Kalin/Wheeler, Morgantown – Taskforce, 03-08-2011 · PDF file2016/01/11 · Kalin/Wheeler, Morgantown – Taskforce, 03-08-2011 1 A Review of the Role of Phosphate Mining...

Kalin/Wheeler, Morgantown – Taskforce, 03-08-2011

1

A Review of the Role of Phosphate Mining Waste: A Chemical or Biological Reagent for AMD Prevention?

Margarete Kalin and W.N. Wheeler

Boojum Research Ltd Telephone (41‐33‐971‐1434)

Email: [email protected]) 1.0 Introduction As the title suggests, phosphate mining wastes are reported to have a chemical role in reducing the oxidation of sulphidic wastes and improving the effluent (acid rock‐ or acid mine‐ drainage). In the nineties, extensive laboratory and field tests were carried out using various phosphate sources in sulphidic mining wastes with positive results. Reductions in acidity, mainly contributed from iron, aluminum and other metals of up to 90 % [1‐12] were reported. The effects on oxidation were attributed to a chemical precipitation of oxidized iron with phosphate, leading to a coating over the sulphides, reducing oxygen access to the mineral surface. Explanations, based purely on chemistry, omit the fact that the oxidation rate is governed, after initial oxidation and acidification, by microbial actions on the mineral surface. Thiobacillus ferrooxidans (Acidithiobacillus) is one of the microbes accelerating pyrite oxidation. It has been used extensively in bioleaching. However, when the chemical conditions on the mineral surface change, the very same microbe is able to switch growth modes, from chemolithotroph (requiring oxygen and electrons from the pyrite), to heterotroph (requiring inorganic carbon and anaerobic conditions)[13]. This capacity to adapt to environmental conditions is widespread in the microbial world. This paper re‐interprets selected field and laboratory experiments, including the microbiological perspective and centered on processes taking place on the mineral surface. 1.1 Phosphate as a chemical reactant Phosphate has been added to sulphidic mining wastes in lab and field experiments using either phosphate fertilizer (orthophosphate) in liquid form or phosphate rock (apatite) in granular form. Phosphate rock occurs in different mineralogies but generally contains both apatite and calcium carbonate. Baker [7] concluded that the addition of apatite to coal plant wastes gave best results with an application rate five times less by weight than the amount of limestone needed. Meek [8] stated that, regardless of the acid‐producing potential of the wastes, an addition of 3% of apatite by weight was effective in achieving a satisfactory effluent. Renton [9 and 10] recommended that application rates of less than 1% by weight were ineffective but 5 % produced a greater than 90 % reduction in generated acid.


2

Based on laboratory and field work, it was postulated by Evangelou [1} and in 2011 again by Mauric and Lottermoser [14] that the acidity reduction was the result of a metal‐phosphate coating over the sulphides. This coating was well‐documented on sulphide minerals, but, to our knowledge, has never been corroborated in the field tests. In the laboratory the pyritic wastes were treated with peroxide to speed up oxidation of the waste (pers. comm. V.P. Evangelou). The phosphate was added as a water‐soluble solution to the wastes in columns or funnels. In the field, where these coatings were not found, phosphate (acid‐soluble and neutralizing form) was added as gravel of different grain sizes. These laboratory conditions create a very different pyrite mineral surface, influencing the native microbial activity, and producing results which would probably differ with field‐generated results. From a practical perspective, achieving an efficacious mixture of phosphate and sulphidic wastes is not an easy chore. Due to the vast volumes of mining wastes it is also economically unattractive. One of the most glaring differences between using phosphate rock and phosphate fertilizer is its solubility. Phosphate rock does not dissolve above pH 4.5. Thus, acid generation has to take place before any phosphate rock can release reactants. Calcium carbonate in the rocks would tend to buffer any incipient acidity, slowing the release of phosphate. Microbes colonize the pyrite surface rapidly upon acidification. Thus the phosphate rock addition has a great potential to change the surface conditions and act as highlighted by Meek [8] as an in‐situ point source control of AMD. The phosphate/AMD literature generally shows no stoichiometric relationship between ferric iron and phosphate application. Further, no relationship could be found between the dosage of added phosphate and the improvement of the drainage. Lower phosphate application rates produced results as good or better than higher additions. Thus, very different recom‐mendations emerged about how much phosphate material is needed to achieve a significant reduction in acid generated[1‐14]. From our perspective, these results and observations, together with our discussions with other authors, lead us on one conclusion, that is, that the mode of action for phosphate is not chemical alone but also biological. Our confidence increased when a report in a monthly trade journal described a copper leach dump at the Gibraltar Mine in British Columbia which no longer leached copper. After some time, acidic drainage, which was cycled over the dump, gradually appeared as neutral drainage with low metal content, clearly to the distress of the shareholders [15]. We inquired with the company if they had followed up on the cause of the cessation of acid generation from the dump. Indeed, sections of the dump had been taken apart, finding rocks covered with a solid coating of iron‐hydroxide. They sent us some rocks from these sections. We examined the coatings with SEM – EDX identifying very few locations with a phosphate signal. For the first time a link between the inhibition of acid generation and iron precipitates on rocks from the dump could be forged. It appeared as if phosphate were acting as catalyst rather than a chemical participant. Since phosphate minerals were present in the mineralogy of the Gibraltar Mine, we should have found traces of phosphate in the coating.


3

1.2 A different perspective: Phosphate mining wastes as rock fertilizers Phosphorus is an essential element to all life forms. Phosphate in rocks as mineral or in ore, in soils and sediments is present as an inorganic orthophosphate. This form of phosphate, however, cannot be utilized well by organisms. A conversion is required to make it water‐ soluble and available to an organism’s metabolism. This conversion occurs through geochemical and biochemical reactions which take place at many stages of the global phosphorus cycle. Microbes can catalyze the conversion and are also major players in the oxidation of all rocks or the weathering of minerals [16, 17 and 18]. Natural phosphate rock (NPR) is mined from phosphatic, sedimentary rock deposits. Those deposits are commonly accompanied by or inter‐bedded with shales, cherts, limestone, dolomites and sometimes sandstone [19]. The NPR from such deposits releases not only orthophosphate, but also organic phosphate and calcium‐carbonate from the fossils and shells. Such deposits have been used a fertilizer for acidic soils .We will use the term NPR here to represent the phosphate mining waste from the Aurora Mine in North Carolina, U.S.A. We suspect that such NPR provides neutralization through the calcium carbonate dissolution and possibly the release of bio‐available phosphate to the microbes on the mineral surface. Even a release of a very small amount of phosphate would provide nutrition to surface‐bound heterotrophs, which, in turn, would outcompete Thiobacillus on the oxidation pit sites. It is also possible, although not yet tested, that the phosphate may also induce a change in the growth mode of some Thiobacillus, from chemolithotroph to heterotroph, further outcompeting the chemolithotrophs. In the laboratory, a phosphate solution acts chemically, altering surface charges on the pyrite due to its buffering capacity. When added in this form in a field test (e.g. large‐scale), at best, it would buffer some of the acidity before it was used up by microbial, fungal and algal populations on the rocks, with the remainder being washed out of the waste dumps. We postulate that, if microbial metabolism is the rate‐limiting factor for the release of phosphate within sulphidic mining wastes, then both high and low application rates would produce similar results. Microbes grow at a given rate, no matter how much fertilizer is present, which would result in similar amounts being released. Secondly, and also consistent with the same effect of high and low additions of phosphate, is the fact, that not all portions of the wastes produce AMD or ARD. The effluent is only produced in that portion of the wastes where rain water runs over the oxidized pyrite surface, dissolving the weathering products, e.g. in a defined seepage path within and through the waste deposits. It is therefore inappropriate to calculate phosphate application based on the pyrite content of the wastes. Calculations should rather be based on how much seepage is produced from the waste pile and that is also affected by the climate. Further, it is essential to consider how the reactant could be transported to the surface of the oxidizing minerals in the wastes, and which application method would ensure that particles would be carried with the rain water to the mineral surfaces within the wastes. This would facilitate the expected in‐situ control of


4

oxidation. In support of this suggestions, Renton observed “that using particles larger than 1.6 mm were ineffective…. but a fine grained apatite ‘spike’ resulted in substantial reduction of acid generation“ [9 and 10]. The published work on phosphate as an ameliorant to acid mine drainage far exceeds the quoted work in this paper but we have surveyed the scientific literature regularly. Most of the work is carried out with the same premise and in the laboratory for relatively short periods of time. Thus, the conclusions are generally consistent with the idea that amelioration is due to the chemical iron phosphate precipitation/coating process, but no definite proof is provided to date. Also missing has been a testable hypothesis of how the additions influence the oxidation of the sulphides, as the coating and iron phosphate precipitate process alone cannot adequately and consistently support the results. Without knowing the controlling factors leading to the reduction of sulphide oxidation, it is not possible to control it or to scale it up. 2.0 Integrating experimental design to counteract economic and technical hurdles A series of long‐term field tests was set up focusing on overcoming the application hurdles and economic objections. Our premise was that we had to work in the field, as natural conditions cannot be simulated adequately in the laboratory. Further, we felt that the industry needed a solution to acid mine drainage which has been proven under field conditions. The application methods had to be chosen such that a mining company could easily integrate them into their operating procedures. Finally, the experiments had to focus on economic and sustainable solutions and this meant: To be economic:

• The application should be one time only addition. • The test plots must remain in the field for at least one year before samples are retrieved

from the field to evaluate the effect. • The phosphate‐containing material needs to be acid‐soluble and cost‐effective. • The set up of the plots in the field or outdoors use techniques usable by industry.

To be sustainable : • After field exposure with the NPR, the field wastes need to be subjected to acid‐

generating conditions, and those need to be monitored for a least year. • The process of inhibition should be as universal as is the oxidation process. Inhibition

process should work in sulphidic mining wastes with widely differing reactivity’s. • The application mode should address the location in the wastes where oxidation takes

place. This requires a different mode of application for tailings (oxidation takes place in the vadose zone) and waste rock (the mineral surface in the seepage path).

Evaluation criteria of the effect of the additions consisted of:

• Use comparable conditions and methodology to evaluate all effluents. • Subject material from field/outdoor tests to detailed surface investigations.


5

• Provide proof of process controls based on field samples.

2.1 Growing vegetation and reducing oxygen ingress into the tailings surface When bare tailings surfaces are investigated at depth, it is evident that the oxidation of the tailings occurs in thin fine layers, which are interspaced by un‐oxidized tailings, as for example in Figure 1. Evaporation during the warm season leads to upward movement of metal laden pore‐water and forms encrustments. These two physically controlled processes can not be changed and are essentially the main cause of the metal‐laden acid seepages. The encrustments consist of water soluble acidic evaporates. These penetrate into the tailings with spring and fall rains carrying the internal oxidation products with them. They pass slowly through the tailings mass, precipitating and re‐dissolving more oxidation products on the way, eventually emerging as AMD or ARD schematically presented in Figure 2 for a waste rock pile. We started to work with fresh tailings to address the inhibition of the weathering shortly after discharge of tailings, when they have just left the mill. We envisage seeding tailings after the depository is completely filled and about to be decommissioned. Effective, in‐situ control of oxidation must attempt to immediately prevent the oxidation of the very top surface. This could be accomplished by removing the oxygen, if we introduce a microbial community within or below the root zone of the needed grass cover on the surface. These heterotrophs (consuming oxygen), if present in high enough concentrations, would consume the oxygen. A layer of NPR below this would lead to precipitation of the metal‐laden porewater, gradually forming a hard pan below the root zone. Ideally, with time ultimately clean water would run off the tailings pile.

Figure 1: split tailings core at South Bay Mines. Note the uneven oxidation due to irregular channeling

through the tailings. Figure 2: Schematic presentation of seepage formation in a waste rock pile

Initially, we added the seeds directly into the discharge spigot distribution with the tailings, but this was unsatisfactory. The seeds only germinated in drying cracks but did not cover the entire surface (Figure 3a). However, rolling out seed with slow‐release fertilizer in a straw type matting onto the fresh tailings proved useful (Figure 3b).


6

Figure 3: a) grass seeds growing in cracked, fresh tailings. b) test plot on fresh tailings using a straw

matting to hold in moisture. As oxidation starts rapidly in tailings, we needed to evaluate of the vegetation cover and especially the root structure after the first winter. A living grass cover directly in the tailings will produce oxygen in the root zone and support oxidation. Further, roots will loosen the tailings and add oxygen. However, roots will also add organic matter, which would support heterotrophs. So, we needed to stimulate oxygen consumption at the same time we seeded. To add heterotrophs, we first needed to quantify microbial activity. The details of these experiments are given in [22]. The field trials demonstrated conclusively that a grass cover can be established on fresh tailings and form a continuous organic layer on the tailings surface. This vegetation grew after the first winter and biomass at the end of the second year produced a standing crop of 10 t per hectare, producing ample biomass to support the heterotrophic decomposition. A vegetation cover was also established on old acidic tailings with the addition of horse manure (Figure 4).

Figure 4: Vegetation cover with one time application of NPR and horse manure on pH 2.5 abandoned uranium tailings.


7

Although not all of the pre‐field test led to the desired results, they were encouraging enough to set up field test plots on uranium, pyhrrotite, polymetallic tailings between 1992 and 1996. Electromagnetic surveys were used to detect a hard pan below the root zone in the tailings. This was only partially successful, due to the extremely high conductivity of the pyhrrotite tailings. The use of comparable methodologies to evaluate effluents and pore‐water was mandatory. We sampled the test plots in the field from pits dug to about 0.5 m depth and collected tailings above, in and below the NPR layer, which generally was found about 15 cm below the shallow root zone of the vegetation. The samples were stored in plastic bags for various periods of time before slurries were prepared. The slurries were made from 1 part tailings and 5 parts water. The supernatant was removed for testing and fresh water added back to the slurry. After stirring, the beakers remained at room temperature, lightly covered to prevent evaporation. These conditions should have enhanced oxidation.

The samples all should have generated acid. However, they did not. Nor did we find any phosphate, even after analysis by ICP [22]. The amount of NPR added to the tailings was relatively small and thus diluted by the tailings to levels below the analytical detection limit, which is relatively high for P in the iron‐rich mining wastes. Phosphate could be determined in the polymetallic concentrate spill material, as here the mixes were made in open bottom buckets, and those were sampled and analysed [23].

2.2 Using NPR in coarse coal wastes and waste rock piles

Tests were also set up in coarse coal wastes [unpublished]. In the same year, an outdoor experiment with about two tonnes of sulphidic waste rock to which NPR was added was monitored for 2.7 years starting in 1992. After the first winter, the effluents from the drums with NPR produced neutral drainage in contrast to the drum without NPR [21].

The waste rock, after the 2.7 years of outdoor exposure, was stored in an industrial basement for an additional 4. 5 years. Thereafter, the rocks were re‐exposed outdoors without NPR. The water collecting in the drums was monitored for a year. The drums were then dismantled and stored for 8 years in covered buckets outdoors, again experiencing rain, heat and freezing. In 2010 further analysis of the stored rocks was carried out at the University of Toronto. The coating had persisted, but changed in appearance [26]. Only traces of the biofilms could be found on the waste rock in all SEM EDX searches on many different rocks. It should be noted, that the coatings cannot be seen by the naked eye or under low magnification, so rocks could not be pre‐selected for investigation.

The mineral surfaces of these rocks were investigated after several years of storage at the University of Ottawa [24]. These investigations concluded that a coating exists over the sulphidic minerals. It consisted of inorganic precipitates embedded within an organic material, referred to as EPS (Extracellular Polymeric Substances) a biofilm. In 2004, the same coating on stored rocks from the same experiment, was confirmed by an independent group of


8

researchers, as part of a project addressing passive treatment methods for sulphidic mining wastes at the University of South Australia [25].

Figure 5: a) (Left) Note size of test piles (scale red truck); (right) Hardpan within the pile, with self‐sealing erosion channels; b) Schematic application methods used in test piles, replace limestone with NPR. Unfortunately estimates of oxidation rates in coal wastes could not be included, as our experiments in a field coal pile (Figure 5a, b) and with coal columns in the laboratory have not been completely analysed. The coarse coal test pile in Cape Breton ceased to deliver acid seepage or any kind of seepage at all, which is what would be expected from the formation of a hardpan within the pile. An autopsy was commissioned which identified a self‐sealing hardpan in an erosion channel (Figure 5a,b). At the same time, columns were set up in the laboratory to parallel the field tests. In the first year, the results of the column tests were promising as drastic acidity reductions were measured in some (Figure 5c; arrow) but not all columns with NPR. The PERD columns were run for up to 5 years (Figure 5d). The effluents collected between 1993 the first year and intermittently up to 1998 at the time when the columns were taken apart. The effluents were measured for pH, redox and electrical conductivity, and re‐measured after storage for up to 7 years in 2000. The colour alone had changed over the course of monitoring (either over time or in storage; Figure 5e). Reasons for this have not been investigated or related to the chemical analyses.


9

Figure 5 cont. c) coarse waste rock columns. Acidity generated in columns with the the NPR =LHS addtions in the first year columns watered in the laboratory

Figure 5 cont. d) PERD columns in the laboratory. e) effluent from coarse coal waste with an NPR layer in year 1993 left clear to 1998 right last effluent collected prior to dismantling of column

3.0 Conclusions and the way forward Our current working hypothesis is that NPR additions not only play a chemical role, but act as a fertilizer, enhancing the growth and competition of heterotrophs in biofilms. The chemolithotrophs are outcompeted, slowing oxidation. Our EM findings corroborate the presence of microbes and biofilms, as indicated by the EPS coating with its enclosures of inorganic precipitates. These findings have also been verified by two independent labs. Further, there is now extensive recent scientific literature on the formation of biofilms on minerals which strongly supports our conjectures of the role of microbes and their inhibitory effects. The evidence that the inhibition process is quite universal comes from estimates of oxidation rates (i.e. rates generated from pure pyrite and under very different conditions, covering 1–90 % sulphide)[27]. Estimates of sulphide oxidation rates (log r) were derived for all field and outdoor tests with NPR using the Eh and pH values of the effluent from the waste rock drums and pore‐water slurries. The log r values of the controls with no NPR were only slightly lower


10

than the laboratory standard rates. In the presence of NPR they were 1 to 3 orders of magnitude lower than the controls. Only for pyhrrotite were the rates similar, but this was expected from very reactive 90 % sulphide wastes (which burn when it rains) [27]. Another piece of evidence for the universality of the inhibition process is provided by the presence and longevity of the EPS coatings which did not deteriorate under conditions which should have generated acid. NPR appears to act primarily as fertilizer for biogeochemical microbial activity on the mineral surface. This hypothesis explains most of the previous work and has paved the way for future work. A definite proof of process control can’t yet be provided. What we have achieved with our field work and follow up investigations is to provide indirect evidence that supports an alternative hypothesis for the action of phosphate. Not only does this hypothesis explain the current work and literature, but it opens the door to large‐scale testing on waste rock and tailings. Minute amounts of phosphate appear to regulate the oxidation process. Using waste phosphate appears to provide even better results than more commercial phosphate compounds. We feel that this technology is ready for large‐scale testing. Would it not be worthwhile giving microbes half a chance to work with us, instead of continuing to generating acid? 4.0 Literature Cited 1. Evangelou VP, 1995. Pyrite oxidation and its control. CRC Press, New York, 285 pp. 2. Spotts E, Dollhopf DJ, 1990. Evaluation of phosphate sources for control of AMD in coal

overburden. Reclamation Research Unit,‐Montana State University,. p 1‐69. 3. Ziemkiewicz PF, 1990. Advances in the prediction and control of AMD. In: Proceedings of the

1990 Mining and Reclamation Conference and Exhibition, Vol. 1, Charleston W. Virginia, p 51‐54.

4. Meek AF, 1991. Assessment of acid prevention techniques employed at the Island Creek

Mining. In: Twelfth Annual West Virginia Surface Mine Drainage Task Force Symposium, April 1991, pp 1‐9.

5. Hart W, Stiller AH, Rymer TE, Renton JJ, 1990. The use of phosphate refuse as a potential

acid mine drainage ameliorant. Proceedings, 1990 Mining and Reclamation Conference and Exhibition, April, 1990, Charleston, WV.

6. Hart W, Stiller AH, 1991. Application of phosphate refuse to coal mine refuse for

amelioration of acid mine drainage. Proceedings, Second International Conference on Abatement of Acidic Drainage, Montreal, Canada.


11

7. Baker BK, 1983. The evaluation of unique acid mine drainage abatement techniques. Master’s thesis , College of Engineering. West Virginia University, Morgantown. WV. 159 p

8. Meek FA, 1983. Research into the use of apatite rock for acidic drainage prevention. In

Proceedings of the Fifth Annual West Virginia Surface Mine Drainage Task Force Symposium, Morgantown WV. http://www.wvmdtaskforce.com/proceedings/84/mee/84mee.htm

9. Renton JJ, Stiller AH, 1988a. Use of phosphate materials as ameliorants for acid mine

drainage. Volume 1. The use of rock phosphate (apatite) for the amelioration of acid mine drainage from the mining of coal. Final report #PB‐93‐216448/XAB.

10. Renton JJ, Stiller A.H., Rhymer T, 1988b. The use of phosphate materials as ameliorants for

acid mine drainage. Mine Drain. Surf. Mine Reclamat., Mine Water Mine Waste; U. S. Bureau of Mines Information Circular 9183, vol 1:67‐75.

11. Gorgopoulou ZJ, Fytas K, Soto H, Evangelou B, 1996. Feasibility and cost of creating an iron‐

phosphate coating on pyrrhotite to prevent oxidation. Environmental Geology 28:61‐69. 12. Belzile N, Maki S, Chen Y, Goldsack D, 1997. Inhibition of pyrite oxidation by surface

treatment. Science Total Environment. 196(2):177‐186.

13. Tabita R, Lundgren DG, 1971. Heterotrophic metabolism of the chemolithotroph Thiobacillus ferrooxidans. J Bacteriology 108(1): 334‐342.

14. Mauric A, Lottermoser BG, 2011: Phosphate amendment of metalliferous waste rocks,

Century Pb‐Zn mine Australia: Laboratory and field trials. Applied Geochemistry 26: 45‐56.

15. Scott D, 1991 Gibraltar's dump leaching‐an insight into acid effluent control. The Northern

Miner, September. 16. Ruttenberg KC, 2004. The Global phosphorcycle. Chapter 8.13 In: Schlessinger WH: (ed.)

Biogeochemistry, Elsevier Treatise on Geochemistry. p 585 ‐633. 17. Jansson M, Olsson H, Pettersson K, 1988. Phosphatases; origin, characteristics and function

in lakes. Hydrobiologia. 170(1):157–175. 18. Tiessen H, 1995. Phosphorus cycles and transfers in the global environment. SCOPE 54;

Scientific Committee on the Problems of the Environment , UNEP John Wiley & Sons pp 459

19. Phosphorite: http://en.wikipedia.org/wiki/Phosphorite#Phosphorus_Cycle.2C_Formation_and_Accumulation


12

20. Kalin M, Fyson A, Smith MP, 1993. Heterotrophic bacteria and grass covers on fresh, base metal tailings. In: Proceedings of the Canadian Land Reclamation Conference 18th Annual Meeting, Lindsay, Ontario, August 11‐13, p. 81‐88.

21 Kalin M, Harris B, 2005. Chemical precipitation within pyritic waste rock. Hydrometallurgy,

78(3‐4):213‐229.

22. Kalin M, Fyson A, Smith MP, Werker A, 2003. Tailings surface cover development through integration of reactive phosphate and organic matter. Proceedings of the Mining and Environment III and Annual Meeting of CLRA (CD), Sudbury, Ontario, May 25‐28. PDF #80,10 p.

23. Kalin M, 2004a. Improving pore water quality in reactive tailings with phosphate mining

wastes.. Proceedings of the Fifth International Symposium on Waste Processing and Recycling in Mineral and Metallurgical Industries” and the 43rd Annual Conference of Metallurgists of CIM, Hamilton, Ontario, August 22‐25 p. 427‐437

24. Kalin M, Ferris G, Paulo C, 2009. Reducing sulphide oxidation in pyritic mining wastes – phosphate mining wastes stimulate biofilm formation on mineral surface. In: Proceedings of ICARD “Securing the Future/8th”, Skellefteå, Sweden. http://www.securing.skelleftea.se/securingkonf2009.

25. Ueshima M, Fortin D, Kalin M, 2004. Development of iron‐phosphate biofilms on pyritic

mine waste rock surfaces previously treated with natural phosphate rocks”, Geomicrobiology Journal (ISEB 16 presentation), 21(5):313‐323.

26. AMIRA, 2009: Alternate treatment for ARD control: http://www.amira.com.au/WEB/site.asp?section=projects&page=projectdetails&ProjectLink=2861&Source_ID=1

27. Kalin M. Paulo C, Sleep B, 2010. Proactive Control of Acid generation: Reduction/ inhibition of sulphide Oxidation , In Mine Water and Innovative Thinking. IMWA, September 5‐9th Sydney Nova Scotia p 479‐482.

Kalin/Wheeler, Morgantown – Taskforce, 03-08-2011 · PDF file2016/01/11 · Kalin/Wheeler, Morgantown – Taskforce, 03-08-2011 1 A Review of the Role of Phosphate Mining...

Documents