mercury emissions from point sources in south africa

Mercury Emissions from Point Sources in South Africa

J Leaner, James Dabrowski, R Mason, P Ashton, K Murray,

T Resane, M Richardson, M Ginster, R Euripides,

E Masekoameng and G Gericke

© CSIR 2006 www.csir.co.za

Background

• Coal combustion provides the largest source of energy to South

Africa. Many households burn coal for heating and cooking purposes.

• Due to South Africa’s reliance on coal as a primary energy source, an

increase in Hg emissions seems inevitable unless better control

measures are applied or more renewable energy sources are used.

• South Africa’s emissions was estimated to contribute about 16% of

global Hg emissions (1590.7 tonnes), following China (Pacyna et al.,

2003; 2006).

• South Africa released 256.7 tonnes of Hg to the atmosphere during 2000, with

most Hg emissions originating from industry production (gold), and stationary

combustion (coal combustion).


Background

• The published assessment signalled the need for a critical evaluation of

the major anthropogenic sources of Hg in South Africa.

• Information on actual Hg emission measurements of Hg in products

and resources are poorly understood in South Africa.

• A South African Mercury Assessment (SAMA) Programme was

established: A co-ordinated attempt towards assessing Hg levels in a

systematic approach in the country by Government, academia, research

councils and key industries recently initiated a (Leaner et al., 2006).

• Preliminary results of a Hg inventory and that of some Hg monitoring

undertaken at Cape Point’s Global Atmospheric Water Station (Baker et

al., 2002) in South Africa provided some information towards refining

the published Hg emission estimates.


Current Situation: Sources

!( Towns

#* Cement production units

#* Cement millling/blending units

Rivers

Highveld National Priority Area

Vaal Triangle Air-shed Priority Area

Residential areas

Water Management Areas

1-Limpopo

2-Luvuvhu and Letaba

3-Crocodile (West) & Marico

4-Olifants

5-Inkomati

6-Usutu to Mhlathuze

7-Thukela

8-Upper Vaal

9-Middle Vaal

10-Lower Vaal

11-Mvoti to Umzimkulu

12-Mzimvubu to Kieskamma

13-Upper Orange

14-Lower Orange

15-Fish to Tsitsikamma

16-Gouritz

17-Olifants/Doring

18-Breede

19-Berg

PowerStations

# Coal Fired (Base Load)

#0 Coal Fired (Mothballed)

") Gas Turbine

") Nuclear

") Pumped Storage



!( Towns



Rivers



Residential areas


1-Limpopo



4-Olifants

5-Inkomati


7-Thukela

8-Upper Vaal

9-Middle Vaal

10-Lower Vaal



13-Upper Orange

14-Lower Orange


16-Gouritz

17-Olifants/Doring

18-Breede

19-Berg

PowerStations



") Gas Turbine

") Nuclear

") Pumped Storage

!( Towns



Rivers



Residential areas


1-Limpopo



4-Olifants

5-Inkomati


7-Thukela

8-Upper Vaal

9-Middle Vaal

10-Lower Vaal



13-Upper Orange

14-Lower Orange


16-Gouritz

17-Olifants/Doring

18-Breede

19-Berg

PowerStations



") Gas Turbine

") Nuclear

") Pumped Storage


Highveld Priority Area

!( Towns



Rivers



Residential areas


1-Limpopo



4-Olifants

5-Inkomati


7-Thukela

8-Upper Vaal

9-Middle Vaal

10-Lower Vaal



13-Upper Orange

14-Lower Orange


16-Gouritz

17-Olifants/Doring

18-Breede

19-Berg

PowerStations



") Gas Turbine

") Nuclear

") Pumped Storage

!( Towns



Rivers



Residential areas


1-Limpopo



4-Olifants

5-Inkomati


7-Thukela

8-Upper Vaal

9-Middle Vaal

10-Lower Vaal



13-Upper Orange

14-Lower Orange


16-Gouritz

17-Olifants/Doring

18-Breede

19-Berg

PowerStations



") Gas Turbine

") Nuclear

") Pumped Storage


Coal Consumption in South Africa (DME, 2004)

Electricity Generation

61%

Iron and Steel

3%

Domestic

3%

Industry

3%

Mining

1%

Mettalurgical

1%

Synthetic Fuels

28%


Coal: Mine, Control Devices/Factors, Consumption

Power Plant Coal mine Emission

Control Device

Emission

Reduction

Factor

(UNEP, 2005)

Coal

Consumption

(Mtonnes/yr)

Arnot Arnot FF (Eskom, 2006) 0.5 6.0

Duvha Middleburg FGD (Eskom, 2001) 0.5 8.4

Hendrina Optimum FF (Bateman, 1998) 0.5 6.5

Kendal Khutala ESP + FGD (Eskom, 2001) 0.5 15.1

Kriel Kriel ESP + FGD (Eskom, 2001) 0.5 11.1

Lethabo New Vaal ESP + FGD (Eskom, 2002) 0.5 17.3

Majuba Various FF (Bateman, 1998) 0.5 10.0

Matimba Grootgeluk CS-ESP 0.1 17.4

Matla Matla ESP + FGD (Eskom, 2001) 0.5 15.5

Tutuka New Denmark CS-ESP 0.1 4.9


Total Product Consumed / Produced by Major Industries

during 2004

Source CategoryProduct Consumed

/ Produced(Mtonnes / year)

Total Product Consumed / Produced

(Mtonnes / year)

Coal-fired Power Plants 112.20

Coal Gasification 41.02

Consumer Products 27.33

Solid Waste (Landfills) 20.00

Crude Oil Refining 18.10

Ferrous Metals: Iron and Steel: 13.62

- Coke Production / Alloy Steel 7.62

- Pig Iron 6.00

Residential Heating 5.00

Cement Production 1.95

Non-Ferrous Metals: Primary Metals: 0.63

- Gold 0.0003

- Zinc 0.24

- Copper 0.35

- Lead 0.04

Incineration 0.03

Total 239.88


2D G raph 2

Source C ategory

Power P

lants

Cement P

roductio

n

Coal Gasifi

cation

Residentia

l Heatin

g

Incin

eratio

n

Primary

Meta

l Pro

duction

Iron a

nd Ste

el

Coke Pro

duction

Min

erals

C rude O

il Refin

ing

Solid W

aste (L

andfills)

Consumer P

roducts

To

tal

Hg

(to

nn

es

/ y

ea

r)

0

1

2

3

6

9

12

Total Hg Emissions by Major Industries during 2004

Total Hg emissions: 20 tonnes


Coal Combustion: Power Plants

Estimated 9.75 tonnes of Hg was emitted from coal-fired power plants during 2004.

Hg emission estimate based on:

• control devices: electrostatic precipitators, fabric filters to flue-gas desulphurization

methods (Bateman, 1998; Eskom, 2001 and 2006);

• corresponding emission distribution factors (0.5 – 0.9; UNEP, 2005);

• average Hg level of 0.15ppm for Highveld coals in Mpumalanga Province (Wagner and

Hlatswayo, 2005); and

• amount of coal consumed by Eskom (112 Mtonnes).

Note: Highveld Coalfield is the second largest coal field for coal production in RSA.

Hg emissions are substantially higher than all other sectors, particularly other coal combustion

processes (e.g. residential heating, iron and steel processing).

With South Africa’s increasing demand for energy, the commissioning of new coal-fired power

plants and the de-mothballing of existing coal-fired power plants (MRA, 2003), Hg emissions to

the environment will inevitably increase over the next decade.


Coal Combustion: Gasification

Sasol converts low-grade coal into petroleum products (IEA, 2007): via the Fischer

Tropsch process to produce about 30% of RSAs liquid fuel requirements (Van Dyk et al.,

2006)

Estimated that Sasol’s process produced about 1.88 tonnes of Hg (air + waste) during

2004 :

- amount of coal consumed (41.2 Mtonnes: 70% gasified & 30% process steam / electricity),

- 0.15ppm of Hg in coal (Wagner and Hlatswayo, 2005),

- fabric filter control devices in the coal-fired power plants (0.5 emission distribution factor).

If much Hg is trapped, then emissions would be substantially lower: ~ 0.18 tonnes Hg

would be released to the atmosphere if one were to assume that about 90% of the

estimated 1.88 tonnes of Hg is trapped during coal gasification.


Ferrous Metal Production: Iron and Steel

Major source of Hg emissions from this activity is from coke production (Pacyna et al., 2006). Small

amount of coal is used in the iron and steel industry in South Africa.

Coke (33%) and Steel Production / Scrap Smelting (66%) estimated to emit ~1.0 tonnes of Hg:

- combustion of ~ 7.62 Mtonnes of coal

- 0.15 ppm Hg in coal (Wagner and Hlatswayo, 2005)

- emission distribution factors (0.9; UNEP 2005)

Pig iron production estimated to emit ~0.29 tonnes of Hg:

-production of 6 Mtonnes of coal

- 0.15ppm Hg in coal (Wagner and Hlatswayo, 2000)

- emission distribution factors (0.95; UNEP 2005)

The estimated Hg emissions are much higher than Hg emissions reported for the African continent

in 2000 (0.4 tonnes; Pacyna et al., 2006).

South Africa is experiencing industrial growth, particularly in this sector of the economy and Hg

emissions from this source are expected to increase.


Coal Combustion: Residential Heating

Informal settlements and rural area households in South Africa use coal for heating and

cooking (Spalding-Fecher and Matibe, 2003).

Residential heating estimated to emit ~0.75 tonnes of Hg:

- combustion of ~5 Mtonnes of coal during 2004 (DME, 2005)

- no emission control devices (1),

- 0.15 ppm Hg in in Highveld coal (Wagner and Hlatswayo, 2005).

Since about 90% of Hg emitted from this source is gaseous Hg0 and Hg2+ (Pacyna et al.,

2003), individuals would be directly exposed to about 0.66 tonnes of Hg during heating or

cooking in their homes.

Increased Hg emissions, concomitant with an increase in coal burning will occur during

winter, similar to what has been reported in China (Wang et al., 2006).


Crude Oil Refining

South Africa is not a major oil producer, but has the second largest oil refining capacity in

Africa, following Egypt (EIA, 2007).

Mercury levels in crude oils vary substantially depending on its origin.

> 80 % of South Africa’s crude oil is imported from the Middle East (Iran and Saudi Arabia;

EIA, 2007), Hg emissions from this source would not vary much.

Crude oil refining emitted ~0.16 tonnes of Hg:

- RSA oil refineries processed approximately 18 Mtonnes of crude oil (DME, 2005a);

- Total Hg in crude oil used in refining process: 10 mg/tonne (UNEP 2005);

- Emission distribution factor (0.9; UNEP 2005).

As South Africa’s oil consumption has increased steadily during 1986 – 2006 (EIA, 2007),

with no signs of slowing in the near future, Hg emissions from crude oil refining are likely

to increase, with increasing demand in South Africa, in the future.


Non-Ferrous Metals Production: Primary Metals

Recent Hg emissions for South African gold production was estimated to be 174.1 tonnes during

2000 (Pacyna et al., 2006). Gold production uses cyanidation and not Hg amalgamation to extract

gold from ore.

Gold production estimated to emit ~0.07 tonnes of Hg:

- 255 tonnes of gold produced (CoMSA, 2007),

- Hg emissions range between 103 – 444 g Hg per tonne gold produced (Schröder et al.,1982)

- emission distribution factor (1)

Hg emissions from gold production in South Africa appear to be lower than in other gold producing

regions of the world (e.g. Nevada have Hg levels as high as 0.1% by weight, and is the largest Hg

by-product source in the USA; Jones and Miller, 2005).

Production of other primary (virgin) metals estimated to emit a total of ~0.32 tonnes of Hg from:

- 0.24 tonnes zinc (Kumba Resources, 2005) at 10 g Hg/tonne;

- 103 Ktonnes copper (DME, 2006) at 1 g Hg/tonne;

- 42 Ktonnes lead (DME, 2006) at 10g Hg/tonne;

- emission distribution factor: 0.1 (UNEP 2005).


Cement Production

Coal for firing cement kilns and producing clinker are the major sources of Hg in cement production.

Pacyna et al. (2006) reported that Hg emissions from cement production in Africa were 5.3 tonnes

in 2000.

The annual coal consumption was estimated by using annual cement production data, taking

account that ~15 tonnes of coal is burned in order to produce 100 tonnes of cement clinker (DME,

2005b).

Approximately 3.77 tonnes of Hg estimated to be released during 2004 (includes Hg emissions

from coal combustion for firing cement kilns) based on:

- 12.98 Mtonnes cement produced in 2005 = ~1.95 Mtonnes coal used for firing cement kilns;

- using appropriate distribution factors (0.9; UNEP, 2005);

- ~0.15 ppm of Hg in coal (Wagner and Hltaswayo, 2005).

Cement production is increasing as new infrastructure is required in RSA, and so Hg emissions for

the country and the continent will increase.

It is important that Hg emissions from this source be monitored in order to provide more robust Hg

emission assessments for South Africa.


Artisanal Gold Mining

Although artisanal gold miners operate in the Mpumalanga and Limpopo Provinces

(CoMSA, 2006), the extent to which artisanal and small-scale gold mining activities

contribute to Hg emissions in South Africa are virtually unknown.

This activity is illegal in South Africa, and has an estimated 8 000 to 20 000 gold miners

operating in the country. Mercury emissions from artisanal gold mining in other African

countries are low (e.g. 3 to 5 tonnes per year in Zimbabwe; Veiga, 2004).

For South Africa, the Hg emissions from these activities may even lower, considering the

estimated number of gold miners in the country is lower than that found in neighbouring

Zimbabwe.


Waste Deposition and Incineration

About 95% of waste was disposed of in landfills prior to 2000 (DWAF, 1998). Total Hg levels in

landfill gas have not been measured in RSA.

Coal ash generated by Sasol’s coal gasification process is also primarily landfilled. Coal

consumption of 41 Mtonnes in 2004 (DME, 2005), of which 70% is consumed in coal gasification

equates to approximately 3.87 and 0.43 tonnes of Hg being trapped as gaseous Hg0 and waste

(coal ash), respectively.

South Africa’s National Waste Management Strategy: medical waste be sorted prior to disposal

or incineration; and disposal of potentially hazardous medical waste to landfills be avoided.

Medical waste estimated to emit ~0.60 tonnes of Hg:

- authorised medical waste treatment capacity (commercial services; public / private hospitals) was

~0.03 Mtonnes in 2005 (DEAT, 2006b);

- emission distribution factor (1).

Poor on-site incinerators in public hospitals or clinics, if present; and the burning or illegal dumping

of waste in residential areas will likely increase Hg emissions to the South African environment.


Monitoring Hg emissions in RSA

Atmospheric monitoring of Hg mostly made at Cape Point’s Global Atmospheric Watch (GAW) Station:

- Since 1995 at the Cape Point GAW Station (Baker et al., 2002; Slemr et al., 2006):

- Average yearly total gaseous Hg between 1995 and 2004 ranged between 1 and 1.5 ng m-3

- Similar to measurements on board ship in the South Atlantic, and only slightly elevated compared to those

measured at Neumayer on the Antarctic Peninsula (Baker et al., 2002; Slemr et al., 2006).

Initial atmospheric Hg studies in Pretoria (Gauteng Province):

-Total gaseous Hg are elevated about 2 ng m-3, especially during the day and indicative of a diurnal trend (low

concentrations at night).

- Concentrations and variations are consistent with substantial local and regional sources in the vicinity, and are

not surprising given the high level of industrial activity in Gauteng Province.

Rainfall collections (weekly bulk phase) at Cape Point’s GAW station:

- Average rainfall concentration for seven weeks of weekly rain sampling (July-/August 2007) was 6.3 ± 3.0 ng/L.

- Scaling these data to a yearly flux suggests that wet deposition is around 3 μg m-2 yr-1;

Consistent with estimates of Mason et al. (1994) for the remote southern Hemisphere.

Lower than that of similar locations in the North Atlantic (Bermuda, for example (8 μg m-2 yr-1), or most locations

on the east coast of the USA (e.g. Mason et al., 2000).

Rain collections in Pretoria in Gauteng Province are now underway and contrasting the concentration ranges

between Pretoria and Cape Point will provide important information on the extent of anthropogenic emissions in

South Africa.


Gaps in our Current Understanding

The above-mentioned industries are all important in South Africa and further monitoring

and research is required to verify the reported Hg estimates.

Biomass burning as a potential source of Hg to the South African environment needs to be

explored. Measurements of total gaseous Hg at Cape Point, downwind of a fire on the

Cape Peninsula, suggested that biomass burning could be a substantial source of Hg in

the southern Hemisphere (Brunke et al., 2001). The impacts of Hg from these sources

have also not been characterised in South Africa.

The importance of determining Hg levels in artisanal gold mining areas in South Africa are

needed to evaluate the potential impacts that such activities may have on human health

and surrounding aquatic ecosystems.

Although these are potential sources of Hg, no information is available on the Hg content

of their emissions. More so, the fate and/or life cycle of gaseous Hg through the entire

electricity generation process require further investigation.


Research Needs

All estimates included in this study are based on the best available

information.

While data on the Hg content of coal exists, this is only for the Highveld

Coalfield and the Hg content of coal and emissions resulting from its use

are likely to vary throughout the country.

Major gaps in our understanding of point source Hg emissions include the

Hg content of raw materials used in industry (i.e. in iron and base metal

ores, limestone in the cement industry, etc.) and the type and efficiency of

control devices used in various industrial sectors.

mercury emissions from point sources in south africa

Documents