SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria

Minerals Engineering xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Minerals Engineering journal homepage: www.elsevier.com/locate/mineng

SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria Guy Nkulu a, Stoyan Gaydardzhiev b,⇑, Edouard Mwema a, Philippe Compere c a

GECAMINES Sarl, Lubumbashi, Democratic Republic of the Congo University of Liege, Mineral Processing and Recycling, Sart-Tilman, B52, 4000 Liege, Belgium c University of Liege, Department of Biology, Ecology and Evolution & Centre of Applied Technology in Microscopy, Sart-Tilman, B6c, 4000 Liege, Belgium b

a r t i c l e

i n f o

Article history: Received 21 July 2014 Revised 23 November 2014 Accepted 5 December 2014 Available online xxxx Keywords: Carrollite Bioleaching Mesophilic bacteria SEM Mineralised polymer substances

a b s t r a c t Bioleaching of high purity carrollite minerals with mesophilic bacteria was carried out and monitored by observations in scanning electron microscopy (SEM) and elemental X-ray microanalysis (EDS) to provide evidence of the interaction pattern between carrollite and microorganisms. A bacterial consortium involving three different acidophilic chemolithotrophs was adopted. The evolution of the surface topography, inside alteration effects and elemental composition of the mineral with leaching time was followed. It could be postulated that bacterial adhesion takes place on the mineral surface, resulting in the formation of dissolution pits of various shapes and continues by boring elongated channels deep inside the mineral grains. Enhanced concentration of ferric iron and sulphur could be assumed in vicinity of the zones where mineralized polymer substances are precipitated. It could be inferred that carrollite dissolution is governed by cooperative bioleaching involving oxidation induced by bacteria attached to the surface and ferric iron re-oxidized by planktonic bacteria in suspension. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction The ore deposits belonging to the Central African Copperbelt situated between the Democratic Republic of Congo (DRC) and Zambia are known to host vast amounts of cobalt. About half of the global production of cobalt originates from this region (Yager, 2010). These deposits are mainly of sulphide types with carrollite being the major cobalt bearing mineral. Hydrometallurgy is by far the most common way of treating cobaltiferous ores encompassing copper sulphides with cobalt and copper recovered from the pregnant leach solutions through solvent extraction. The metal bearing sulphate solutions are purified and high-purity copper and cobalt are electrowon (Habashi, 1997). Beyond being a natural phenomenon, bioleaching has turned into a technique which is based on the ability of microorganisms (bacteria) to transform solid phases into soluble and extractable compounds which can be subsequently recovered (Akcil and Deveci, 2010; Ehrlich, 2004). Recent trends, like growing demand for metals coupled with the increasing complex mineralogy of the ore deposits leading to high exploitation costs with conventional methods are additional factors that promote bioleaching as

⇑ Corresponding author. E-mail address: [email protected] (S. Gaydardzhiev).

an attractive hydrometallurgical process. Plenty of studies have proved that the oxidation rate of many sulphide minerals is markedly accelerated when mesophilic bacteria are present (Dziurla et al., 1997; McGuire et al., 2001; Rawlings, 2002). Recently published figures indicate that about 20–25% of the world copper production is derived from bioleaching (Brierley, 2008). Apart from copper, uranium and cobalt, other metals such as nickel and zinc are potential candidates for bioleaching (Brierley, 2010; Gericke et al., 2009). As of recently, biomining has been expanding into full-scale operation in Talvivaara, Finland for recovery of nickel and cobalt (Riekkola-Vanhanen, 2012). Microorganisms related to the species Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Leptospirillum ferrooxidans are characterized as Gram-negative, aerobic and chemoautotrophic, operating at ambient temperatures (mesophilic bacteria). They are known to play an important role in the leaching and recovery of valuable metals from sulphide ores due to their acidophilic character (Brierley, 1997; Krebs et al., 1997). To explain degradation of sulphides, two indirect mechanisms have been proposed (Schippers et al., 1996; Schippers and Sand, 1999). The first one is based upon the oxidative attack of ferric iron on acid-insoluble metal sulphides involving thiosulphate as the main intermediate compound. The second one supposes likewise proton and/ or ferric iron attack on acid-soluble metal sulphides, but with polysulphides and sulphur as intermediates.

http://dx.doi.org/10.1016/j.mineng.2014.12.005 0892-6875/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005

2

G. Nkulu et al. / Minerals Engineering xxx (2015) xxx–xxx

Deeper understanding of the bioleaching phenomena requires fundamental studies not only on physiology, biochemistry and genetics of the microorganisms, but on the nature of reactions occurring at bacteria–mineral interface as well. Therefore, bacterial activity on sulphide mineral surfaces has been investigated with the ultimate aim of overcoming some process-limiting factors for rendering bioleaching as attractive alternative to conventional roasting or pressure oxidation. For instance, the observations that bacteria attach on the pyrite surface by means of a secreted biofilm encapsulating the mineral, as shown in the studies by Mustin et al. (1993), Rojas-Chapana et al. (1995, 1996) and Tributsch et al. (1998), imply that bacterially assisted dissolution of sulphides does not involve ‘‘non-contact’’ mechanism only with leaching by bacterially regenerated iron (III), but a ‘‘contact’’ bioleaching as well involving electrochemically supported dissolution occurring at the interface bacterial cell/mineral. The latter assumes a physical attachment of bacterial cells to the surface of sulphide minerals under aerobic conditions (Silverman, 1967). Scanning electron microscopy (SEM) and energy dispersive spectroscopy of X-rays (EDS X-ray microanalysis) have been intensively used to investigate the attachment patterns of bacteria on mineral surfaces (Jordan, 1993; Gómez et al., 1996; Sampson et al., 2000; Liu et al., 2011; Jiang et al., 2009). However, regardless the number of studies performed so far, research focusing on carrollite behaviour during leaching virtually does not exist. To fulfil this need, the objective of the present study is to contribute in advancing the knowledge on bacteria–substrate interaction through observations into mineral surfaces following their contact with bacteria and to interpret the results in view their relevance to Co and Cu bioleaching. 2. Materials and methods 2.1. Mineral samples High purity carrollite samples accompanied by their dolomitic gangue were handpicked from rich mineralized zones of the Kamoya deposit located in Katanga province of DRC. Selected samples between 25–50 mm in size have been then gently sliced in order to separate intact carrollite crystals from their gangue. The thus obtained mono-crystals of carrollite have been dry-ground using disc mill to obtain material with particle size below 0.053 mm for the leaching. The elemental composition of both solid and liquid samples was determined by atomic absorption spectroscopy (Analytic Jena CONTRAA 300). X-ray diffraction unit Bruker D8-Advance was employed to obtain the mineralogical characteristics of the pure carrollite. The chemical analysis gave the following concentration for the main elements: Co 36%, Cu 19.8%, Ni 1.3%; Fe 2.18%, S 21.5%. The mineralogical composition given in Fig. 1 indicated carrollite as major mineral phase with a small amount of chalcopyrite, bornite and pyrite being detected. 2.2. Microorganisms and culturing medium A bacterial consortium involving three different mesophilic chemolithotrophic bacterial strains belonging to the species A. ferrooxidans, L. ferrooxidans and A. thiooxidans was used. The native strains have been isolated from various acid mine drainage waters and dumps in Bulgaria and kindly supplied for this study by Prof. S. Groudev (UMG – Sofia). The isolation, identification and enumeration of the microorganisms were carried out by described methods (Karavaiko et al., 1988). Bacteria have been successively adapted through several stages on carrollite substrate in iron-free 9 K

C : Carrollite Cp : Chalcopyrite Py : Pyrite D : Dolomite Qz : Quartz

C

C

C C

C C

C D Cp

Py

Qz

Fig. 1. XRD pattern of pure carrollite.

medium (Silverman and Lundgren, 1959), pH 1.8–2.0 and temperature of 33 °C. Once the inoculum has reached its log phase indicated by redox potential in the range 620–640 mV, the suspension was filtered and the resulting solution used as lixiviant during all the experiments. 2.3. Bioleaching procedure Bioleaching of carrollite samples was carried out in triplicate inside shake flasks of 250 mL (working volume) at pulp solids loading of 2% (w/v). The shaker has been placed in a thermostatized room at 33 °C and agitated at 120 rpm. At predetermined time intervals, 2 mL have been sampled from the bioleaching solutions, filtered and delivered for Co and Cu analysis. Eh and pH of the suspension have been monitored on a regular basis and losses due to evaporation compensated by addition of distilled water. Concentrated sulphuric acid has been used to keep pH value between 1.8 and 2. For microscopical study of the evolution of carrollite surface during bioleaching, 10 mL of the pulp was collected at day 5, 10, 15, 20 and 30 and subsequently filtered through filter paper. The remaining solids on the filter surface have been gently washed with 9 K solution and transferred into glass tubings containing 10 mL of 9 K solution before being delivered to SEM-preparation. 2.4. Sample preparation for SEM and EDS analysis The filtered-solids from the bioleached material have been recovered from the glass tubings by sedimentation and pipetting, then glutaraldehyde-fixed for 2 h at 20 °C in 1 mL of iron-free 9 K solution with addition 100 lL of 25% glutaraldehyde, and finally rinsed in iron-free 9 K medium then in distilled water before being separated into two sets. The samples from the first set have been freeze-dried, glued on conductive carbon tape (on aluminium stubs) and coated with 20 nm Pt in a sputtering unit (Balzers SCD-030) before SEM observation. Samples from the second set have been ‘‘en bloc’’ stained in aqueous 2% uranium acetate for 2 days, rinsed in distilled water, dehydrated through an ethanol series and embedded in Epofix resin (Struers, cat no. 40200029). Mirror-polished slices have been realized by hand polishing with SiC grit papers (up to P4000) followed by final polishing with a non-aqueous 1 lm-sized diamond suspension (ESCIL, 1PS-1MIC). SEM observations were carried out by use of secondary electron (SE) and backscattered electron (BSE) detectors on bulk samples and polished slices respectively in a FEI ESEM-FEG XL-30 system working in high vacuum mode and at 15 kV of accelerating voltage. The ESEM was coupled with an EDAX energy dispersive X-ray spectrometer (EDS) for elemental microanalysis and with a sapphire

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005

G. Nkulu et al. / Minerals Engineering xxx (2015) xxx–xxx

super-ultrathin window for detection of light elements. Weight and atomic percentages of elements were estimated with normalized standardless EDAX-ZAF quantitative method. 3. Results and discussion 3.1. Dissolution of carrollite The time course for the extraction rate of Co and Cu in the bioleaching system and the redox potential variation are presented in Fig. 2. It could be noted that the majority of copper and cobalt leaching is achieved within two weeks. Following that period, curves begin to flatten with the percentage of leached Co and Cu after 20 days reaching around 51% and 48% respectively. Between day 20 and 30 there is virtually no recovery of copper and cobalt which remains nearly stationary. It could be noted that the redox potential in the system decreases slightly during the first two days and then rises quickly from the day five on until reaching values of about 630 mV with further on acquiring a plateau-trend after twenty days of leaching. 3.2. Morphology of mineral surfaces in the presence of bacteria

700 50 650 40 600

30

550

copper cobalt Eh

20

500

10

0

Redox potential, mV

Recovery in pregnant leach solution, %

To explain the observed incomplete metals dissolution, the surface morphology of mineral samples has been investigated. It was conceivable that the observations will also provide evidence about the role which microorganisms play in the system. Fig. 3 presents BSE-SEM views of polished sections of carrollite grains before and after 30 days of bioleaching. It is obvious how at the end of the leaching period carrollite is degraded with voids being developed from the grain surface towards the bulk. Further on in the study more detailed observations at various leaching times were performed. The SE-SEM images in Fig. 4 show a micro-topography of leached carrollite grains sampled at day 5, 15, 20 and 30. Settlement of bacteria (b) on the surface, EPS produced near bacteria or covering some grains (asterisks) as well as associated secondary precipitates (p) could be noted. Rod-shaped bacteria are easily distinguished from the bended or S-shaped ones. These findings could be interpreted by taking into account the three phases of surface degradation progress during bioleaching: (1) an initial settlement phase in the first 5 days, (2) an exponential growth phase between day 5 and 15 during which most of mineral leaching occurs, and (3) a final stationary phase after 15 days. Fig. 4A shows

0

5

10

15

20

25

30

450

Leaching time, days Fig. 2. Leaching kinetics of Co and Cu and redox potential with time in the bioleaching system.

3

a typical example of bacterial cells attached to a quasi-intact carrollite surface after 5 days. It could be argued that during and just after the settlement phase, the majority of bacteria remains randomly dispersed on carrollite surface and starts to produce and expel EPS (Extracellular Polymeric Substances). The generated EPS surrounding bacteria become visible after 10 days (Figs. 4B– F). EPS precipitates could encase secondary mineral husks (micro particles) seen in Fig. 4D. The presence of traces of exopolymer and precipitates together with bacteria in the vicinity of the pits seen after day 15 (Fig. 4C and D) and in channel-like cavities of deep alteration zones (Fig. 4E and F) suggest that these compounds together with the bacteria play a role in establishing surface corrosion zones onto carrollite grains. It could be assumed that the intensive bacterial growth occurring between days 5 and 20 is accompanied by a reversible settlement/detachment of bacteria on/from the mineral surface. Among them, the iron oxidizing planktonic bacteria are contributing to re-oxidation of the brought-into-solution ferrous iron. Since copper and cobalt dissolution virtually seized after two weeks, it has been interesting to look in more detail at the morphology of carrollite surface and the alterations occurring after that period – Fig. 5. These observations witnessed weak (A, B, C) and deep alteration patterns (D, E, F). Square-shaped isolated pits (A), locally aligned ones (B) or such covering a large area of carrollite crystals (C) are visible. Corrosion grooves forming parallel-crossed patterns resembling the hexagonal crystallographic structure of carrollite were observed locally, along edges and deep in the bulk of grains. In each case observed, these alterations patterns are accompanied by colonization through rod-shaped (rb), bended (cb) or S-shaped (sb) bacteria, together with secondary precipitates (p), sometimes met as aggregates Fig. 5E. Fig. 6 provides details on examination of carrollite grains in the period of days 20–30. On images A–B, an organic-mineral biofilm (bf) covering the surface (c) could be noted, associated with bacteria (b) and evidenced by U-staining (B) as a sheath around grains. Images C–D illustrate aggregates of Fe-rich secondary precipitates (p) between carrollite grains (C) and containing bacteria evidenced by U-staining (D). Images E and F show details of bacteria (b) at the surface of Fe-rich precipitates proved by U-staining between carrollite grains. Altogether, the micrographs shown in Figs. 4–6 indicate that bacterial cells with different morphology are present on the surface and inside grooves or channel-like cavities. Rod-shaped bacteria seen in Figs. 5A and 4A, C, probably related to Acidithiobacillus species, are usually met in proximity to small corrosion pits or surface micro-fissures. These finding are consistent with previous studies (Edwards and Rutenberg, 2001; Ndlovu and Monhemius, 2005) showing that A. ferrooxidans would be attached preferentially to the edges of the corrosion pits. On the other hand, bended and S-shaped bacteria, probably related to L. ferrooxidans, are spotted mostly inside corrosion pits (Fig. 5D) or along deep alteration zones having abundance of voids (Figs. 4E–F and 5E and F) as well as onto carrollite grain surfaces covered with organic–mineral precipitates (Fig. 6A). In deeply altered zones these bended bacteria increase in number while residing inside tunnels and cavities (Figs. 4E and F, 5D). The ability of L. ferrooxidans to change in shape and size during pyrite bioleaching is known as pleomorphism and has been previously interpreted in line with the different stages in bacterial life cycle by Sand et al. (1995), and demonstrated by Rojas-Chapana and Tributsch (2004). Areas of grains as corners, edges and pre-existing micro-fissures where bacteria preferentially settle and degrade mineral surface would constitute ‘‘energetically-favourable’’ sites. Also, secondary surface precipitates that form in the vicinity of bacteria and EPS biofilms (Figs. 4 and 6A), could accumulate between mineral grains to form large aggregates (Fig. 6B). Bacterial

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005

4

G. Nkulu et al. / Minerals Engineering xxx (2015) xxx–xxx

Fig. 3. BSE-SEM micrographs of initial (non-contacted) carrollite grains (A, B) and after 30 days in the bioleaching medium (C, D).

B

A b

p

*

2 µm

C

* 2 µm

D

b

*

b

*

p

*

2 µm

2 µm

E

* *

b

F

b

* b

500 nm

5 µm

Fig. 4. SE-SEM views of carrollite surface being in contact with bacteria for 5 (A), 10 (B), 15 (C), 20 (D, E) and 30 (F) days. Symbols indicate: EPS produced near bacteria or covering grains (⁄); bacteria (b); secondary precipitates (p).

cells are also detected onto mineral surface (Fig. 6C) as previously seen on SE-SEM views (Figs. 4 and 5). BSE-SEM observation revealed an ‘‘en bloc’’ U-stained organic matter in the bioleached samples. This finding suggests that a bacteria-secreted biofilm is wrapping carrollite grains (Fig. 6E), as well as that bacteria are present inside aggregates of secondary precipitates (Fig. 4E).

3.3. X-ray EDS analysis of mineral surface Table 1 reports the quantitative analysis of EDS spectra of carrollite grains as well as of secondary Fe-rich precipitates from lyophilized bulk samples and polished sections of resin-embedded samples. The analyses were performed on both intact and deeply

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005

5

G. Nkulu et al. / Minerals Engineering xxx (2015) xxx–xxx

A

B cb

rb rb

5 µm

5 µm

cb

C

D sb

rb

cb 5 µm

E

cb

p

F rb

5 µm

2 µm

rb

cb

cb

2 µm

Fig. 5. SE-SEM views of carrollite grain surface after 20–30 days bioleaching. Symbols indicate: rod-shaped (rb), bended (cb) or S-shaped (sb) bacteria; p – secondary precipitates.

altered areas of carrollite grains taken at different bioleaching times. It should be noted that the elemental atomic percentages data for intact and altered areas were pooled because they did not show any significant compositional differences. On the other hand, analysis data coming from the bulk samples and the polished sections for the iron precipitates were treated separately. The values are normalized after taking into account the Pt metal coating for the bulk samples and the C content from the embedding resin used for the polished samples. For the amount of C in the bulk samples, non-normalized data are provided. The perusal of the results shown in Table 1 indicates that the Co + Cu/S ratio is higher in the intact carrollite than in the altered one. Moreover, this ratio has dropped to 0.57 in altered grains of bulk samples. The comparison with the intact carrollite shows that carrollite degradation consists in gradual dissolution of Co and Cu, while sulphur remains in place or re-precipitates locally as elemental compounds (S4 or S8). This hypothesis is supported by the lack of S-rich altered grains in the polished slices, knowing that elemental sulphur is soluble in the organic solvents (ethanol, propylene oxide) used during the embedding procedure. It is therefore logical to assume that the degradation of carrollite in time is accompanied also by enrichment in sulphur. The EPS from bacteria probably lead to formation of Fe3+-rich sites (Fe-precipitates) having mean Fe concentration of 11.13% in contrast to 0.47% in the bulk leached carrollite. Those precipitates in turn could serve as local reactive sites for development of dissolution pits. This phenomenon is to be attributed to the L. ferrooxidans due to its capability of accumulating high concentration of EPS complexed Fe3+ (Rojas-Chapana and Tributsch, 2004). The results shown in Table 1 infer also that small differences in the concentration of O and in Co + Cu/S ratio, lower and higher respectively, are detected in the bulk samples in comparison to

those detected in the polished sections. These differences could be due to surface oxidation of the polished sections and to the lower detection efficiency for sulphur in the bulk samples. The elemental composition of the carrollite in the polished sections corresponds to a (Co, Cu)S stoichiometry, giving Co/Cu and Co + Cu/S ratios around 2 and 1 respectively. On the other hand, examination of the deeply altered areas (on carrollite bulk samples seen in Figs. 4F and 5F) reveals presence of about 27 atomic per cents of carbon, which is very likely to be an organic carbon from bacteria, bacterial remnants and exopolymers, as previously seen in the SEM micrographs. The above observations confirm the assumption that living microorganisms and/or their secretions are in close contact with the altered mineral and play a direct role in the alteration process. Similar phenomenon has been already highlighted for arsenopyrite and pyrite by Monroy and Dziurla (1994) and chalcopyrite and sphalerite by Rodriguez et al. (2003). All these studies suggested that during bioleaching, an important proportion of bacteria can be fixed to the surface of the mineral under the influence of the developed precipitates. These bacteria could be present on the surface of the mineral without necessarily having an oxidative activity. It could be likewise assumed for our system, that the Fe-rich secondary precipitates mentioned earlier which contribute to passivation of the surface could capture also bacteria and thus inhibit overall microbial activity. The EDS analysis of the precipitated phases formed on the carrollite surface, seen respectively in Fig. 6A and B and indicated as ‘‘p’’, suggest that apart from being rich in Fe and O, they contain organic C most probably coming from the microorganisms themselves or their secretions. The presence of bacteria, in association with or met inside these Fe-rich precipitates has been evidenced by the ‘‘en block’’ U-staining for BSE-SEM observation of polished

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005

6

G. Nkulu et al. / Minerals Engineering xxx (2015) xxx–xxx

b

A

bf

D

b p c

bf

bf

2 µm

5 µm

B

E

p b

p

p

2 µm

20 µm

p

C

F

b b

b

b

5 µm

5 µm

Fig. 6. SE-SEM views of bulk samples (A, B, C) and BSE-SEM views of polished thin slices (D, E, F) from ‘‘en bloc’’ U-stained resin embedded carrollite grains after 20–30 days of bioleaching. Symbols indicate: organic-mineral biofilm (bf); carrollite surface (c), bacteria (b); secondary precipitates (p).

Table 1 Quantitative data (atomic%) of elemental X-ray microanalysis of intact carrollite (before leaching), altered carrollite and Fe-rich precipitates during the bioleaching process. Intact pure carrollite Bulk samples n = 2 Mean ± SD Elements C 0 ± 0.0 O 3.3 ± 3.3 S 29.0 ± 2.6 Fe 1.6 ± 0.2 Co 43.7 ± 3.0 Cu 18.7 ± 0.7 K Ratio Co/Cu Co + Cu/S

2.34 2.18

Altered carrollite

Fe-rich precipitates

Polished sections n = 3 Mean ± SD

Bulk samples n = 4 Mean ± SD

Polished sections n = 5 Mean ± SD

Bulk samples n = 4 Mean ± SD

Polished sections n = 4 Mean ± SD

– 16.16 ± 8.84 44.72/6.88 0.21/0.36 26.23/1.20 12.68/0.48

26.95 ± 16.73 6.08/8.94 42.74/4.95 0.47/0.43 20.34/3.12 10.36/2.26

– 16.63/5.51 42.17/6.23 0.86/0.32 27.46/1.83 12.88/0.63

30.84 ± 18.98 44.28/17.56 14.67/6.07 11.13/3.72 3.22/3.53 2.01/1.26 3.13/1.76

–/– 73.30/6.07 9.84/1.61 8.09/3.58 6.24/3.74 2.53/1.49 1.94/1.06

0.14 0.27

2.07 0.88

0.02 0.11

1.58 0.57

slices - Fig. 6(D–F). Most likely the precipitates contain mixtures of different ultra-fine particulate material from altered carrollite. It could not be excluded that jarosite-type Fe-rich oxides whose precipitation has been strongly favoured by the presence of bacteria or their products form also part of these precipitates. 4. Conclusions The examination on the bio-assisted leaching of high-purity carrollite in presence of consortium consisting of three different bacterial species has led to the following findings:

0.10 0.08

2.13 0.98

0.04 0.20

1.71 0.34

1.15 0.18

2.40 0.86

0.26 0.59

(1) SEM observation has outlined the role bacteria play in carrollite bioleaching system with pitting patterns for which specific bacteria strains present in the consortium are responsible. (2) EDS X-ray analyses suggested the existence of ‘‘biomineralized’’ cells and localized ‘‘biofilm mediated’’ etching of the surface, an assumption supported by the observed ultra-fine particles and obvious organic–mineral aggregates deposited on the carrollite grain surface as well as by the enrichment in S of altered carrollite and by the presence of Fe-rich secondary precipitates.

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005

G. Nkulu et al. / Minerals Engineering xxx (2015) xxx–xxx

(3) The EPS of bacteria probably lead to the formation of Fe3+rich spots which in turn serve as local reactive sites for the development of dissolution pits. This may occur by oxidoreductive mobilisation of Co and Cu, leaving in place S0-compounds as by-products from the bacterial metabolism. This phenomenon is to be attributed to the L. ferrooxidans due to its ability to accumulate high concentration of EPS enriched in iron coming from pyrite. (4) The observed interruption of leaching after two weeks could be due to secondary precipitates, possibly of jarosite-type, forming passivation zones and also due to bacteria ‘‘captured’’ in proximity to the surface without exercising metabolitic activity. It could be concluded that carrollite dissolution is governed by cooperative mechanism involving bacteria attached to the surface and planktonic bacteria in suspension. The attached bacteria degrade the carrollite electrochemically through direct contactleaching activity liberating iron (II) in solution, which is then used as an energy source by the planktonic bacteria. Acknowledgements The authors would like to thank Gécamines for providing the carrollite sample. G.N. wishes to acknowledge the BTC (Belgian Technical Cooperation) for granting him a PhD fellowship. The authors are thankful to the Centre for Applied Technology in Microscopy, University of Liege (Catl-ULg) for providing access to ESEM and EDX equipment. References Akcil, A., Deveci, H., 2010. Mineral biotechnology of sulphides. In: Rai, M.K. (Ed.), Geomicrobiology. Science Publishers, Enfield, New Hampshire, USA, pp. 101– 137 (Chapter 4). Brierley, C.L., 1997. Mining biotechnology: research to commercial development and beyond. In: Rawlings, D.E. (Ed.), Biomining: Theory, Microbes and Industrial Process. Springer-Verlag, Berlin, pp. 3–17. Brierley, C.L., 2008. How will biomining be applied in future? Trans. Nonferr. Metals Soc. China 18, 1302–1310. Brierley, C.L., 2010. Biohydrometallurgical prospects. Hydrometallurgy 104, 324– 328. Dziurla, M.A., Monroy, M., Domestre, A., Barrès, O., Berthelin, J., 1997. In: Hoberg, H., Von Blottnitz, H., (Eds.), Proceedings of the XX International Mineral Processing. Germany, Aachen, pp. 52–60. Edwards, K.J., Rutenberg, A.D., 2001. Microbial response to surface microtopography: the role of metabolism in localized mineral dissolution. Chem. Geol. 180, 19–32. Ehrlich, H.L., 2004. Beginnings of rational bioleaching and highlights in the development of biohydrometallurgy: a brief history. Eur. J. Miner. Process. Environ. Prot., 102–112. Gericke, M., Neale, J.W., Van-Staden, P.J., 2009. A Mintek perspective of the past 25 years in minerals bioleaching. J. SAIMM 109, 567–585. Gómez, E., Bla˘zquez, M.L., Ballester, A., Gonza˘lez, F., 1996. Study by SEM and EDS of chalcopyrite bioleaching using a new thermophilic bacteria. Miner. Eng. 9, 985– 999.

7

Habashi, F., 1997. Handbook of Extractive Metallurgy, John Wiley & Sons. Jiang, L., Zhou, H., Peng, X., Ding, Z., 2009. The use of microscopy techniques to analyze microbial biofilm of the bio-oxidized chalcopyrite surface. Miner. Eng. 22, 37–42. Jordan, M.A., 1993. The oxidation of base metal sulphides and pyrite gold concentrations with particular reference to mechanism and preferential release of ferrous iron. Ph.D. Thesis. Camborne School of Mines, University of Exeter. Karavaiko, G.I., Rossi, G., Agate, A., Avakyan, Z., 1988. Biogeotechnology of Metals. Manual. GKNT International Projects, Moscow. Krebs, W., Brombacher, C., Bosshard, P.P., Bachofen, R., Brandl, H., 1997. Microbial recovery of metals from solids. FEMS Microbiol. Rev. 20, 605–617. Liu, H., Gu, G., Xu, Y., 2011. Surface properties of pyrite in the course of bioleaching by pure culture of Acidithiobacillus ferrooxidans and a mixed culture of Acidithiobacillus ferrooxidans and Acidithiobacillus thiooxidans. Hydrometallurgy 108, 143–148. McGuire, M.M., Edwards, K.J., Banfield, J.F., Hamers, R.J., 2001. Kinetics, surface chemistry and structural evolution of microbially mediated sulfide mineral dissolution. Geochim. Cosmochim. Acta 65, 1243–1258. Monroy, G., Dziurla, M.A., 1994. A laboratory study on the behavior of Thiobacillus ferrooxidans during pyrite bioleaching in percolation columns. Adv. Bioprocess Eng., 509–517. Mustin, C., de Donato, P., Berthelin, J., Marion, P., 1993. Surface sulphur as promoting agent of pyrite leaching by Thiobacillus ferrooxidans. FEMS Microbiol. Rev. 11, 71–78. Ndlovu, S., Monhemius, A., 2005. The influence of crystal orientation on the bacterial dissolution of pyrite. Hydrometallurgy 78 (3–4), 147–286. Rawlings, D.E., 2002. Heavy metal mining using microbes. Annu. Rev. Microbiol. 56, 65–91. Riekkola-Vanhanen, M., 2012. Talvivaara mining company – from project to a mine, Paper Presented at Biohydrometallurgy’12, Falmouth UK, 18–20 June 2012. Rodriguez, Y., Ballester, A., Blasquez, M., Gonzalez, F., Munoz, J., 2003. Study of bacterial attachment during the bioleaching of pyrite, chalcopyrite, and sphalerite. Geomicrobiol. J. 20, 131–141. Rojas-Chapana, J.A., Tributsch, H., 2004. Interfacial activity and leaching patterns of Leptospirillum ferrooxidans on pyrite. FEMS Microbiol. Ecol. 47, 19–29. Rojas-Chapana, J., Giersig, M., Tributsch, H., 1995. Sulfur colloids as temporary energy reservoirs for Thiobacillus ferrooxidans during pyrite oxidation. Arch. Microbiol. 163, 256–352. Rojas-Chapana, J.A., Giersig, M., Tributsch, H., 1996. The path of sulfur during the bio-oxidation of pyrite by Thiobacillus ferrooxidans. Fuel 75, 923–930. Sampson, M.I., Phillips, C.V., Blake, R.C., 2000. Influence of the attachment of acidophilic bacteria during the oxidation of mineral sulphides. Miner. Eng. 13, 373–389. Sand, W., Gehrke, T., Hallmann, R., Schippers, A., 1995. Sulfur chemistry, biofilm, and the (in)direct attack mechanism – a critical evaluation of bacterial leaching. Appl. Microbiol. Biotechnol. 43, 961–966. Schippers, A., Sand, W., 1999. Bacterial leaching of metal sulfide proceeds by two indirect mechanisms via thiosulfate or via polysulfides and sulfur. Appl. Environ. Microbiol. 65, 319–321. Schippers, A., Jozsa, P.G., Sand, W., 1996. Sulphur chemistry in bacterial leaching of pyrite. Appl. Environ. Microbiol. 62, 3424–3431. Silverman, M., 1967. Mechanism of bacterial pyrite oxidation. J. Bacteriol. 94, 1046– 1051. Silverman, M.P., Lundgren, D.G., 1959. Studies on the chemoautotrophic iron bacterium Ferrobacillus ferrooxidans. I. An improved medium and a harvesting procedure for securing high cell yields. J. Bacteriol. 77, 642–647. Tributsch, H., Rojas-Chapana, J.A., Bärtels, C.C., Ennaoui, A., Hofmann, W., 1998. Role of transient iron sulfide films in microbial corrosion of steel. Corrosion 54, 216– 227. Yager, T.R., 2010. The mineral industry of Congo (Kinshasa), 2007 Minerals Yearbook, US Department of the Interior, US Geological Survey: Reston, VA, March 2010, 11.3.

Please cite this article in press as: Nkulu, G., et al. SEM and EDS observations of carrollite bioleaching with a mixed culture of acidophilic bacteria. Miner. Eng. (2015), http://dx.doi.org/10.1016/j.mineng.2014.12.005