Possible role of hydrogen sulphide gas in self-heating of pyrrhotite-rich materials
Introduction
Self-heating of sulphide minerals is the result of oxidation reactions. The material self-heats when, without any external heat input, the rate of heat release by exothermic reactions exceeds that of heat loss. Self-heating of sulphides may give rise to dramatic consequences. An historical example is the sinking of the N.Y.K. liner S/S Boyko Maru in 1939, while transporting copper concentrate (Hikami, Katuyuki, 1942 cited by Kirshenbaum (1968)). Oxygen depletion and fires in mines have been reported at various times (Tally et al., 1917, Harrington et al., 1923, Rachilly and Butte, 1923, Lukaszewski, 1969, Farnsworth, 1977, Good, 1977). Based on 53 years of observation (from 1869 to 1923), Harrington et al. (1923) reported an average loss of 10 lives per year due to “metal-mine fires”. Self-heating of sulphides when transporting iron–copper ore was reported as early as 1871 (Stevens (1871) as quoted by Kirshenbaum (1968)). Good (1977) described streams of molten metal from draw holes and sulphide dust explosions in the Sullivan mine (British Columbia). A photograph of the incandescent sulphide ore was the cover page of the CIM Bulletin vol. 70/782, year 1977. Self-heating and smouldering of sulphides poses problems of air quality in and around mines (Brown and Miller, 1977) and contributes to acid mine drainage.
The reactions initiating self-heating at low temperature are not known (Good, 1977, Rosenblum et al., 1982, Rosenblum and Spira, 1995). Oxidation of sulphides by oxygen (air) (Lukaszewski, 1969, Good, 1977, Ninteman, 1978, Rosenblum and Spira, 1995), by dissolved FeIII (Lukaszewski, 1969) or by nitrates (Lukaszewski, 1969) has been considered. Good (1977) enumerated 12 exothermic reactions, based on on-site observation of reaction products where self-heating occurred. These products included ferrous and ferric sulphates and hydroxysulphates, ferric oxides and hydroxides, sulphuric and sulphurous (SO2 dissolved in water) acids and elemental sulphur. Incipient heat was attributed to reactions giving rise to elemental sulphur (Good, 1977, Rosenblum and Spira, 1993, Rosenblum and Spira, 1995, Rosenblum et al., 2001).
Elemental sulphur is considered to result from reactions between ferric oxide and sulphuric acid or from the reaction between pyrrhotite (FeS) and ferric sulphate. Hydrolysis of iron species, formation of water soluble iron sulphate (Good, 1977, Steger, 1982) and heat of hydration were suggested (Rosenblum and Spira, 1995) as responsible for temperature increases up to 100 °C. Rosenblum and Spira (1993) monitored temperature increase from room temperature in self-heating tests. They showed that the weight gain measured at the end of experiments conducted at 40 °C, 50 °C and 70 °C was proportional to the S0 content. Goethite and iron sulphates were also reported as reaction products. However, synthetic mixtures of elemental sulphur and sand or hematite did not heat (Rosenblum and Spira, 1993, Rosenblum and Spira, 1995).
Self-heating risk assessment methods were reviewed by Rosenblum and Spira (1981). Based on experience using column tests (Rosenblum et al., 1982), a self-heating assessment method was developed and applied to various sulphide-bearing materials (Rosenblum and Spira, 1993, Rosenblum and Spira, 1995, Rosenblum et al., 2001). The second generation technique was based on automated calorimetric cells. A standard test comprises measuring heat output during a sequence of air injections for about one week at two temperatures, typically 70 °C (stage A) and 140 °C (stage B). By comparing heat output in stages A and B, a sample could be ranked for self-heating risk. Most nickel sulphide concentrates were classified with a high potential of self-heating, a response presumably due to their pyrrhotite content, while the reactivity of copper, lead and zinc sulphide concentrates was site specific (Rosenblum et al., 2001). The technology is now available to McGill researchers and was used for the present study.
The standard self-heating test is more direct although more complex than an alternative procedure that has been proposed based on measuring weight gain over a few weeks (e.g., Steger, 1976, Rosenblum and Spira, 1993, Wu and Li, 2005). It is more sophisticated than the standard risk assessment procedure adopted by the United Nations (1995) which is commonly applied for transport of dangerous goods.
The present work aims to better understand how self-heating is initiated as a step towards mitigation and control. Pyrrhotite (Po) was chosen as test mineral because it is commonly regarded as the most reactive among the common sulphides. The tests followed the protocol of Rosenblum and Spira (1995), using mixtures prepared with varying proportions of pyrrhotite and coarse quartz sand (Test series I). A wider range of controlled Po content was achieved in this manner compared to previous test work (Rosenblum and Spira, 1995). With increasing Po content, self-heating rates progressively increased but the samples visually appeared less and less oxidized. A high Po content, it was hypothesized, may give a more reducing environment favouring formation of hydrogen sulphide (H2S). The procedure was then modified to test for hydrogen sulphide gas (H2S) release (Test series II). Finally, the role of H2S on self-heating was indirectly tested: it is shown that by capturing H2S as a stable sulphide before it reacts with oxygen, self-heating was suppressed (Test series III). To our knowledge, this is the first time that a role of H2S has been considered with regard to self-heating of sulphides. It introduces a potentially important avenue of investigation into self-heating mechanisms.
Section snippets
Sulphide samples
The material in Tests series I and II were samples from the pyrrhotite tail (PoT) stream of the Falconbridge (now Xstrata Nickel) Strathcona mill (Sudbury, Canada). This stream runs ca. 70% Po with particle size 80% − 48 μm (Wells et al., 1997). There is a magnetic separation step and the Po in the samples is predominantly monoclinic. While questions of oxidation and contamination with flotation reagents can be raised the overwhelming advantage is the material is available in large quantities, an
Test series I: pyrrhotite/sand mixtures
In terms of risk assessment as described by Rosenblum and Spira, 1995, Rosenblum et al., 2001, all samples were in the high risk zone 5 (“Mitigation methods are required”).
The thermograms (Fig. 2) illustrate a fast, exothermic response as soon as air is injected, except for the initial air injections on the 10% (and less, not shown) PoT samples. For these Po-poor samples, it seems there is an induction period. A slight (up to 4 °C) and progressive increase in heating rates and of the base
Discussion
The abrupt and fast response to air injection (Fig. 2) suggested exothermic reactions between air and a gas/vapour or a catalysed reaction (by mineral surfaces or by water), rather than liquid water-based reactions or direct gas (air)–solid reactions. Because the temperature returned quickly to the base level once air injection was stopped, and at about the same rate for all samples, this excluded a major contribution to self-heating from physical parameters (differences in heat conductivity)
Conclusion
Standard self-heating tests on pyrrhotite (Po)–sand mixtures showed increasing self-heating as Po content increased but less and less visual evidence of oxidation. It was hypothesized that high Po content produced conditions favouring hydrogen sulphide (H2S) formation and that it was oxidation of H2S which contributes to self-heating. A gas phase reaction is suggested by the abrupt and rapid response to air injection. Formation of H2S also offers an explanation of the importance of moisture in
Acknowledgments
We are grateful to Frank Rosenblum and Jan Nesset for training on the self-heating apparatus, sharing their experience and their dedicated support of the project; and to Dr. In-Ho Jung for thermodynamic calculations using FactSage. We thank our colleagues, Ray Langlois for sample preparation, and Jean-François LeBerre, Lang Shi and Monique Riendeau for analytical support. This work was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) Collaborative Research and
References (47)
- et al.
The formation of Cu2S from the elements: I. Copper used in form of powders
Thermochimica Acta
(2000) - et al.
The formation of Cu2S from the elements: II. Copper used in form of foils
Thermochimica Acta
(2001) - et al.
X-ray photoelectron spectroscopy of oxidized pyrrhotite surfaces: I. Exposure to air
Applications of Surface Science
(1985) - et al.
Anoxic dissolution of troilite in acidic media
Journal of Colloid and Interface Science
(2006) - et al.
Troilite oxidation by hydrogen peroxide
Journal of Colloid and Interface Science
(2006) - et al.
Electrochemical characterization of pyrrhotite reactivity under simulated weathering conditions
Applied Geochemistry
(2005) - et al.
The synthesis of some copper sulfides and copper sulfosalts in 500–700 gram quantities
Materials Research Bulletin
(1973) - et al.
Spectroscopic and XRD studies of the air degradation of acid-reacted pyrrhotites
Geochimica et Cosmochimica Acta
(2002) - et al.
XPS identification of bulk hole defects and itinerant Fe 3d electrons in natural troilite (FeS)
Geochimica et Cosmochimica Acta
(2004) The stability of certified reference ores MP-1, KG-1 and SU-1 towards air oxidation
Talanta
(1976)
Oxidation of sulfide minerals: VII. Effect of temperature and relative humidity on the oxidation of pyrrhotite
Chemical Geology
The role of surface sulfur species in the inhibition of pyrrhotite dissolution in acid conditions
Geochimica et Cosmochimica Acta
Kinetic factors for oxidative and non-oxidative dissolution of iron sulfides
Minerals Engineering
A mechanism to explain sudden changes in rates and products for pyrrhotite dissolution in acid solution
Geochimica et Cosmochimica Acta
Oxide and sulphide layers on copper exposed to H2S containing moist air
Corrosion Science
(Water + hydrogen sulphide) Association. Second virial cross coefficients for (water + hydrogen sulphide) from gas phase excess enthalpy measurements
The Journal of Chemical Thermodynamics
Nouveau Système de Minéralogie
Pollution problems associated with Sullivan Mine hot muck
CIM Bulletin
NIST-JANAF Thermochemical Tables
Introduction to and background of sulphide fires in pillar mining at the Sullivan Mine
CIM Bulletin
The oxidation of sulphide minerals in the Sullivan Mine
CIM Bulletin
Cited by (38)
Microstructure and mechanical behavior of cemented gold/tungsten mine tailings-crushed rock backfill: Effects of rock gradation and content
2023, Journal of Environmental ManagementMechanochemical effect of spontaneous combustion of sulfide ore
2022, FuelCitation Excerpt :Therefore, spontaneous combustion disasters of the sulfide minerals can cause a series of safety and environmental problems and huge economic losses. From the literature cited by Kirshenbaum (Stevens, 1871), accidents concerning the self-heating of sulfide ore during transportation were first reported in 1871 [3]. The issue of ore spontaneous combustion has also received increasing attention since the International Maritime Solid Bulk Cargoes Code (IMSBC) has been in effect since 2011 [6].
Oxidative self-heating modeling of iron sulfides during the processing of high sulfur oil
2022, Process Safety and Environmental ProtectionExperimental study on microbial desulphurization of sulfide ores and self-heating simulation of ore heaps under ultrasonic and microwave
2022, Process Safety and Environmental ProtectionSurface and interface analysis of iron sulfides in aqueous media using X-ray photoelectron spectroscopy of fast-frozen dispersions
2021, Applied Surface ScienceCitation Excerpt :The spectra collected at RT from pyrite samples dried in vacuum or air are quite similar and are not shown in Figures. Acidic etching of metal sulfides is applied as a pretreatment for cleaning the material (mineral) surfaces and removing metal oxides; the leaching in hydrometallurgy and natural processes, particularly weathering and AMD production occur in acidic media [5–15,33–49]. The interaction of pyrrhotite with diluted acids is known to proceed either releasing hydrogen sulfide under non-oxidative conditions or, much more slowly, without perceptible formation of H2S at higher electrochemical potentials induced by aqueous low-concentration ferric ions [40–42,44–49].
- 1
Present address: COREM, Quebec, QC, Canada G1N 1X7.