nagaraj DR, Somasundaran P, (2010) Impact of pulp rheology on selective recovery of value minerals from ores, XI international seminar on mineral processing technology, 15–17 December, NML, Jamshedpur, IndiaGoogle Scholar
  11. 11.
    Fitzpatrick EA (1993) Soil microscopy and micromorphology. John Wiley and Sons, New York ISBN-10: 0471938599Google Scholar
  12. 12.
    Doi M, Edwards SF (1986) The theory of polymer dynamics. Clarendon Press, Oxford ISBN: 9780198520337Google Scholar
  13. 13.
    Doi M, Edwards SF (1978) Dynamics of rod-like macromolecules in concentrated solution: part 1, J Chem Soc, Faraday Trans 2, (74), 560–570.  https://doi.org/10.1039/F29787400560

Copyright information

© Society for Mining, Metallurgy & Exploration Inc. 2019

Authors and Affiliations

  1. 1.Department of Earth and Environmental EngineeringColumbia UniversityNew YorkUSA
  2. 2.Cytec Solvay Industries Inc.Stamford Technology CenterStamfordUSA

Personalised recommendations