CA3043821C - Remediation and stabilization of thixotropic and/ or colloidal mines tailings using pulverized anhydrite (naturally occurring anhydrous calcium sulphate (caso4)) - Google Patents

Abstract

Metallurgical Rehabilitation of Oil Sands Mines Tailings. Pulverised, locally abundant, natural anhydrite (anhydrous calcium sulphate (CaSO4)), renders colloidal Mature Fine Tailings ('MFT'),"grandfathered" from mining of the Alberta Oil Sands, sufficiently solid to be load bearing. The reactions take place at 20°C, slightly alkaline pH, and 101.3 kilopascals atmospheric pressure. One prior successful patent used quicklime, others used gypsum (hydrated CaSO4.2H2O) and calcined gypsum (CaSO4.1/2H2O) as a flocculant. Various concentrations, from 0.5% to 12.0% of anhydrite, and quicklime (CaO) for comparative purposes, were added to 50gm aliquots of MFT, resulting in stiffening to a firm paste within minutes to days. All reaction times were markédly more rapid after prior drying the MFT for 96 hours; reagent concentrations required were also lower. Mineralogical composition of the reagents and the products were all tested with X-Ray Diffraction. The reactions are irreversible, yielding some free fluid and a hard, load-bearing, substrate.

Description

Application Number 3,043,821 REMEDIATION AND STABILIZATION OF THIXOTROPIC AND/
OR COLLOIDAL MINES TAILINGS USING PULVERIZED
ANHYDRITE (NATURALLY OCCURRING ANHYDROUS
CALCIUM SULPHATE (CaSO4)) DESCRIPTION
Technical Field:
Metallurgical Remediation, Stabilization and Rehabilitation of Thixotropic and/ or Colloidal Oil Sands Mines Tailings.
Background:
The extent of the problem here addressed was obtained from documentation provided by Canada's Oil Sands Innovation Alliance (COSIA), most particularly their "Tailings Clay Challenge"
(Anon, 2017) and "Soft Tailings Capping Technology Challenge" (Anon, 2017),.
The problem isn't solved or addressed according to present knowledge. The tailings merely accumulate in ponds, capped by water. As of 2014, the tailings ponds contained 109 cubic metres of Mature Fine Tailings ('MFT') covering more than 130 square kilometres in the oil sands region, north of Fort McMurray, Alberta. There are several other technologies being tested and used, but none seem to be wholly satisfactory. Unless the colloid can be broken, the tailings are not load bearing, so they have historically been sequestered in ponds, which cannot be rehabilitated. The large volume of MFT requiring safe containment and the vigilant management of capping waters represent not only a significant management challenge but also a liability for the industry.
The extraction of bitumen from Oil Sands of Northern Alberta and Saskatchewan using hot or warm water processes produces a slurry waste that is hydraulically transported and stored within surface tailings ponds. The fast-settling sand particles segregate from the slurry upon deposition at the edge of the tailings ponds while the fine fraction, comprising mineral particles smaller than 44 i.tm, and clay particles of the order of 2 m, accumulates in the center of the pond, and settles to Page! of 13 Application Number 3,043,821 become MFT. Although most of the water is released and recycled back into the process, up to 86%
by volume of MFT consists of water. MFT only settle to about 30% to 35% solids content after a few years of placement. The MFT are a colloid, which can be broken, but to-date a solution that is both logistical and economical has been elusive.
With one single exception, there appears to have been no attempts to address the problem using simple, locally available natural products. The BGC Engineering (2010) produced a comprehensive review of the state of the art, which proved an excellent start point for appreciating the myriad approaches to a solution that have been, and are currently being, pursued. Liu et al., (2016) successfully achieved an enhanced rate of sedimentation and settling using gypsum as a flocculant. However, despite this success, there appears to be some reluctance by industry to use this technology because the addition of the calcium ion in the gypsum would interfere with the effect of sodium on bitumen extraction when water is recycled. Wang's (op cit) work concentrated on the use of Portland Cement and an artificial sodium silicate polymer (NS) and aluminosilicate polymer (AAAS) on the MFT. Wang is not alone in this field, quoting several other researchers following similar paths. In both these studies, the calcium appears to be instrumental in tackling the colloidal suspension of the clays. See the list of References and Bibliography below.
A patent search revealed the following patent registrations and patent applications that relate, even remotely, to the process here described. The only potential contenders for prior art are CA
2143396, CA 2183380, CA 2921835 and CA 2522031; but none tackle the problem in the simplistic way here advocated. Patents CA 2143396 and CA 2183380 both claim uses as a flocculant of calcium sulphate hemihydrate (CaSO4.1/2H20) in very low concentrations for treatment of mine tailings; whereas the present study claims the use of anhydrite (naturally occurring anhydrous Page 2 of 13 Application Number 3,043,821 calcium sulphate (CaSO4)) as a reagent/ additive in concentrations of 6000 ppm and greater. Patent CA 2921835 claims a process for the treatment of MFT using gypsum inter alia.
CA 2143396 describes the use of calcium sulphate hemihydrate essentially as a flocculant, where the concentration of calcium sulphate is in the 125 to 175 ppm range, to accelerate fines settlement from desanded aqueous tailings from mining the oil sands; and as a detoxifying agent, to precipitate naphthenates and sulfonates.
Patent CA 2183380 is a use of gypsum (naturally occurring hydrated calcium sulphate (CaSO4.21120)), and calcium sulphate in the form of the hemihydrate (CaSO4.1/2H20) as a flocculant in very low concentrations (100-200 ppm) for aqueous tailings containing coarse sand and clay fines and to remove naphthenates and sulfonates from these aqueous tailings produced from oil sands.
Patent CA 2522031 uses quicklime (CaO) or slaked lime (Ca(OH)2) as a reagent/
additive to fine tailings in process, to thicken them prior to remixing them with sand tailings;
this use of quicklime (CaO) was used as a basis for comparison in the present study.
Patent CA 2921835 is a process for the treatment of mature fme tailings (MFT) derived from oil sands extraction that include contaminants comprising bitumen, naphthenic acid, arsenic, selenium and suspended solids, by using an immobilization chemical that comprises or consists of gypsum (Claim #11).
CA 2927082 and JP 2010207659 are concerned with the use of gypsum (CaSO4=2H20), calcined gypsum (CaSO4=1/2H20), and anhydrite (CaSO4) in precipitating and solidifying industrial and construction sludge and wastewater for the containment of hazardous substances present in such effluent.
Page 3 of 13 Application Number 3,043,821 It should be noted that whereas other patents refer to the chemical compound "calcium sulphate" and shows the symbol "CaSO4", only the terms "gypsum", "calcined gypsum" and "calcium sulphate hemihydrate" are used. Clearly, there is no ambiguity here:
only the mineral species gypsum mentioned is intended by those patents.
It should be further noted that anhydrite (CaSO4) and gypsum (CaSO4.21120) are very distinctly different mineral species. Anhydrite crystals are orthorhombic, with a hardness of 3.0 to 3.5, specific gravity of 2.90 to 2.98; whereas gypsum crystals are monoclinic, hardness 1.5 to 2.0, and specific gravity of 2.3.
Broad Description Of The Invention Mix thixotropic and/ or colloidal mine tailings with pulverized anhydrite (naturally occurring anhydrous calcium sulphate (CaSO4)), with or without other reagents or inert agents, to form a stiffer product.
The Problem To Be Addressed As of 2014, 109 cubic metres of colloidal MFT have accumulated in 130 km2 of water-capped ponds north of Fort McMurray, Alberta. The tailings are not load bearing, so they cannot be rehabilitated by reforestation, but require containment and vigilant management. They are an environmental liability for the industry. Rendering these tailings load-bearing would enable their reforestation.
Objectives Of The Invention To change the physical properties of these quasi-liquid tailings to make them sufficiently solid to be load bearing, and capable of sustaining revegetation; without the danger of collapse under stress or change in local physical conditions (temperature, water wind, overloading).
Page 4 of 13 Application Number 3,043,821 Most Appropriate Use Of The Invention The invention specifically targets, and was tested on, the "grandfathered"
colloidal MFT of McMurray oilsands mining operations.
Time and again, background studies in the literature suggest that the sodium ion is the villain, with what is essentially an electrical bond holding the clays in suspension. The silica silt fraction seems to play no part, being merely entrained by the clays into the colloid. Simplistically, challenging this electrical bond with a valence bond, challenges the equilibrium of the colloid.
Following previous workers, calcium might offer a realistic challenge. By like token, because physical dehydration of the clays is impossible, thought was given to tackling dehydration by chemical means. The author's experience in polymetallic epithermal and porphyry mineralization in the American cordillera and southeast Asia, showed naturally occurring anhydrite veinlets absorbing water on exposure, altering rapidly and irreversibly to the hydrated form.
Theoretically, therefore, the following should occur:
anhydrite fine tailings gypsum fine tailings LcaSO4 2H20 + (silt & clay) CaSO4.2H20 + (silt & clay) SOLID COLLOID SOLID SOLID
If the idea works, the question then is how much anhydrite would be necessary to achieve this. Anhydrite occurs naturally, in abundance, in the McMurray area. It could be quarried, with little or no overburden. The theory is to use a "hydroklept" (from the ancient greek tiowp = "water"
and Iola-I-co ¨ "I steal") to address the water entrained in the colloid.
Features That Are New And Distinct From What Has Been Done Before The use of anhydrite (CaSO4) has never been tried on MFT as a technique to stiffen them.
The guiding principle here is simplicity, with the intent of using locally abundant, inexpensive material which would have minimal short, medium and long term environmental impact and Page 5 of 13 Application Number 3,043,821 footprint. The goal is to convert the MFT from an environmental liability to an asset: whether that be as a viable substrate for revegetation of the tailings, a sterile capping for other MFT, or a useful raw material for a secondary industry being irrelevant. In short, the object is to make the MFT solid enough to walk and drive on with safety.
Scope Of The Invention:
The targeted materials are the very fine fraction (<45 ,m) of mine tailings containing an appreciable amount (>5%) of interstitial, capillary or adsorbed moisture that can only be removed by other methods with difficulty, if at all. In the case of the McMurray MFT
these are a very stable colloid, containing over 50% water, and the solids being composed of a preponderance of clays, kaolin predominating, with entrained very fine silica silt.
Wang's (2017) Table 2.2 (p 28) details the clay mineralogy of MFT, as determined by him and by other earlier researchers. The consensus depicts kaolinite at around 28% to 37% of the solid fraction of MFT, illite ranging from 11% to 36%, with minor smectite, muscovite (sericite), chlorite and rare iron oxide material. Typically, adsorbed on the individual grains of clay is a thin film of water that cannot be mechanically removed (Mitchell and Soga, 2005).
The reactions that comprise this invention take place at standard (20 C) temperature, near neutral to slightly alkaline pH, and standard (101.3 kilopascals) atmospheric pressure.
1. Results Of Laboratory Tests A 2 litre sample of Mature Fine Tailings (MFT) was received from Innotech on 2018-11-05, and allowed to rest, sealed in the original plastic container at room temperature (-20 C) until 2019-01-10. The MFT had a thin cover of supernatant liquid, mostly water with a strong petroleum odour, slightly discoloured with a fme suspension of black tar flecks. This liquid was decanted, Page 6 of 13 Application Number 3,043,821 along with about 30 per cent of the MFT, to ensure relative physical homogeneity of samples being tested.
From the 70 per cent remaining in the original container, approximately 50 gm aliquots of MFT, drawn off with a turkey baster, were accurately weighed into separate beakers at room temperature. Two of these samples ('A' and 'B') were used as controls. Sample 'A' had no additives, but was stirred vigorously for two minutes, then left to rest. To sample 'B' was added approximately gm, accurately weighed, of 20-40 mesh clean silica 'frac' sand, ground to 80%
<200 mesh and stirred vigorously for two minutes, then left to rest.
To 7 MFT 50gm samples approximate aliquots of 0.5 gm to 5 gm finely pulverized, pure, 'Laboratory Grade', CaO were accurately weighed and added, then vigorously stirred for two minutes. These form a further set of controls, essentially repeating prior experiments for comparison to the current experiment. To another 7 MFT 50gm samples, approximate aliquots of 0.5 gm to 7.5 gin of anhydrite from the Keg River formation of the Devonian Elk Point Group (from core from oil well 12-05-110-04 W5) were added, then stirred vigorously for two minutes.
All samples were then allowed to rest, and were observed.
Page 7 of 13 Application Number 3,043,821 MFT REAGENT ID REACTION
gm gm % min 50.04 5.01 9.1% CaO 1 0 50.73 2.73 5.4% CaO 12 2 50.91 2.46 4.6% CaO 3 1 49.92 0.67 1.3% CaO 11 11 50.28 0.51 1.0% CaO 2 36 49.87 0.42 0.8% CaO 13 120 49.90 0.42 0.8% CaO 14 120 n = 7; linear regression slope -1132; intercept 79; correlation coefficient -0.6551; covariance -0.9860.
MFT REAGENT ID REACTION
gm gm % min 50.69 7.63 15.0% Anh 14 2 49.04 7.52 13.3% Anh 3 0 47.92 5.18 10.8% Anh 13 5 49.76 5.54 10.0% Anh 1 5 50.02 2.67 5.3% Anh 12 13 50.32 0.69 1.4% Anh 11 10 49.86 0.51 1.0% Anh 2 120 n = 7; linear regression slope -486; intercept 62; correlation coefficient -0.6295; covariance -1.3154.
The same series of tests were then done on six samples of 50 gm aliquots that had been left at room temperature in open beakers to dehydrate by themselves for 96 hours prior to testing. The idea was to ascertain whether results could be improved by using partially ameliorated material.
MFT REAGENT ID REACTION
gm gm % min 49.55 2.02 4.8% CaO 8 0 50.04 1.48 3.5% CaO 7 1 50.74 1.12 2.6% CaO 9 15 50.05 1.05 2.4% CaO 6 8 50.13 0.54 1.3% CaO 5 10 49.96 0.25 0.6% CaO 4 15 n = 6; linear regression slope -355; intercept 17; correlation coefficient -0.8209; covariance -0.0679.
Once it had been established that the reagents had the desired effect on MFT, whether the product was stable in the presence of further water was tested, to see if the process would merely reverse itself. Approximately 2 gm of the material, 96 hours after the original mixing, was shaken vigorously for 2 minutes in 3 to 4 gm of water, then left to rest for 24 hours in sealed vials. The Page 8 of 13 Application Number 3,043,821 relative volumetric separation into clear supernatant liquid and un-separated "sludge" was then measured.
MFT REAGENT ID REACTION
gm gm % min 50.19 2.50 6.0% Anh 9 0 50.90 2.08 4.9% Anh 8 1 50.14 1.50 3.5% Anh 7 3 49.68 1.01 2.5% Anh 6 3 50.07 0.52 1.3% Anh 5 9 50.10 0.25 0.6% Anh 4 11 n = 6; linear regression slope -203; intercept 11; correlation coefficient -0.9361; covariance -0.0717.
Observations The received sample of MFT, before any test work, has the consistency of yoghurt.The samples ('A') to which no reagents had been added, and ('B') to which inert silica powder was added showed no discernable change in two hours. After 96 hours both were slightly more viscous, having lost about 13 to 20 percent (by weight) by evaporation.
ID

gm gm WATER ATION
A 1.79 3.42 66% 4%
SiO2 B 1.05 4.75 82% 40%
CaO 1 1.45 5.12 78% 32%
CaO 2 1.89 5.10 73% 31%
CaO 3 1.24 5.75 82% 36%
Anh 1 1.51 4.64 75% 28%
Anh 2 2.12 3.94 65% 16%
Anh 3 1.69 3.74 69% 23%
Both calcium oxide and calcium sulphate reacted with the "as delivered" MFT, but with erratic results. Below 1.0% calcium oxide, there was little perceptible thickening of the MFT within two hours, although after 24 hours the MFT had thickened to almost paste-like consistency. This was not discernable in the two control samples, 'A' and 'B'. After 96 hours these MFT samples with Page 9 of 13 Application Number 3,043,821 low concentrations of quicklime had been reduced to a much thicker paste, again after losing about 20% by weight.
ID ID
WATER LOSS WATER LOSS
A -22% S102 B -13%
CaO 1 -21% Anh 1 -22%
CaO 2 -20% Anh 2 -20%
CaO 3 -26% Anh 3 -17%
CaO 4 -17% Anh 4 -22%
CaO 5 -16% Anh 5 -20%
CaO 6 -15% Anh 6 -21%
CaO 7 -19% Anh 7 -19%
CaO 8 -19% Anh 8 -20%
CaO 9 -17% Anh 9 -21%
When 1.0% anhydrite was added to the MFT, there was no discernable reaction in

2 hours, nor after 24 hours. However, after 96 hours the samples were noticeably thicker than the two control samples ('A' and IV), although not as viscous as the low concentration quicklime samples, again after losing 20% water by weight.
Where concentrations of both quicklime and anhydrite were in excess of 1.0%, the reaction time was very rapid, ranging from immediate where 5% or more of calcium oxide or 12% or more anhydrite was added to the MFT to ten minutes to half-an-hour where the concentrations were dropped to about 1%. After reaction, the MFT was reduced to a firm paste, rather like cookie dough. Within 24 hours of being left open in beakers at room temperature, these mixtures were all solid, albeit damp, and after 96 hours all but the lowest concentrations were rock hard and apparently dry, all having lost about the same 20% water content.
Not surprisingly, all reaction times were markedly more rapid after the samples had been allowed to dry out for 96 hours. Furthermore, the concentrations of reagent required to initiate the Page 10 of 13 Application Number 3,043,821 reactions were also much lower. The MFT seemed to become firmer, and to attain a harder final consistency after a further 96 hours of standing when quicklime rather than anhydrite was used.
The process is irreversible in every case.
Conclusions There is a clear, tangible change associated with the addition of both calcium oxide and anhydrous calcium sulphate, the former being more apparent than the latter.
The lack of tangible change in 'A' and 'B' indicates that:
= The change is not a function of time;
= The change is not a function of stirring;
= The change is not a function of mere bulking by addition of material.
= Regardless of concentration of reagent, water was not released initially from the product of the reaction, indicating that both reagents are hydroklepts.
= The change is much more rapid and occurs at lower concentrations where the reagent is quicklime, not anhydrite.
= Prior dehydration of the MFT accelerates and accentuates the change.
= The change is irreversible.
= The change cannot be ascribed only to dehydration of the colloid by the reactions:
CaO + H20 Ca(OH)2 and CaSO4 + 2H20 CaSO4.2H20 Page 11 of 13 Application Number 3,043,821 While such dehydration no doubt occurs, it could only be partly responsible for the steep increase in viscosity that is observed in all but the 1% anhydrite addition. Evidently, there is something more fundamental taking place.
Practical Use of the Invention Anhydrite occurs in abundance within a few hundred kilometres to current mining operations of the Canadian Oil Sands. It can be mined by open pit quarry, transported, pulverized, added to the MFT, and thoroughly mixed with it. This will produce a firm, load-bearing substrate; which will allow the complete revegetation and rehabilitation of the extensive (>1301=2) currently water-capped tailings ponds.
References Cited Anon 2017 Soft Tailings Capping Technology challenge Canada's Oil Sands Innovation Alliance (COSIA), 6pp.
Anon 2017 Tailings Clay Challenge Canada's Oil Sands Innovation Alliance (COSIA), 7pp.
Anon 2017 Improving Environmental Performance through Open Innovation Canada's Oil Sands Innovation Alliance (COSIA) Project Portfolio, 74pp.
BGC Engineering Inc. 2010. Oil Sands Tailings Technology Review. Oil Sands Research and Information Network, University of Alberta, School of Energy and the Environment, Edmonton, Alberta. OSRIN Report No. TR-1. 136 pp.
Liu H, Tan S.Y, Yu T, Liu Y. (2016) Sulfate reducing bacterial community and in situ activity in mature fine tailings analyzed by real time qPCR and microsensor.
Journal Of Environmental Sciences 44; pp141 ¨ 147 Mitchell J.K. and K. Soga, (2005) Fundamentals of soil behaviour. 3rd ed., NY: Wiley.
Wang Y. (2017) A study on chemical stabilization of Oil Sands Mature Fine Tailings M.
Eng. Sc. thesis University of Western Ontario 144pp.
Page 12 of 13

Claims

CLAIM

The use of pulverized anhydrite (anhydrous Calcium Sulphate (CaSO4)), to cause an obvious, tangible change immediately and irreversibly, when admixed, at 0.5%
to 12.0% by weight, to thixotropic mine tailings, and colloidal Mature Fine Tailings (MFT) derived from mining and processing of the Alberta Oil Sands, to become load bearing, wherein prior dehydration of the tailings accelerates and accentuates said change.