 Commonwealth of Australia 2010


Environmental transport and distribution



Download 2.29 Mb.
Page16/58
Date20.10.2016
Size2.29 Mb.
#6264
1   ...   12   13   14   15   16   17   18   19   ...   58

23.3Environmental transport and distribution

23.3.1Free cyanide


Alkaline solutions (e.g. pH >9.2) of sodium cyanide contain a high proportion of the free cyanide as the cyanide ion CN-. Free cyanide is very reactive and does not occur commonly in nature (USEPA, 1980). In aqueous solutions of simple alkali cyanides, the CN group is present as the cyanide ion (CN-) and molecular hydrogen cyanide (HCN), which may dissociate to form the cyanide ion depending on pH according to the reaction:

H+ + CN-  HCN(g)

‘Free cyanide’ in water is generally considered as the sum of cyanide present as both HCN and CN- USEPA, 1980; Staunton et al., 2003. In most natural waters (e.g. pH 6 to 8), HCN predominates (APHA, 1998; Brix et al., 2000).

At any particular pH and temperature, the system is in equilibrium and the relative amounts of each can be determined from the following equation (interpreted from Smith and Mudder, 1993 and USEPA, 1985):

Ka = [H+].[CN-] / [HCN] = ~4.5x10-10 to 6.0x10-10, pKa = 9.31-9.35 (at 20°C).

At a pH corresponding to the pKa value (i.e. pH ~9.3 at 20°C), the concentration of HCN and CN- in solution are equal. Using a pKa value of 9.33, at 20°C the calculated concentrations of CN- and HCN at other pHs are approximately as follows:



  • at pH 10.5: 94% CN- and 6% as HCN;

  • at pH 9: 32% CN- and 68% HCN;

  • at pH 8: 5% CN- and 95% HCN; and

  • at pH 7: 0.5% CN- and 99.5% HCN.

The pKa value decreases with increasing temperature, e.g. pKa = 9.63 at 10°C, 9.09 at 30°C and 8.72 at 45°C (calculated from an equation in USEPA, 1985). Under typical environmental conditions of pH <8 and water temperature <25°C, >94% of free cyanide exists as HCN. In the processing plant, temperatures may be elevated by residual heat in the slurry due to the milling process. At pH 9, HCN contributes ~55% of the free cyanide at 30°C and ~34% at 45°C. At pH 10.5, HCN contributes ~4% of the free cyanide at 30°C and ~0.5% at 45°C.

A reaction related to the release of HCN by decreasing pH is that with carbon dioxide (CO2) in the presence of water:

CN- + CO2 + H2O  HCN + HCO3-

It is noted that because of the toxicity and flammability hazards presented by HCN gas, carbon dioxide is not used for control of fires involving sodium cyanide.


23.3.2Volatilisation of HCN


HCN has a vapour pressure of 100 kPa (750 mm Hg) at 26°C and is regarded as highly volatile (Chatwin et al., 1987). The Henry’s Law Constant (KH = 0.132 L atm/mol at ~25°C – Lye, 2002; W. Staunton, pers. comm. 2005) indicates HCN also has high volatility from water. HCN is lighter than air, with a relative vapour density = 0.94, compared to air = 1 (CDC, 2005).

Volatilisation can be a significant removal process for free cyanide from aqueous solutions and soils to air (Smith and Mudder, 1993). Volatilisation was considered a major loss mechanism for HCN from surface soils of pH <9.2 (Hagelstein and Mudder, 1997a).

The volatilisation rate of HCN from aqueous solutions (i.e. HCN(aq)  HCN(g)), with pH standardised, increases with increasing temperature (as does the Henry’s Law Constant) and solution agitation (Lye, 2002; Staunton, pers. comm. 2005). There are also indirect influences which change the amount of HCN which volatilises with increasing ionic strength (lowering the pKa value), but effects of ionic strength are only significant for the hypersaline operations in WA (Staunton, pers. comm. 2005). Laboratory testwork reported in Adams et al, (2008abc) indicated that 90% HCN occurs at about pH 8.0 for fresh water solutions and 8.5 for solutions of total dissolved solids (TDS) of ~200,000 mg/L. Solution depth may also be an important factor affecting HCN volatilisation from surface waters (Chatwin et al., 1987).

The main factors affecting emission of HCN from soil include soil pH and availability of continuous vapour pathways (e.g. pores, cracks; Smith and Mudder (1993)).

Volatilising HCN gas in soils has five possible fates (listed below). Of these, the latter two are considered least significant (Chatwin et al., 1987):


  • volatilisation of HCN to the surface and then the atmosphere;

  • biodegradation and metabolism by micro-organisms;

  • reactions with soil constituents in the subsurface, particularly at low soil moisture content;

  • dissolution in soil moisture, particularly if the solution pH is high; and

  • entrapment in cavities in the subsurface soil.

Volatilisation from groundwater is unlikely to be a significant transformation process.

HCN is lighter than air and rises from surfaces where it is released. Hence volatilised HCN is not expected to accumulate when released to the external environment, and its flammability is not an environmental concern.


23.3.3Atmospheric fate of HCN


In air, inorganic forms of cyanide are found mainly as HCN gas and a smaller amount as fine dust particles (ATSDR, 2006). Cicerone and Zellner (1983) reviewed the atmospheric fate of HCN, and more recent observations by various means of HCN concentrations in the atmosphere have added to this knowledge (Li et al, 2000; Singh et al, 2003; Rinsland et al, 1996, 2001, 2005; Kleinböhl et al, 2006; Pumphrey et al, 2006). HCN is slightly lighter than air (Section 23.3.2) and therefore tends to rise from the immediate point of release, becoming mixed in the air as it does so. Atmospheric modelling indicates that HCN is well-mixed in the troposphere, but decreases in concentration in the stratosphere with increasing altitude.

In the troposphere, HCN may degrade via reactions with hydroxyl and oxygen radicals, with that involving hydroxyl radicals (OH) thought to be the principal one, ultimately producing nitric oxide in a series of rapidly occurring reactions:

HCN + OH   NO + CHO

In the stratosphere, a UV photolysis reaction may also occur, followed by oxidation of the CN radical which is produced:

HCN + uv  H + CN

CN + O2  NCO- and NCO-  N + CO

Photolysis of HCN in the troposphere and lower stratosphere is negligible, and precipitation of cyanide in rainfall is a negligible sink for atmospheric HCN. HCN is a strongly bound molecule and Cicerone and Zellner (1983) estimated its atmospheric lifetime (the time taken for the concentration to fall to approximately 37% (1/e) of its initial concentration) based on reaction with hydroxyl radicals to be approximately 2.5 years (range 1 to 5 years).

However, Li et al. (2000) argued that the observed seasonal variations in atmospheric HCN concentrations imply an atmospheric lifetime of only a few months for HCN, i.e. much shorter than is commonly assumed if oxidation by OH is the main sink (i.e. as discussed above). They proposed that uptake by dry deposition to the ocean provides the missing sink and examined this using a global 3-D model simulation, with ocean uptake as the main sink and biomass burning as the main source. From this, they estimated an HCN atmospheric lifetime of 2-4 months on their standard assumptions, or 4.4 months with some different assumptions. Singh et al. (2003) and Li et al. (2003) incorporated measured data from in situ aircraft observations of HCN in the troposphere (rather than simply data from remote sensing of the total column of HCN in the atmosphere). Using different modelling approaches, they arrived at a tropospheric lifetime for HCN of 5-5.3 months.

HCN taken up by ocean waters is expected to be degraded by biological processes, and this uptake limits the amount of NO generation that would have been anticipated if the principal method by which HCN dissipated from the atmosphere were via reaction with OH. Typical measured concentrations (mixing ratios) of HCN in the troposphere are ~180-250 parts per trillion by volume (pptv) in various studies.

Based on figures from Cicerone and Zellner (1983) and an estimate of the world’s usage of NaCN for gold mining in the early-mid 1990’s, De Vries (1996) argued that gold mining sources contributed only a minor proportion of annual HCN infiltration to the atmosphere and was an insignificant contributor to nitrogen oxide formation in the atmosphere. Total HCN release to the atmosphere from various biogenic and anthropogenic sources has more recently been estimated to be of the order of 1 million tonnes (1 x 1012 g) per annum globally in terms of the contained nitrogen (Li et al., 2003). To update the argument of de Vries (1996), if it is assumed that use of NaCN for gold mining is ~330 000 tonnes per annum and that at worst ~30%-50% of this quantity is ultimately released to the atmosphere as HCN (Sections 23.6.3 and 23.6.4), this equates to ~28 000-47 000 tonnes per annum N as HCN, or ~3-5% of estimated total HCN release. This is at most a relatively minor contribution to overall HCN release, comparable in scale to estimated global release via car exhaust (as listed by Li et al., 2003).

Small amounts of metal cyanides may be present as particulate matter in air, from which it is removed by both wet and dry deposition (HSDB, 1991). The average half-life and lifetime for particles in the troposphere are estimated to be about 3.5 to 10 days and 5 to 15 days, respectively.

23.3.4Complexation


In solutions of simple metal cyanides, the CN group may also react with metals to form complex metal-cyanide anions of varying stability (ANZECC/ARMCANZ, 2000a; APHA, 1998).

Cyanide complexes, of which there are many combinations (Lye, 2002; Smith and Mudder, 1993; Table 3.), are traditionally classified according to their stabilities as weak (cyanocomplexes of Ag, Cd, Cr, Cu, Hg, Mn, Ni, Zn) or strong (cyanocomplexes of Au, Co, Fe, Mo, W, Re, Pt-group) (Flynn and Haslem, 1995). Many simple metal cyanides are sparingly soluble in water (weak; e.g. CuCN, AgCN, Zn(CN)2); however, a variety of soluble complex metal cyanides may be formed in the presence of alkali cyanides (APHA, 1998).

Some iron-cyanide complexes are not completely stable between aqueous and solid phases (Ghosh et al., 1999). Only Prussian Blue (Fe4(Fe(CN)6)3) and Turnbull’s Blue (Fe3(Fe(CN)6)2) are stable in a wide range of pH and redox conditions. Other less complex iron cyanides, such as ferrocyanide (Fe(CN)64-) and ferricyanide (Fe(CN)63), may be present in solution, with the ability to speciate to free cyanide depending on the pH and redox potential of the water. Ferro- and ferri-cyanides also act as non-adsorptive solutes in sandy gravelly aquifers (Ghosh et al., 1999).

23.3.5Adsorption and mobility in soil


Distribution co-efficients (Kd) for soil and sediment adsorption by cyanide (sodium cyanide at pH 10.5) range from 5.04 to 14.5, indicating that the CN- ion can bind to these media (USEPA, 2006c). However, due to its high water solubility, HCN is only weakly bound or partitioned to sediments, soils and organic matter, and its mobility in waters is potentially high (Callahan et al., 1979). In addition to low pH, cyanide mobility is relatively low in soils with a high concentration of free iron oxides, positively charged particles and clays such as kaolin, chlorite, gibbsite (Callahan et al., 1979). Alesii and Fuller (1976) indicated that soils with a high anion-exchange capacity, soils with high concentrations of manganese and hydrous oxides of iron, are more likely to attenuate cyanide. Mobility of cyanide is greater in soils with high pH, high concentration of free calcium carbonate (high negative charge) and low clay content (Callahan et al., 1979). Adsorption of HCN by soils containing strongly cation-exchanging materials (e.g. montmorillonitic clays) is fairly weak and is decreased by the presence of water (Alesii and Fuller, 1976; Callahan et al., 1979).

Chatwin et al. (1987) investigated the attenuation of cyanide in several soil types using soil column tests, revealing that soils can have significant cyanide attenuation abilities. Chatwin et al. (1987) identified two major physicochemical mechanisms that attenuate cyanide in soils including:



  • volatilisation; and

  • reaction or adsorption to alumina (bauxite or kaolinite) and organic matter, and to a lesser extent, plagioclase, potassium feldspar, ilmenite and hematite.

Cyanide adsorbed weakly to ferric forms of iron in the soils tested by Chatwin et al. (1987). Increasing the pH resulted in the solubilisation of this cyanide. Volatilisation was significant due to the moderately alkaline to acidic nature of the soils they tested.

Soil attenuation capacity for free cyanide is not limitless, as indicated by the Australian Minerals Industry Research Association (Staunton, 1991c-d). AMIRA (1991c) found, using trickle-flow column tests with soil from one mine site, that good initial attenuation of cyanide was exhibited, but that this fell rapidly as more cyanide solution passed through. In the results shown in Table 6., ~50% of the initially applied cyanide was converted to other forms, notably cyanate (OCN). However, attenuation soon fell to ~15%, with little evidence of continuing reactions with soil constituents (Table 6.). In contrast, the soil continued to show good pH buffering capacity, with the soil pH remaining at a value intermediate between that of the applied solution (pH 10) and that in the original soil (pH ~5.5). The authors suggested that cyanate formation was due to organic carbon in the soil.

Table 6.. Composition of leachate and % attenuation of applied cyanide in soil columns in laboratory tests (from Staunton, 1991c).


No. Pore Volumes

pH

WAD CN** (mg/L)

Cu(CN)32-

(mg/L)

Fe(CN)64-

(mg/L)


OCN*** (mg/L)

% CN attenuation

0.7

6.6

13

6.6

0.64

13

50

2.9

6.0

22

1.0

0.20

3

15

4.9

7.6

22

0.48

<0.14

<1

15

8.0

5.9

20

<0.06

<0.14

<1

23

11.0

7.2

21

0.37

<0.14

4

15

* Initial leachant concentration 26 mg/L with pH 10.0. Initial soil pH 5.6. ** Total CN concentration was in each case = WAD CN concentration. *** SCN concentration in each case <0.5 mg/L.

The persistence of cyanide in the groundwater environment is influenced by groundwater chemistry, aquifer composition and groundwater microbiology (Meehan, 2000). Free cyanide concentrations may decrease through chelating with transition metals (complexation; Smith and Mudder, 1993), which may result in precipitation from solution (Theis and West, 1986). Stable, insoluble ferro- and ferricyanide complexes may form, which precipitate in a range of pH and redox conditions. Based on its relative abundance in soils, Smith and Mudder (1993) indicated that the majority of cyanide metal complexation reactions in soils would involve iron.

Although dissolution of iron-cyanide complexes in soils can result in the release of free cyanide, these compounds are quite stable in soil, with long half-lives. Under acidic (pH 4) and reduced conditions the half-life of iron complexes may be in the order of 1 year; however, under conditions which would be considered normal for near surface soils (pH >6, aerobic conditions) the half-life is in the order of 100-1000 years (Kjeldsen, 1999).

Kjeldsen (1999) reported that free cyanide concentration in groundwater at Dutch gas works sites contaminated with iron-cyanide complexes constituted less than 1% of the total cyanide content, and spent bog iron leachate tests revealed no detectable free cyanide (Theis et al., 1994).

The sparingly soluble metal cyanides (e.g. copper, nickel, zinc) may adsorb onto particulates and partition to the sediment compartment (Hagelstein and Mudder, 1997a).


23.3.6Cyanide complex precipitation


Ferrocyanide and ferricyanide ions form insoluble iron-cyanide precipitates with Fe, Cu, Ni, Mn, Pb, Zn, Cd, Sn, Cd, and Ag (Weast, 1969), through the pH range 2 to 11 (Hendrickson and Daignault, 1973). Examples include the iron complexes Prussian blue and Turnbull’s Blue (Section 23.3.4). Iron complexes may react with thiocyanate to form stable complexes if sulphur is present in the soil (Smith and Mudder, 1993).


Download 2.29 Mb.

Share with your friends:
1   ...   12   13   14   15   16   17   18   19   ...   58



The database is protected by copyright ©ininet.org 2024
send message
    Main page
harmful effects
australian government department
many thousands
future need
director provides applicants
other focussing