Subscriber access provided by MEMORIAL UNIV
Article
Carboxymethylcellulose Adsorption on Molybdenite: The Effect of Electrolyte Composition on Adsorption, Bubble-Surface Collisions, and Flotation Mohammad Kor, Piotr M Korczyk, Jonas Addai-Mensah, Marta Krasowska, and David A Beattie Langmuir, Just Accepted Manuscript • DOI: 10.1021/la503248e • Publication Date (Web): 18 Sep 2014 Downloaded from http://pubs.acs.org on October 1, 2014
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 31
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Carboxymethylcellulose Adsorption on Molybdenite: The Effect of Electrolyte Composition on Adsorption, Bubble-Surface Collisions, and Flotation Mohammad Kor, Piotr M. Korczyk†, Jonas Addai-Mensah, Marta Krasowska, and David A. Beattie* Ian Wark Research Institute, University of South Australia, Mawson Lakes SA 5095, Australia.
* Corresponding Author: Email:
[email protected], Telephone number: +61 8 8302 3676 †
Current Address: Institute of Fundamental Technological Research, Polish Academy of
Sciences, Warsaw, Poland.
ACS Paragon Plus Environment
1
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 31
ABSTRACT The adsorption of carboxymethylcellulose polymers on molybdenite was studied using spectroscopic ellipsometry and atomic force microscopy imaging, using two polymers of differing degree of carboxyl group substitution and at three different electrolyte conditions: 1 x 10-2 M KCl; 2.76 x 10-2 M KCl; and simulated flotation process water of multicomponent electrolyte content, with an ionic strength close to 2.76 x 10-2 M. A higher degree of carboxyl substitution in the adsorbing polymer resulted in adsorbed layers that were thinner and with more patchy coverage; increasing ionic strength of the electrolyte resulted in increased polymer layer thickness and coverage.
The use of simulated process water resulted in the largest layer
thickness and coverage for both polymers. The effect of the adsorbed polymer layer on bubbleparticle attachment was studied with single bubble-surface collision experiments recorded with high speed video capture and image processing, and also with single mineral molybdenite flotation tests. The carboxymethylcellulose polymer with a lower degree of substitution resulted in almost complete prevention of wetting film rupture at the molybdenite surface under all electrolyte conditions. The polymer with a higher degree of substitution only prevented rupture when adsorbed from simulated process water. Molecular kinetic theory was used to quantify the effect of the polymer on dewetting dynamics for collisions that resulted in wetting film rupture. Flotation experiments confirmed that adsorbed polymer layer properties, through their effect on the dynamics of bubble-particle attachment, are critical to predicting the effectiveness of polymers used to prevent mineral recovery in flotation.
ACS Paragon Plus Environment
2
Page 3 of 31
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
INTRODUCTION Flotation is a mineral separation process that relies on the differences in wettability of solid particles suspended in water 1. Bubbles are used to harvest hydrophobic valuable particles, and transport them to the top of flotation cells for collection. In complex flotation systems, such as flotation of metal sulfide ores, selective separation of minerals does not take place without the aid of added reagents in the mineral suspension. Water-soluble polymers are used to selectively alter particle/bubble interactions for unwanted minerals, to aid the separation. One such polymer that has seen extensive use is carboxymethyl cellulose (CMC). CMC is an anionic polysaccharide with a broad application as a flocculant 2, dispersant
3, 4
and depressant (a
polymer that prevents bubble-particle attachment) 5, 6. The adsorption of CMC on silicate minerals in the flotation of sulfide ores (particularly on talc) has been studied extensively 7, 8, 9, 10, 11, 12, 13, 14 by our research group and others. One of the main approaches of our research has been to focus on understanding the effect that adsorbed polymer layer properties have on bubble-surface interactions and flotation outcomes. Parameters such as coverage, layer thickness, hydration content, and rigidity (in addition to hydrophobicity) have been shown to be critical in the action of the polymer in a flotation experiment 15, 16. In some instances, polymer depressants like CMC can also influence the recovery of valuable minerals, such as molybdenite. In the commercial processing of copper-molybdenum ores, the common flotation practice of separating molybdenite from chalcopyrite relies on a two-stage flotation process. First, a bulk chalcopyrite–molybdenite concentrate is produced, followed by separation of the collective concentrate. In the second separation stage, chalcopyrite is depressed from the bulk concentrate typically using sodium hydrosulfide, sodium cyanide, ammonium sulfide or Nokes reagent. However, given the toxic nature of these chalcopyrite depressants,
ACS Paragon Plus Environment
3
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 31
there have been a number of studies conducted to determine the effect of non-toxic organic polymers on molybdenite and chalcopyrite flotation recovery 17, 18, 19, 20, 21, 22. In most cases, the polymers affect molybdenite flotation much more than chalcopyrite (the same is also true for the use of polymer depressants in molybdenite/talc flotation
23, 24
), thus
making them ineffective for traditional copper-moly circuits. As a result, it is important to determine what factors of the flotation solution environment cause such strong adsorption of flotation depressants on molybdenite, and also to see if there are structural modifications that can be made to alter their impact on flotation. This information will be invaluable in the search for novel environmentally benign depressants for targeting chalcopyrite over molybdenite. In this study, two CMC polymers with specific variation in functional group degree of substitution and distribution of substitution are studied in their adsorption onto molybdenite. The adsorption of the polymers on molybdenite is characterized with AFM and ellipsometry, and the effect of the adsorption on the mineral surface properties is characterized with bubble-surface collision studies and single mineral flotation testing. In addition, the effect of background electrolyte (simple KCL of various ionic strength, or simulated process water) on polymer adsorption, bubble surface collisions, and flotation, is studied.
EXPERIMENTAL Materials: Molybdenite particles used for flotation were purchased from Fluka (>99 % pure). The Brunauer-Emmett-Tellet (BET) surface area was measured at 3.1 m2g-1. The particle size distribution was determined by a Mastersizer X (Malvern instruments, UK), giving a particle distribution of 0.5-100 µm, with a D10 of 3.0 µm, a D50 of 7.4 µm, and a D90 of 15.5 µm (see Figure 1 for a SEM image of the molybdenite particles). For in situ AFM and bubble-surface
ACS Paragon Plus Environment
4
Page 5 of 31
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
collision experiments, molybdenite from the Northern Territory was obtained from the Mineralogy Department of the South Australian Museum (see Figure 1 for a photograph of the bulk mineral sample), and cleaved flat surfaces were obtained by peeling the top layer of the mineral sample using clean tweezers and/or a scalpel blade, revealing a flat freshly cleaved basal plane. All molybdenite was free from significant impurities, as verified by X-ray photoelectron spectroscopy (XPS) analysis. High purity Milli-Q water with a resistivity of 18.2 MΩ cm, a surface tension of 72.4 mN m-1 at 22 °C, and a total organic carbon component of less than 4 ppm was supplied by an Advantage A10 system (Millipore, USA). The solution pH was adjusted by adding of small quantities of volumetric grade of 0.1 M HCl and KOH solutions. All experiments were carried out in 10-2 M KCl, 2.76×10-2 M KCl background electrolyte and simulated process water at pH 9, unless otherwise stated. For AFM, ellipsometry and bubble collision experiments ultra-pure KCl was used. KCl (99.9%) was purchased from Merck and further purified via calcination (8 h at 550 °C), recrystallization and a second calcination (8 h at 550 °C). This ensures removal of any surface active impurities that could affect adsorption of CMC and bubble rise velocity. In addition to experiments in KCl solutions, experiments were also performed in simulated process water, in this case, simulated process water based on that from a mine site in Australia that processes molybdenite ore. The composition of the water is as follows: CaSO4 2.745 mmol/L; Mg(NO3)2 0.411 mmol/L; Na2SO4 4.653 mmol/L; and KCl 1.458 mmol/L. This water is hereafter referred to as SPW. Two carboxymethylcellulose polymers (water-soluble sodium salts, for structural diagram, see Figure 1) were used as supplied from CP Kelco. These polymers consist of different degree of substitution (the average number of carboxymethyl groups attached into the glucose unit) and different relative blockiness (carboxymethyl sequence distribution within the cellulose chain;
ACS Paragon Plus Environment
5
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 31
low blockiness implies random/even distribution along the polymer chain): HSHB – high substitution (around 0.7 DS), high blockiness; LSLB – low substitution (around 0.4 DS), low blockiness. The molecular weight averages and polydispersity index (Mw/Mn), determined by size exclusion chromatography (SEC)
25
are 165,000 kDa and 3.8 for HSHB, and 157,000 kDa
and 3.6 for LSLB. Polymer stock solutions of 2000 mg.L-1 were prepared by weighing the appropriate mass of solid polymer and dissolving in background electrolyte solution and stirring them overnight to ensure complete hydration. Prior to use, solutions of the desired concentration (5 mg.L-1) were prepared by dilution in the relevant electrolyte, and then pH modified to pH 9 using volumetric grade potassium hydroxide 3. Background electrolyte solutions (KCl and SPW) for AFM and bubble collision experiments, in order to remove dissolved CO2, were purged with high purity (99.99%) dried N2 before having their pH adjusted. All solutions were prepared and used within 24 h.
Figure 1. (a) photograph of molybdenite sample type used for AFM imaging, ellipsometry, and bubblesurface collisions (with water droplet); (b) SEM image of molybdenite particles used for flotation; (c) chemical structure of carboxymethylcellulose (adapted from Coultate 26)
Spectroscopic Ellipsometry: Ellipsometry measurements were performed on a vertical variable angle spectroscopic ellipsometer from J. A. Woollam V-VASE (J.A. Woollam Co) equipped with a 5mL vertical liquid cell (model TLC-100) to obtain the in situ thickness of the polymer
ACS Paragon Plus Environment
6
Page 7 of 31
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
layer on top of the mineral surface. The experiments were performed in full spectroscopic (3501100 nm) mode to determine the adsorbed layer optical properties and equilibrium layer thickness. The ellipsometry parameters, amplitude ratio (ψ) and phase difference of reflected p and s polarized light (∆), were measured for the required range of wavelengths at the angle of incidence close to 75°. The WVASE32 software (J.A. Woollam Co) was employed for modelling and data analysis of the ellipsometry measurements.
Equilibrium measurements
involved conditioning the molybdenite surface in the liquid cell with the relevant polymer solution for 30 minutes, followed by a rinsing stage with background electrolyte solution. The measured experimental ψ and ∆ are fitted to ψ and ∆ values calculated from a four layer model (molybdenite/roughness layer/CMC/ambient layer) where the parameters in the calculation are the thickness and optical properties. A regression analysis was used to vary the parameters of the model to obtain a best fit to the experimental data. Optical constants for ambient layer (water-KCl) were available in the library of the WVASE software. For determination of the molybdenite optical constants, ellipsometry spectra were taken without a liquid cell at 65, 70, and 75° angle of incidence in air from the freshly cleaved molybdenite: the refractive index, and the extinction coefficient, as a function of wavelength using the WVASE program. Furthermore, an intermediate roughness layer containing 50% volume CMC and 50% volume molybdenite was introduced to account for variations in the substrate over the area analysed (less than 0.3 nm in all cases). The fitting to obtain the thickness of the CMC layer used a Cauchy medium
27
to represent the adsorbed polymer layer (assuming the layer is a
homogeneous transparent medium), with a refractive index, n, defined by the following formula:
݊ሺߣሻ = ܣ+ ఒమ + ఒర
(1)
ACS Paragon Plus Environment
7
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 31
where λ is the wavelength of the light, and A, B, and C are constants determined from the fitting process. Atomic Force Microscopy Imaging: AFM imaging was carried out in situ, in peak force tapping (PFT) mode using a Nanoscope MultiMode 8 AFM (Bruker) with a Nanoscope V controller. In PFT AFM, the cantilever is oscillated with an amplitude of 100-300 nm and at a frequency of 2 kHz (much below its resonance frequency). In addition, the z-piezo is driven with a sinusoidal waveform (as opposed to a triangular waveform used in contact or tapping mode). Such system allows precise control and minimization of any normal and lateral forces. This prevents alteration/deformation of soft samples
28
for instance when imaging an adsorbed
polymer layer on a mineral surface. In situ imaging was performed using a commercially available quartz fluid cell in open configuration, i.e. without an o-ring. Bare mineral surfaces were conditioned for 30 minutes in a desired polymer solution. After conditioning the polymer solution was exchanged for the desired background electrolyte solution. In order to minimize the effect of solution evaporation (and therefore an increase in the ionic strength of the solution) during scanning, a small amount of solution (~150 µL) was added every 30 min. A piezoelectric scanner E, with a maximum 10 µm x 10 µm scan size in the XY direction and nominal 2.5 µm in the Z direction was used to collect several 2 µm x 2 µm images of the sample. To acquire images silicon nitride cantilevers with a resonance frequency between 40 and 75 kHz, a spring constant between 0.12 and 0.48 N/m, and a sharp silicon (nominal tip radius 2 nm) tip (SCANASYSTFLUID+, Bruker) were used. All images were taken at high resolution (768 x 768), resulting in a pixel size of 2.6 nm, similar to the nominal tip radius. Scan rates employed in imaging were 0.99 Hz or lower. Images were analysed using WSxM 4.0 SPMAGE 09 Edition (Nanotec)
29
and
ACS Paragon Plus Environment
8
Page 9 of 31
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
NanoScope Analysis v1.5 (Bruker) software packages. The AFM images were fitted with a firstorder plane fit (to remove the image tilt). Bubble-Surface Collisions and Receding Contact Angle Measurements: Bubble-surface collisions and the evolution of the receding contact angle on bare molybdenite and polymertreated molybdenite surfaces were carried out using a rising microbubble apparatus. Flat, ~ 1.5 cm x 1.5 cm samples of molybdenite were attached (using thermal glue – EPON 1004) to a clean glass slide. The fresh molybdenite surface was exposed by peeling the top layer off. Such surfaces were immersed in the desired polymer solutions for 30 minutes prior to measurements. The molybdenite samples were then rinsed with background electrolyte prior to immersion into the microbubble apparatus borosilicate glass column of a square (30 mm x 30 mm) cross-section containing the same background electrolyte. A microfluidic chip with a T-junction was mounted at the bottom of the column to generate microbubbles (bubble diameter between 428 and 484 microns). The time interval between subsequent bubbles was 30-60 s ensuring the bubbles were rising isolated from any effects coming from the previous and/or the next bubble. A schematic diagram of the bubble-surface collision experimental setup is given in the supporting information. The molybdenite surface was located at 8 cm from the point of bubble formation, ensuring bubbles were rising with their terminal velocity (51.75 – 60.10 mm/s for bubbles in the size range 428 – 484 µm a) prior they started to approach and interact with molybdenite surface (Reynolds numbers, Re, for such the bubbles are between 24 and 33). The measured terminal
a
The choice of the bubble size was governed by the fact that, in order to study the bubble attachment process (in terms of stability of wetting film under quasi-equilibrium condition), kinetic effects such as the bubble bouncing off the surface must be minimised. Larger bubbles have higher terminal velocity and would need first to dissipate their kinetic energy before staying ‘entrapped’ beneath the surface for the film drainage to start30 [30. Zawala, J.; Krasowska, M.; Dabros, T.; Malysa, K. Influence of Bubble Kinetic Energy on its Bouncing During Collisions with Various Interfaces. Can. J Chem. Eng. 2007, 85, 669-678.]
ACS Paragon Plus Environment
9
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 31
velocities are in a good agreement with the ones predicted by Klaseboer et al. for bubbles of 0
