UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mercury and artisanal and small-scale gold miners in China Gunson, Aaron James 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2004-0224.pdf [ 17.51MB ]
Metadata
JSON: 831-1.0081074.json
JSON-LD: 831-1.0081074-ld.json
RDF/XML (Pretty): 831-1.0081074-rdf.xml
RDF/JSON: 831-1.0081074-rdf.json
Turtle: 831-1.0081074-turtle.txt
N-Triples: 831-1.0081074-rdf-ntriples.txt
Original Record: 831-1.0081074-source.json
Full Text
831-1.0081074-fulltext.txt
Citation
831-1.0081074.ris

Full Text

Abstract This thesis determines that over four hundred thousand artisanal and small-scale gold miners all over China systematically use amalgamation and release approximately 240 tonnes of mercury per annum. Artisanal and small-scale gold mining ( A S G M ) is one of the largest sources of mercury pollution in China. This thesis outlines an approach to investigating A S G M in China based on a series of three studies. The main findings of the thesis were that A S G M and amalgamation are widespread in China, and that mercury releases from A S G M have serious health and environmental impacts on the miners themselves and on their surrounding communities and environment. The use of whole ore amalgamation was demonstrated to be the primary source of these releases, and Muller mi l l amalgamation was pinpointed as being responsible for over 70% of the A S G M mercury releases. In addition, the practice of following amalgamation with cyanidation was indicated to increase the solubility of mercury and perhaps further increase its risks. The test work completed on an ore sample from the community of "Go ld Mountain" demonstrated that alternative technology could largely replace whole ore amalgamation (although not cyanidation) with gravity methods and thus greatly mitigate the health and environmental impacts of A S G M . ii Table of Contents A B S T R A C T n T A B L E OF CONTENTS HI LIST OF T A B L E S vm LIST OF FIGURES ix LIST OF A B B R E V I A T I O N S x A C K N O W L E D G E M E N T S xu 1 Introduction 1 1.1 S T A T E M E N T OF P R O B L E M 1 1.2 B A C K G R O U N D 2 1.2.1 Artisanal and Small-scale Mining 2 1.2.1.1 Definition 2 1.2.1.2 Extent and Impact 2 1.2.1.3 Health, Safety and Environment 3 1.2.1.4 Gender and Social Issues 4 1.2.1.5 Sustainable Livelihoods & Communities 4 1.2.2 Mercury as a Global Pollutant 5 1.2.3 Mercury as a Local Health Concern for ASGM. 8 1.2.3.1 Mercury Vapour 8 1.2.3.2 Methylmercury 10 1.3 O U T L I N E OF THESIS 12 2 Study One: Artisanal & Small-scale Gold Mining in China 14 2.1 INTRODUCTION '• 14 2.1.1 Literature Review 14 2.2 M E T H O D O L O G Y 18 2.2.1 Site Visits IS 2.2.2 Interviews 19 2.2.3 HgEXExpert System 20 2.3 R E S U L T S 22 2.3.1 Site Visits - Gold Processing Equipment and Methods in China 22 2.3.1.1 Muller Mills 22 iii 2.3.1.2 Ba l l M i l l / Sluice 24 2.3.1.3 Gold Dredges and Placer Min ing 25 2.3.1.4 Retorts 27 2.3.1.5 Cyanidation & Zinc Strip Smelting 28 2.3.1.6 Gold Shops 29 2.3.2 Interviews 30 2.3.3 HgEXExpert System 31 2.3.3.1 Alpha Factor Determination for China 31 2.3.3.2 Bayuan Site Contamination 38 2.4 DISCUSSION - A S G M IN C H I N A 40 2.4.1 Significance of ASM in China 40 2.4.2 Scale of ASGM in China 41 2.4.2.1 Definition 41 2.4.2.2 Scale of A S G M in China 44 2.4.3 Economic Impact of ASGM 47 2.4.3.1 National Impact 47 2.4.3.2 Local Impact 48 2.4.3.3 Impact on Formal Min ing Companies 48 2.4.4 Legal Framework. 50 2AAA Government Framework 50 2.4.4.2 Government Policy 50 2.4.4.3 Laws and Enforcement 51 2.4.4.4 Tax 53 2.4.4.5 Illegal A S G M 53 2.4.5 Social Context: ASGM in the Community 54 2.4.5.1 A S G M Labourers 54 2.4.5.2 A S G M Income 55 2.4.5.3 Women 55 2.4.5.4 Children 56 2.4.5.5 Labour Unions 56 2.4.6 Health, Safety, and Environmental Impact 56 2.4.7 Technical Assistance Programs 58 2.4.7.1 Technical Assistance 58 iv 2 . 4 . 7 . 2 Research Cooperation 5 8 2 . 4 . 7 . 3 Domestic Financial Institutions 5 9 2 . 4 . 7 . 4 Non-Governmental Organizations 5 9 2 . 4 . 7 . 5 Multilateral Organizations 5 9 2 .5 C O N C L U S I O N 6 0 3 Study Two: Artisanal and Small-scale Gold Miners in "Gold Mountain" 61 3.1 INTRODUCTION 61 3 .2 " G O L D M O U N T A I N " - C A S E S T U D Y M E T H O D O L O G Y 6 1 3.3 " G O L D M O U N T A I N " C A S E S T U D Y - R E S U L T S 6 2 3.3.1 "Gold Mountain " Mining and Processing Practice 62 3.3 .1 .1 Min ing Practice 6 3 3 . 3 . 1 . 2 Comminution and Amalgamation 6 3 3 .3 .1 .3 Retorts 6 5 3 .3 .1 .4 Cyanidation and Zinc Smelting 6 7 3 .3 .1 .5 Go ld Purification 6 8 3 . 3 . 1 . 6 "Go ld Mountain" A S G M Cost Structure 6 8 3.3.2 "Gold Mountain " Mercury Releases and Sample Collection 70 3.3 .2 .1 M i l l Operator Estimates on Mercury Lost 7 0 3 . 3 . 2 . 2 M i l l Data on Mercury Lost 7 0 3 .3 .2 .3 Sample Collection Data on Mercury Lost 71 3 .3 .2 .4 A S G M Mercury Release Estimates 71 3.3.3 ASGM Health and Environmental Impact 72 3.3 .3 .1 Health Impact 7 2 3 . 3 . 3 . 2 "Gold Mountain" Aquaculture 7 3 3 .4 DISCUSSION 7 4 3.4.1 "Gold Mountain's " Processing Practice 74 3.4.2 Health and Environmental Impact of Mercury Releases in "Gold Mountain" ... 74 3.4 .2 .1 Health Impact of Mercury Vapour 7 4 3 . 4 . 2 . 2 Bioaccumulation 7 5 3.5 C O N C L U S I O N 7 6 4 Study Three: "Gold Mountain" Alternative Technology Test Program 7 7 4.1 INTRODUCTION 7 7 V 4.1.1 Sample GM1 - "Gold Mountain " Feed Ore 77 4.1.2 Sample GM2 - "Gold Mountain " Muller Mill Tails 77 4.1.3 Sample GM3 - "Gold Mountain " Cyanidation Tails 77 4.2 METHODOLOGY 78 4.2.1 Sample GM1 - "Gold Mountain " Feed 78 4.2.1.1 Sample Preparation and Characterization 79 4.2.1.2 Gravity Recovery Test 79 4.2.1.3 Gravity / Cyanidation Tests 80 4.2.2 Sample GM2 - "Gold Mountain " Muller Mill Tails 82 4.2.2.1 Sample Characterization 82 4.2.2.2 Gravity Recovery Test 83 4.2.2.3 Mercury Solubility Test 83 4.2.3 Sample GM3 - "GoldMountain" Cyanidation Tails 84 4.2.3.1 Sample Characterization ". 84 4.2.3.2 Gravity Recovery Test 84 4.2.3.3 Mercury Solubility Test 85 4.3 RESULTS 85 4.3.1 Sample GM1 - "Gold Mountain " Feed 85 4.3.1.1 Sample Characterization 85 4.3.1.2 Gravity Recovery Test 85 4.3.1.3 Gravity / Cyanidation Tests 86 4.3.2 Sample GM2 - "Gold Mountain " Muller Mill Tails 89 4.3.2.1 Sample Characterization 89 4.3.2.2 Gravity Recovery Test 90 4.3.2.3 Mercury Solubility Test 91 4.3.3 Sample GM3 - "Gold Mountain " Cyanidation Tails 91 4.3.3.1 S ample Characterization 91 4.3.3.2 Gravity Recovery Test 92 4.3.3.3 Mercury Solubility Test 93 4.4 DISCUSSION 93 4.4.1 "Gold Mountain " Samples 93 4.4.1.1 G M 1 94 4.4.1.2 G M 2 95 vi 4 . 4 . 1 . 3 G M 3 9 5 4.4.2 Mercury Solubility 96 4 .5 C O N C L U S I O N 9 7 5 Mercury Releases, Alternatives, and Implementation 98 5.1 C H I N A ' S M E R C U R Y R E L E A S E S 9 8 5.1.1 Estimating China's ASGM Mercury Releases 98 5.1.2 Other Sources of Mercury 101 5.1.3 Comparing Mercury Releases in China 103 5.2 A L T E R N A T I V E T E C H N O L O G Y 1 0 4 5.2.1 Sustainable Alternative ASGM Technology 107 5.3 T E C H N O L O G Y T R A N S F E R 1 1 2 5.3.1 ASGM Processing Centres 113 5.4 A S G M A N D P O L I C Y 1 1 6 6 Limitations and Bias . 117 7 Areas for Additional Research 120 8 Conclusion 122 Bibliography 124 Appendix 1 134 vii List of Tables Table 1.1 A S M Production of Various Minerals and Metals 3 Table 2.1 Site Visits , 2000-2002 18 Table 2.2 Chinese Productions of Several Major Minerals 41 Table 2.3 Major health, safety, and environmental impacts of A S M in China 57 Table 3.1 Typical Miner 's Revenue and Profits Per Tonne of Ore 69 Table 3.2 "Go ld Mountain" Processing Centre Data 70 Table 3.3 Muller M i l l Mercury Loss Estimates 71 Table 4.1 G M 1 Sample Characterization 85 Table 4.2 G M 1 - A Gravity Recoverable Gold Test 85 Table 4.3 G M 1 - B Test Results 86 Table 4.4 G M 1 - C Test Results 86 Table 4.5 G M 1 - D Test Results .' 87 Table 4.6 G M 1 - B and G M 1 - C Cyanidation Tailings Analysis 88 Table 4.7 G M 2 Sample Characterization 90 Table 4.8 G M 2 Gravity Test Results 90 Table 4.9 G M 3 Sample Characterization 92 Table 4.10 G M 3 Gravity Test Results 92 Table 4.11 Mercury Solubility 93 Table 4.12 Grade Differences Between Test Samples 93 Table 5.1 Mercury to Gold Loss Ratio for China 100 Table 5.2 Chinese Annual Mercury Releases attributable to A S G M 100 Table 5.3 L o w Estimate of A S G M Gold Production and Mercury Releases (2001) 101 Table 5.4 Chinese Sources of Mercury Releases Due to Min ing (2001) 103 viii List of Figures Figure 1.1 Number of Artisanal & Small-scale Miners Around the World by Country 7 Figure 1.2 Eh (redox potential) versus pH for the Main Inorganic Mercury Species 11 Figure 2.1 Alpha Factor for HgEX 21 Figure 2.2 "Gold Mountain" Muller Mill 23 Figure 2.3 Jiangxi Ball-Mill Mercury Sluice Circuit 25 Figure 2.4 Guangxi Gravel/Gold Dredge Operation , 26 Figure 2.5 "Gold Mountain" Gravel/Gold Operation 27 Figure 2.6 Jiangxi Province Retort Cooling Tube 28 Figure 2.7 "Gold Mountain" Zinc Strip Boxes 29 Figure 2.8 Burning Dore in Longsheng 30 Figure 2.9 Alpha Factor Comparison 37 Figure 2.10 ASGM Areas Across China 44 Figure 2.11 ASGM Tunnels, Boka, China 49 Figure 3.1 Cleaning the Concentrate From the Muller Mill 64 Figure 3.2 "Gold Mountain" Retort Diagram 66 Figure 3.3 "Gold Mountain" Retort Photo 66 Figure 3.4 Mercury Droplets from "Gold Mountain" Retort 67 Figure 4.1 GM1 Test Program 78 Figure 4.2 GM2 Test Program 82 Figure 4.3 GM3 Test Program . 84 Figure 4.4 GM 1-D Leach Time 87 Figure 4.5 GM1-B and GM1-C Cyanidation Tailings Analysis 88 Figure 4.6 GM2 Size, Gold, and Mercury Distributions 89 Figure 4.7 GM3 Size, Gold, and Mercury Distributions 91 Figure 5.1 China's Mercury Releases Due to Mining 103 Figure 5.2 Gold Particle Size Recovery Range 108 Figure 5.3 Muller Mill / Centrifugal Concentrator, Chile 111 ix List of Abbreviations A S G M Artisanal and Small-scale Gold Miners A S G M A u China's annual gold production due to A S G M A S M Artisanal and Small-scale Min ing A T Appropriate Technology B B C British Broadcasting Corporation B C - M E M P R British Columbia - Ministry of Energy, Mines and Petroleum Resources B S A u % The percentage of gold produced by A S G M ball mill/sluice operations B S R The ratio of mercury lost to gold recovered for ball mill/sluice operations C F Certainty Factor C A S M Communities and Small-Scale Mining C A S M - C h i n a Communities and Small-Scale Mining Regional Network in China C C C Criterion Continuous Concentration C I A Central Intelligence Agency C I D A Canadian International Development Agency C M C Criteria Maximum Concentration D F A I T Department for Foreign Affairs and International Trade DFID Department for International Development (United Kingdom) D o B D E F Degree of Bel ief - Dangerous Environmental Factor D o B H E F Degree of Bel ief - High Emission Factor D o B M A F Degree of Bel ief - Mercury Adsorption Factor E S A T Environmentally Sound and Appropriate Technology G M 1 Gold Mountain Feed Ore Sample G M 2 Gold Mountain Muller M i l l Tails Sample G M 3 Gold Mountain Cyanidation Tails Sample G M P Global Mercury Project G R G Gravity Recoverable Gold H g A u The amount of mercury released per annum due to A S G M activities H g E X Expert System for Risk Assessment for Mercury Discharge from Gold Mining Operations I C P - M S Inductively Coupled Plasma - Mass Spectrometry IIED International Institute of Environment and Development ILO International Labour Organization I T D G Intermediate Technology Development Group K C - M D 3 Knelson Concentrators lab-scale M D 3 centrifugal concentrator M M A u % The percentage of gold produced by A S G M operations using muller mills M M R The ratio of mercury lost to gold recovered for muller mills M M S D Mining , Minerals, and Sustainable Development N G O Non-Governmental Organization N I M D National Institute for Minamata Disease N M A Z National Miners' Association of Zimbabwe N S E R C National Science and Engineering Research Council N R C A N Natural Resources Canada P B C People's Bank of China P M A u % The percentage of gold produced by A S G M placer operations P M R The ratio of mercury lost to gold recovered from placer deposits P R C People's Republic of China R M B Renminbi (Chinese Dollar or Yuan) S C O P E Scientific Committee on Problems of the Environment S W G Southwestern Resources T D U Transportable Demonstration Units T O C Total Organic Carbon T V E Township and Village Enterprises U B C The University of British Columbia U N E C A Unit of Gold Extraction and Controlled Amalgamation U N I D O United Nations Industrial Development Organization U N D E S A United Nations Department of Economic and Social Affairs U S E P A United States Environmental Protection Agency W H O World Health Organization Y P N I Yunnan Province Nuclear Industry Team 209 xi Acknowledgements Part of this research was made possible by the support of the World Business Council for Sustainable Development, through a report written for the International Institute of Environment and Development's (IIED) Mining, Minerals and Sustainable Development (MMSD) project, by the National Science and Engineering Research Council (NSERC) Industrial Postgraduate Scholarship in cooperation with Knelson Concentrators, and UBC's Bridge Program. Special thanks to: Dr. Marcello Veiga, my primary advisor, and Dr. Bern Klein, my other primary advisor, Dr. Chris van Netten, and Dr. Yin Yongyuan, Dr. Lin Yuhuan, Dr. Shen Lei, Dr. Peter Golas, and Dr. Philip Andrews-Speed, Yue Jian, for his invaluable help and friendship, The UBC Sustainability Working Group, especially Jenn Hinton, for advice and help on finding references, Edmund Bugnosen, Anne-Marie Fleury and the MMSD group, Jon Coates with BHP Billiton in China, Ish Grewal, Linda Duncan, Jim Zhang, and Dennis Grimm at Knelson Dr. Kay Teschke and the Bridge Program, and Jeffrey Davidson and Gotthard Walser, at Communities and Small-Scale Mining (CASM). Many thanks to others who helped me out in China, but who would rather not have their names recorded, especially the miners of "Gold Mountain." Finally, special thanks to my wife, Melania Cannon, for all her assistance and patience. xii 1 Introduction This thesis determines that artisanal and small-scale gold mining ( A S G M ) in the People's Republic of China (PRC) employs at least 400,000 miners extracting about 60 tonnes of gold per annum and releases around 240 tonnes per annum of mercury into China's environment. This thesis focuses on describing China's artisanal and small-scale gold miners, estimating their mercury releases, and exploring technological alternatives to reduce these releases. Mercury is one of the most toxic heavy metals, and mercury pollution has led to many tragic incidents around the world. A S G M is a major source of global mercury releases; the health impact of this mercury is of significant concern, both locally and globally. This thesis grew out of an interest to understand China's role in these releases and to learn how A S G M in China differed from A S G M in other regions of the world. 1.1 Statement of Problem A n exploratory literature review found only one article regarding A S G M in China; that article stated that most A S G M has been strictly prohibited since 1996 (Lin et al, 1997). However, a handful o f global reviews o f A S G M prepared by multilateral agencies such as the International Labour Organization (ILO), indicated that China was home to millions of artisanal and small-scale miners. This thesis grows out of an initial research trip to China, conducted in 2000, undertaken to examine the situation regarding A S G M in China. Specifically, this thesis aimed to address the two following questions: 1) Does A S G M still occur in China, and i f so, at what scale and in what socio-economic context? 2) Does A S G M in China use mercury, and i f so how much? H o w does A S G M in China use mercury, and are there methods o f dealing with the resulting environmental and health problems applicable to other parts of the world, or can techniques used in other nations be applicable to China? 1 1.2 Background Artisanal and small-scale gold mining is closely linked with the global emission of mercury. This link is of importance on a number of levels. Locally, A S G M mercury use is a serious health risk faced by millions of artisanal and small-scale miners and their communities around the world. Non-mining communities located downstream from mercury-using A S G M operations are also placed at risk. On a global level, A S G M contributes to mercury releases that may be linked to the atmospheric deposition of mercury around the world. This section reviews artisanal and small-scale gold mining and mercury use in a global context. 1.2.1 Artisanal and Small-scale Mining 1.2.1.1 Definition Artisanal or small-scale mining ( A S M ) is the use of rudimentary processes to extract valuable minerals from primary and secondary ore bodies, and is characterized by the lack of long-term mine planning/control (Hinton et al, 2003a). A S M can be illegal or legal, formal or informal and can encompass everything from individual gold-panners to large-scale operations employing thousands of people. 1.2.1.2 Extent and Impact Artisanal and small-scale mining operations form the backbone of hundreds of developing rural communities around the world and supplement the income of thousands of agricultural communities. In an authoritative ILO study on A S M , Jennings (1999) estimates that there are 11 to 13 mil l ion artisanal miners worldwide, with 80 to 100 mil l ion people directly or indirectly dependent on artisanal miners for their livelihood. A S M extracts a wide range of metals, precious stones, and industrial minerals, and accounts for a significant portion of world production, as can be seen in Table 1.1. 2 Table 1.1 A S M Production of Various Minerals and Metals (Noetstaller, 1995) Metals Mined by Industrial Mined by A S M (%) Minerals A S M (%) Beryll ium 100 Fluorite 90 Mercury 90 Graphite 90 Tungsten 80 Talc 90 Chromium 50 Vermiculite 90 Antimony 45 Pumice 90 Manganese 18 Feldspar 80 Tin 15 Clay 75 Iron 12 Gypsum 70 Lead 11 Barite 60 Zinc 11 Sand and gravel 30 Cobalt 10 Dimension stones 30 Gold 10 Salt 20 Silver 10 Coal 20 Copper 8 Phosphate 10 1.2.1.3 Health. Safety and Environment The employment, income and production generated from A S M often come with significant costs to miners' and nearby communities' health, safety, and environment. A S M hazards are many and varied. In A S M mines, hazards include: poor ground conditions leading to underground failure, methane or coal dust explosions from coal mines, flooding, machinery accidents, poor lighting and ventilation, explosives accidents, and electrocution (Priester et al, 1993). Hydraulic monitoring of secondary deposits can also be extremely unsafe, as a potential exists to undercut h i l l slopes and generate landslides (Hinton et al, 2003a). Mercury, cyanide and other hazardous chemicals are used as reagents for recovering and purifying gold and other precious metals. Fine dust from mineral processing, leading to silicosis, and noise pollution are endemic problems. A S M accidents are often underreported, due to the illegal or semi-legal status of most A S M operations; Jennings 3 (1999) estimates that non-fatal accidents in A S M are still 6-7 times greater than in the formal mining sector. A S M operations can also lead to the destruction of arable and grazing land through excavation, the accelerated erosion of top soils, landslides, the collapse of old workings, unsafe tailings disposal, the lowering of water tables, soil contamination due to dust from mines and tailings, and increased levels of sediment load and flooding in nearby rivers. 1.2.1.4 Gender and Social Issues Perhaps 30% of the A S M miners around the world are women, involved at all stages of mining from ore extraction to processing to market (Hinton et al, 2003b). Women also frequently work part-time as cooks and service providers, or in transporting and processing, and thus the actual number of women involved in artisanal mining could be significantly higher than generally estimated. Processing activities are often conducted in homes by women of childbearing age who are especially susceptible to mercury poisoning (Hinton et al, 2003a). In addition, A S M , especially in the boomtowns that arise from big mineral finds in remote locations, may come with major social challenges such as prostitution, substance abuse, and gambling. Women are often disproportionately harmed by these factors. The roles of women in A S M are significantly different from those of men and thus deserve, but generally do not receive, special attention with regard to policy, development, and research (Hinton et al, 2003b). 1.2.1.5 Sustainable Livelihoods & Communities With the range of i l l effects that arise from A S M , governments often see closing these often-illegal mines as a desirable end. However, A S M frequently provides impoverished remote communities with a rare opportunity to alleviate poverty and contribute to community development (Dahlberg, 1997). A l l ore deposits are finite, thus by definition no single mining operation can be sustainable. Nevertheless, A S M can contribute to the sustainable development of communities by providing opportunities for economic diversification through auxiliary enterprises, such as jewellery production and agricultural 4 development, and by providing enough economic impetus to allow impoverished communities to move toward sustainability. International organizations, such as Communities and Small-scale Min ing ( C A S M ) and the United Nations Department of Economic and Social Affairs ( D E S A ) are focusing on promoting a sustainable communities approach to A S M . A n underlying premise of this thesis is that A S M must be looked at in the context of sustainable livelihoods and communities in order to most effectively mitigate the impact of A S M activities while maximizing its benefits to current and future generations. 1.2.2 Mercury as a Global Pollutant Mercury is a global pollutant due to both its impact through actions such as the disposition of atmospheric mercury in the Arctic, and its local impact in communities worldwide. Present concern about the health impact of mercury can be traced back to the methylmercury poisoning disaster in Minamata, Japan, in the 1950s ( N I M D , 2001). The Minamata outbreak, followed by another incident in Niigata, Japan, both occurred due to the consumption of large quantities of seafood polluted by mercury-contaminated industrial effluent. Both culminated in several hundred deaths and claimed hundreds of more victims with severe mercury induced symptoms (Harada, 1978). These cases created widespread awareness about the i l l effects of mercury and prompted much research on the topic (Horvat, 2002). The global importance of mercury became increasingly evident due to a number of further issues. Mercury-based fungicides and seed disinfectants were found to have caused severe reductions of bird populations (Jernelov et al, 1975). In 1971-2, a large epidemic that kil led over 450 people and may have disabled over 100,000 people occurred in Iraq when farmers consumed foreign aid supplied wheat, intended for seed, treated with alkyl mercury fungicide (Forstner and Wittman, 1997). Unsafe levels of mercury accumulation were found in fish and wildlife in Canada, the northern United States, and Scandinavia during the 1960s (Williamson, 1986). Fish in reservoirs in Quebec and Manitoba were found to have elevated levels of mercury, perhaps due to forest fires or organic material in newly flooded areas (Williamson, 1986; Stokes and Wren, 1987). Mercury vapour poisoning kil led 15 miners, with 205 more miners admitted 5 to hospital, in a mercury mine in Keniya, Turkey (Trakhtenberg, 1974). A report from 1992 estimated that in the previous 40 years, over 20,000 people were afflicted by mercury poisoning, including 1,400 deaths (Larcerda, 1997). This rising concern led to increasing efforts to understand global mercury releases and to create regulations to control its trade and use. First implemented in the 1970s, these regulations have steadily reduced global releases of mercury, primarily through banning mercury-based fungicides and mercury-cell chlor-alkali plants and through the recycling of mercury from solid wastes (Larcerda, 1997). The sale of fish containing mercury levels above 0.5 mg/kg was banned in the 1970s. In the mid to late 1990s, atmospheric deposition of mercury in the Arctic due to both natural and anthropogenic sources became an issue of increased concern (Lu et al., 2001). Anthropogenic releases of mercury originate primarily from: coal and oi l combustion, refuse incineration, solid and urban waste disposal, and agriculture and forestry practices. Anthropogenic releases are four times higher than natural mercury releases (Larcerda, 1997). Pacyna and Pacyna (2002) estimate that the 1995 anthropogenic global mercury emission was about 1900 tonnes, with three quarters of the total emission coming from combustion of fuels, particularly coal combustion in China, India, and South and North Korea. A European Commission Report estimates the current annual market supply of mercury at 3800 tonnes (Maxson, 2004). In the past, mercury mining has been the major source of mercury; Hylander and M e i l i (2003) estimate around 1 mil l ion tonnes of mercury have been mined globally since 1500. A major source of mercury releases has been A S G M . Maxson (2004) estimates total global mercury releases due to A S G M in 2000 to be 650 tonnes, 20% of the total global anthropogenic amount of 3386 tonnes. Mercury has been used to amalgamate precious metals since as early as 2,700 B C , when the Almaden mines in Spain were opened (Lacerda, 1997). Amalgamation is a simple and inexpensive method of recovering gold. The wetting of gold by mercury is not alloying, but a phenomenon of moderately deep sorption, involving some interpenetration of the two elements (Pryor, 1965). A s the 6 surface tension of mercury is greater than that of water, but less than that of gold, mercury adsorbs onto the surface of gold particles. In addition, mercury acts as a dense medium; gold sinks into the mercury while the lighter gangue material floats overtop. The solubility of gold in mercury is only 0.06% at 20°C. When the resulting amalgam is heated, the mercury vaporises, leaving gold dore. The practice of using mercury to extract gold had largely faded in the early twentieth century, until the gold rushes in Latin America began during the 1970s and 1980s (Veiga, 1997). Mil l ions of miners flocked to sites such as Serra Pelada, where 80,000 miners produced 90 tonnes of gold from a single pit. Veiga (1997) estimated that around 5,000 tonnes of mercury was released into forest and urban areas in Latin America over this time. In addition to Latin America, small-scale mining spread through areas of Africa and the As ia Pacific. Mercury, due to its simple effectiveness and practicality, is virtually ubiquitous with small-scale gold processing. These gold miners have been both a major source of mercury releases and have been one of the groups, along with nearby communities, most seriously harmed by mercury pollution. Figure 1.1 Number of Artisanal & Small-scale Miners Around the Wor ld by Country 7 Using the ILO data from Jennings (1999), Mining , Minerals, and Sustainable Development ( M M S D ) (Hentschel et al, 2002), C A S M (Davidson, 2003) and the United Nations Industrial Development Organization's (UNIDO) Global Mercury Project ( G M P ) (Veiga, 2004c), Figure 1.1 was compiled. Countries where estimates have not been made, but A S M is known to occur, are listed as "no estimate available." National A S M numbers were calculated as the mean of the high and low estimate compiled from the reports. This figure does not include any countries with a GDP/capita over $US 15,000, with a population under 150,000, or with an area less than 4,000 square miles, in order to exclude tiny countries and recreational gold panners in affluent countries. 1.2.3 Mercury as a Local Health Concern for ASGM The mercury used in A S G M has wide effects on the miners, the local environment, and the people l iving downstream from A S G M operations. Exposure to mercury mainly occurs through inhalation of mercury vapour or through the consumption of methylmercury. Repeated exposure to mercury eventually leads to progressively more serious health problems. 1.2.3.1 Mercury Vapour Elemental mercury vapour is rapidly absorbed through the lungs (Keating et al, 1997). The people most affected by mercury vapour are the artisanal miners who practice amalgamation, and the gold shop workers, where the gold dore is bought and purified. Amalgam typically consists of about 60% gold and 40% mercury, with varying amounts of silver. Miners often purify their amalgam into dore by either directly evaporating mercury on a stovetop or by open flame. This method can lead to serious mercury vapour poisoning. A n alternative is to use a retort to distil the amalgam into dore while safely collecting and condensing the mercury vapour. Artisanal gold miners frequently purify their dore at gold shops, or at home. This process releases the 2 - 5 % mercury remaining in the dore into the gold shop and thus into the town environment. People living near the mining operations or the gold shops are also affected; the majority of mercury emitted by gold smelters is deposited near the emission source (i.e. within 1 km), contaminating the local areas (Veiga, 1997). 8 M a l m (1991) measured up to 60,000 u.g/m3 of mercury in the air when amalgam was burnt in pans; the recommended limit for public exposure is 1.0 ug/m 3 (Malm et al, 1990) and the recommended health-based exposure limit for metallic mercury is 25 ug/m 3 for long-term exposure and 500 ug/m for short-term exposure ( W H O , 1991). When retorts are used, mercury vapor concentration drops to as low as 10 ug/m 3 (Veiga and Baker, 2004), much lower than the 50 ug/m limit for industrial exposure outlined in the Health, Safety and Reclamation Code for Mines in British Columbia ( B C - M E M P R , 1992). Mercury vapour is oxidized in the lungs and forms mercury complexes that are soluble in many body fluids (Veiga and Baker, 2004). While around 90% of the inhaled mercury is excreted through urine and feces in a few days (Hacon, 1990; W H O , 1991), some of the mercury accumulates in the central nervous system (Mitra, 1986). Mercury can permanently damage the nervous system - the kidneys are most affected in the medium-term, while the brain is the dominant receptor in long-term exposure (Suzuki, 1979). Total mercury elimination through urine can take several years (Veiga and Baker, 2004). High, short-term exposure (1000 to 44,000 ug/m ) to mercury vapour is typically associated with symptoms including chest pains, dyspnoea, cough, haemoptysis, impairment of pulmonary function, and interstitial pneumonitis. Chronic exposure symptoms include a metallic taste in the mouth, gum diseases such as gingivitis, ulcers and the formation of a blue line at gum margins (Stopford, 1979). Long-term, low-level mercury vapour exposure has resulted in less pronounced symptoms of fatigue, irritability, loss of memory, v iv id dreams, and depression. Long-term exposure to high levels of mercury vapour can lead to delirium, hallucinations and suicidal tendency as well as erethism (abnormal irritability), excessive shyness, insomnia, and in some cases muscular tremors. In milder cases, erethism and tremors regress slowly over a period of years following removal from exposure pathways ( W H O , 1991). A person with a mi ld case of mercury poisoning can be unaware of the symptoms because they are mainly psycho-pathological. The symptoms can also be confused with fever, alcoholism, or malaria and other tropical diseases. 9 1.2.3.2 Methvlmercurv Methylmercury exposure is a primary source of mercury uptake for humans and wildlife; methylmercury is easily bioaccumulated and becomes concentrated in fish, particularly in carnivorous fish. Carnivorous species are usually the preferred species of consumption by most people and represent the main exposure pathway of methylmercury to humans (Veiga and Baker, 2004). Methylmercury (CH3Hg+), an organic compound, is the most dangerous form of mercury, as it is easily absorbed through the gastrointestinal tract (Keating et al, 1997). When mercury from artisanal mining (or any other source) is released into water systems, it may react with organic acids and thus be oxidized into methylmercury (Hinton, 2002). This methylmercury is bioavailable, which means it can quickly find its way into the food chain. It can also easily bioaccumulate, or build up over time in an organism's body, and thus methylmercury concentrates at the top of the food chain, in fish and then in humans (Horvat et al, 2000). Around 70 to 90% of the mercury in fish is already in the form of methylmercury, which rapidly reaches levels where excessive consumption by humans can be dangerous. Carnivorous fish are most susceptible to bioaccumulation as they eat a large amount of omnivorous or herbivorous fish, and retain most of the mercury that those fish have accumulated. In fish, methylmercury can impair reproduction, growth and development, lead to behavioural abnormalities, and lead to death. Mercury can also harm plants, birds, and mammals, with reproduction also being effected (Keating et al, 1997). Mercury-laden mine tailings have been linked with higher levels of methylmercury (Baker, 2002). While not of the magnitude of Minamata, fishing communities near A S G M operations have been found to exhibit symptoms of methylmercury poisoning, especially along the Tapajos River in Brazi l (Grandjean et al, 1999; Harada et al, 2001; Mergler, 2002). However, the actual mechanisms that convert metallic mercury to methylmercury are complex and not fully understood. Variables such as sediment E h (standard hydrogen electrode potential), p H , conductivity (uS/cm), total organic carbon (TOC), and sulphate, all contribute to this transformation (Veiga and Baker, 2004). The other variables 10 identified which contribute to methylation include the supply of Hg(II), composition of micro-organisms, temperature, biomass, and iron (Hinton, 2002). Variables able to reduce mercury availability include the presence of hydrous ferric oxides, the presence of clay sediments, and suspended solids (Veiga and Meech, 1995). Figure 1.2 shows an Eh-pH diagram for the main inorganic mercury species. Eh (volts) 2 1.5 -0.5 i 1 r 2 + water oxidized Figure 1.2 Eh (redox potential) versus pH for the Main Inorganic Mercury Species (Meech etal, 1998). It is important to note that there is a wide variety of possible sources of mercury other than A S G M , including geologic weathering and erosion, evaporation from waters and soils, run-off waters, ancient gold and silver mining, plant transpiration and decomposition, waste incineration, and forest fires (Veiga and Baker, 2004). Camourze et al (2001) highlighted the importance of erosion in carrying natural mercury bound to intensively weathered soils to Amazonian aquatic systems. The authors stressed that this is a much more important source of mercury for the entire Amazonian environment than any other source, although A S G M activities are more relevant locally. 11 People with severe methylmercury poisoning, or "Minamata Disease" exhibit all of the following five symptoms: visual constriction, numbness of the extremities, impairment of hearing, impairment of speech, and impairment of gait. Muscular atrophy and mental disturbance are prominent in acute intoxication (Veiga and Baker, 2004). Unborn children are especially susceptible to methylmercury poisoning. Methylmercury can penetrate into the placental barrier, transferring mercury to the foetus. When the intake of mercury is large enough, sterility can occur in both women and men. When the dosage is smaller in women, pregnancy can take place but the foetus can miscarry spontaneously. Even when the pregnancy goes to term, the baby can have severe neurological symptoms, including psychomotor retardation and other developmental delays (Harada, 1978). 1.3 Outline of Thesis Chapter one has provided a background review of artisanal and small-scale gold mining and mercury pollution, in order to underlie the following examination of China's A S G M sector. The chapter introduced mercury as a global health threat, summarized the toxic nature of mercury vapour poisoning and methylmercury accumulation, and defined and summarized A S G M globally and its link to mercury pollution. In addition, it here outlines the structure of this thesis and its overall objective. Chapter Two describes the objective, methodology and results of a general study of China's A S G M industry and communities. At the time this research was undertaken, little research had been done to understand the nature of China's A S G M and its relationship to mercury since the early 1990s, and none had been done since mercury was made illegal in 1996. The objective of this stage of the research was to examine several A S G M sites throughout China to evaluate whether A S G M still took place and under what conditions and if mercury use persisted and represented a potential health threat. A discussion explores the A S G M sector in China in detail, including estimates of the number of gold miners and production figures, basic laws and structures, and socio-economic situations. Chapter Three is devoted to a case study of "Gold Mountain", a small A S G M community in northern China. This study was undertaken to elaborate and verify the findings of the 12 general study described in Chapter Two. It describes the community, surveys its gold processing facilities, describes in detail its gold processing methods, estimates mercury losses, and describes a novel mercury retort used locally. Chapter Four describes the objective, methodology and results of test work undertaken on three "Go ld Mountain" samples. These samples include a feed ore, an intermediate tailing sample from amalgamation process, and a final tailings sample following amalgamation and cyanidation. The objectives of the test work are the following: to verify the mercury losses reported in Chapter Three; to evaluate the potential of centrifugal concentrators, in conjunction with cyanidation, to replace the current amalgamation/cyanidation techniques; and to explore the effect of cyanidation of mercury tailings on mercury solubility. Chapter Five brings together the results of the previous chapters to estimate the extent of mercury released by A S G M in China. In addition, potential technology alternatives to amalgamation that could reduce those releases are discussed and issues surrounding alternative technology and technology transfer are examined. Chapter Six discusses the limitations and bias of the thesis. Chapter Seven identifies areas that would benefit from additional study. Chapter Eight concludes the thesis by highlighting the key results and by discussing recommendations for how the impact of mercury releases from A S G M can be mitigated. The Appendix contains a report of the inaugural meeting of C A S M - C h i n a , an organization formed in part to address the issues raised in this thesis. 13 2 Study O n e : Ar t isanal & Small-scale G o l d M i n i n g in C h i n a 2.1 Introduction As stated in Chapter One, the objective of the research described in this chapter is to examine several A S G M sites throughout China to evaluate whether A S G M was still taking place and under what conditions, and if mercury use persisted and represented a potential health threat. The basis of this research was a field study undertaken during the summer of 2000. The research focused on understanding the use of mercury by A S G M miners and seeing if their methods of dealing with these environmental and health problems could be applicable to other parts of the world, or if techniques developed outside of the PRC could be used by the Chinese. The summer 2000 field study resulted in site visits to six regions, investigating seven mineral processing operations, three underground mines, one gold shop, one dredge operation, and one mine prospect. This survey resulted in a number of small case studies, future contacts, and a sense of how the A S G M industry operated in China, and provided the basis for further research. The trip demonstrated that A S G M and mercury use were common and widespread throughout China. Interviews were also conducted with a variety of experts involved with A S G M in China in different capacities. These interviews, combined with the literature review and site visits, form the basis of the discussion section of this Chapter, an exploration of the place of A S G M in China. 2.1.1 Literature Review The sum total of the literature available before the summer 2000 field study was one article: Lin, Guo, and Gan's Mercury Pollution From Small Gold Mines in China (1997). This article gave a brief overview of China's A S G M industry and a case study description of A S G M in Dexing County, Jiangxi Province. In 1993-4, Dexing had 200 14 "rol l mil ls" , probably muller mills, which each used around 8 kg of mercury per day, of which 10-20% was lost to the environment. Extensive use of mercury was recorded, with losses in Dexing reaching 160-320 kg of mercury per day. If it is assumed that the mines ran for 300 days per annum, 48 to 96 tonnes of mercury were released annually from Dexing County alone. O f four articles in the bibliography, two refer to American mercury guidelines and the other two refer to a difficult-to-find Chinese journal (Environment & Exploitation). The article also makes reference to a S C O P E Ch ina /CAST study in 1993-1994. This study is perhaps altogether lost, but is certainly unattainable. L i n et al conclude their article with the following statement: "Since September 1996, most of the small scale gold mining activities were strictly prohibited by China's national environmental legislation. Recent investigation in late 1996 indicated all those small mines [researched] were shut off (Lin et al, 1997). " Interestingly, the site visit in 2000 found a larger operation than described by L i n et al (1997) and amalgamation was still clearly integrated into their processes. The most valuable item from literature review, however, was L i n et al's (1997) simple sketch of China that indicated areas with small-scale gold mining. This map formed the basis for the field study. Some visiting Chinese professors to the U B C mining department declared there to be no artisanal mining in China and certainly no A S G M mercury use (Veiga, 2000). Thus, when the field study was undertaken, there was no indication that A S G M operations existed in China or continued to use mercury. Other works include Golas's excellent history of mining in China (1999), which provided a great background, but provided little help for a study on contemporary China. Lacerda (1997) includes some erroneous information on China's A S G M in an article on global mercury releases. For example, he reports, "after granting permission to form individual 1 The muller in muller mills is not capitalized - muller comes from an alteration of the Middle English term molour and means "a stone or piece of wood, metal, or glass used as a pestle for pounding or grinding (Merriam-Webster, 2004)." 15 enterprises in China, over 200 small mines were opened in Dixing Province since 1992, increasing gold production by 10% per year." Larcerda was presumably reporting on the 200 A S G M operations in Dexing County, Jiangxi Province, recorded in L i n et al, 1997 (China does not have a Dixing Province). Larcerda also reports, "Major [mercury] impacts are from the Brazilian Amazon, followed by China, where over 200 small gold mining operations were settled and 120 t/yr of H g are being released into the environment." Again, these were, the releases from one county among thousands in China, and are hardly a national figure. These errors demonstrate the importance of field research to aid in understanding a large and complex country. Since the first field study, a handful of other articles relating to A S G M in China have appeared. These include "Mercury consumption and contamination of environment in gold mining in China" by He, L i n , and L i (2000). He et al (2000) estimated China's mercury consumption to be only 80 tonnes per annum, but did so from estimates of 33 smaller state-owned mines, not by visiting A S G M operations. He et al estimated that 13% of gold production was from amalgamation, with 33 mines losing 14.6 g mercury/tonne, with 75% recovered and recycled, leaving 20 tonnes/a discharged. This can be taken as a reasonable estimate of the amount of mercury released by China's formal mining sector, but not its A S G M industry. Dai et al's (2003) article on mercury releases from small gold mines in Tongguan County, Shanxi Province notes how gold has been mined for over 900 years in Tongguan, and that most families in the County are involved in gold mining. A n unpublished 1997 report from L i n Yuhuan is cited, stating that around 120 tonnes of mercury are released in the county annually; 38% into the atmosphere, 62% into the tailings, and less than 1% into the local rivers. Dai et al (2003) report their findings of water sample mercury concentrations, with dissolved mercury ranging from 0.11 to 3.10jj,g/l (average of 0.74ug/l) and particulate mercury levels ranging from 0.10 to 258.62 ug/1. In filtered water, reactive mercury ranged from 7.76 to 62.28% of total mercury. Da ie r al (2003) do not report on the amalgamation processes used. 16 An unpublished paper given by Zhong (1999) at the Small-and Medium-Scale Mining Workshop of the 3rd Environmental Cooperation Workshop on Sustainable Development of Mining Activities in 1999 detailed the small-scale mining industry in China. Zhong provided an overview of ASM in China and explained the importance of ASM to local development and environmental and safety issues arising from ASM activities. While the paper mainly focused on coal, it did provide some excellent information on new policies affecting ASGM. Zhong made no reference to mercury releases. A 2001 report by Canada's Department for Foreign Affairs and International Trade (DFAIT), provides some useful statistics on ASM (although incorrectly sourced), but little else. An extensive review undertaken for the MMSD ASM country report for China (Gunson and Yue, 2001) found a broader review of artisanal and small-scale mining in China yielded applicable laws, journal articles, books, statistical yearbooks, and newspaper reports. There was one book and twelve articles on ASM coalmines, primarily on safety and closure. With respect to mercury, there were three articles on coal combustion and mercury, and three short general articles on mercury in China (primarily coal and industrial sources of emission). There were twelve news articles on small-scale coal disasters (these were a sample - there are many more such small news reports, generally reporting the location and the number of dead miners), nine on a Guangxi ASM tin mine disaster, and one article on a labour protest. Eight laws and regulation guidelines were found to be relevant to ASM, and three general reports on China's mining industry included small sections on ASM. For an industry that directly employs 3 to 15 million people in China and has a major economic and environmental impact on the world's most populous nation, this is a shockingly understudied area. Since 2000, a handful of additional articles on ASM coal policy have been written by Andrews-Speed in cooperation with Chinese academics (2002, 2003a, 2003b, 2003c). These articles, while specific to coal, have implications for the ASGM sector, discussed in later sections. 17 2 . 2 Methodology In order to achieve the f ie ld study objectives, several A S G M sites and the amalgamation practices u t i l i zed at the sites were examined throughout C h i n a . Three m a i n techniques were used: • Site vis i ts ; • Personal interviews w i t h people associated w i t h A S G M ; and • The H g E X Exper t system. 2.2.1 Site Visits Table 2.1 Site V i s i t s , 2000-2002 Area Year County Sites Visited Nor th -Cen t ra l C h i n a 2000 " G o l d M o u n t a i n " 2 One go ld mine , one mine prospect, and three processing plants G u a n g x i P rov ince , South C h i n a 2000 Longsheng Coun ty One G o l d Dredge, One G o l d Shop J i angx i P rov ince , South Cent ra l C h i n a 2000 L e p i n g Coun ty One G o l d Process ing Plant J i angx i P rov ince , South Cent ra l C h i n a 2000 D e x i n g Coun ty One G o l d M i n e & Process ing Plant A n h u i P rov ince , Centra l C h i n a 2000 T o n g l i n g Coun ty One Canad ian Junior Joint Venture copper-gold-s i lver processing plant. Shaanxi P rov ince , Wes t Centra l C h i n a 2000 Bayuan , Q i n g l i n g Moun ta in s T w o G o l d Process ing Plants. Nor th -Cen t ra l C h i n a 2001 " G o l d M o u n t a i n " One go ld mine , two go ld processing plants, one z i n c / A u smel t ing operation, two i ron processing plants, two f i sh ing reservoirs Inner M o n g o l i a 2001 B a o t o u M u n i c i p a l i t y M i d - s i z e d State O w n e d G o l d M i n e w i t h centrifuges. Nor th -Cen t ra l C h i n a 2002 " G o l d M o u n t a i n " Process ing Plant , P lacer G o l d Operat ion, Sand and G r a v e l A S M operation. Site vis i ts were undertaken to research institutes i n B e i j i n g and A S G M areas i n Hebe i , G u a n g x i , J i angx i , A n h u i , and Inner M o n g o l i a provinces , cover ing northern, southern, 2Gold Mountain is a pseudonym. While the miners have local government permits to operate, they are concerned about receiving publicity for fear that higher levels of government will shut them down. 18 south-central, and western China. Table 2.1 describes the site visits made, including visits to six counties, in these five provinces. Sites were chosen from the map included in L i n et al (1997), a small note found in Atiyah et al (1997) and discussions with a wide variety of people. Once in an. area known to have A S G M , locals were queried as to exact locations of mines and processing operations. Upon arriving at a mine, processing site, or gold shop, a manager or operator would be asked for an interview and a tour. 2.2.2 Interviews Over 25 interviews were conducted at A S G M mines, mills, and villages with labourers, operators, managers, financiers, retired miners, woman miners or wives of miners, fishermen, and local government officials. Interpreters were used when appropriate, or necessary, and were available at all times except for the first site visit to "Go ld Mountain" and the site visits to Guangxi and Inner Mongolia. Over the course of the site visits, and especially during interviews, miners were observed for any signs of mercury-related illnesses, such as muscular tremors or gingivitis. A t A S G M mine and process sites, people interviewed were generally asked the following questions: • H o w long has the mine been operating? • H o w did the operation get started? What permits are required to operate? • Is the operation legal? • H o w many employees are there? • What is the mine ownership structure? • H o w many other gold mines/processing plants are in the area? • What tonnage is usually processed per hour? • What are the average gold grades and your general gold recovery? • What is the mine or process operation? • Is mercury used in the operation? How? Is the mercury use legal? • Are they aware that mercury can be dangerous to their health and environment? • Where did they get their mining equipment and how much did it cost? • Where did they learn their techniques? 19 • What is their relationship with the government and how do they pay tax? • What is their relationship with local environmental agencies? In addition, over 15 people from research institutes, government offices and embassies, formal and international mining companies, and multilateral agencies were interviewed. These interviews were conducted in ASGM areas, Beijing and other major centres in China, and internationally. Questions were asked about: • The political, social and environmental situation of artisanal miners in China, including their legal status; • The land-use conflicts and coexistence between artisanal miners and the largely agriculture-based communities of rural China; • The traditional mining concept and the current public perception of mining in China; and • The relationship between artisanal miners and international mining companies. These last interviews were necessary to develop a broad understanding of ASGM in China. 2.2.3 HgEX Expert System During the initial field research, Veiga and Meech's "Expert System for Risk Assessment for Mercury Discharge from Gold Mining Operations", or HgEX (1996), was extensively used. Conventional approaches that attempt to correlate natural variables with mercury biota levels have relied on empirical regression models that often yield poor bioaccumulation predictions (Hakanson et al, 1988). Predictions are fraught with uncertainties and unknowns such as internal correlations between variables and site-specific aspects of biota contamination. HgEX uses a heuristic approach in which mining and amalgamation methods, along with natural variables, are dealt with using inference equations to give varying degrees of belief about the risk of bioaccumulation. The system can handle uncertain or vague data using fuzzy logic and neural network techniques. These procedures were intended to reduce the need for extensive monitoring programs and provide a preliminary diagnosis about bioaccumulation risk (Veiga and Meech, 1992). Despite sparse data and uncertainty, a diagnosis can still be made about the likelihood of a critical situation. HgEX was developed to make risk assessments of 20 mercury discharge from small-scale gold mining operations, primarily in the Amazon region by integrating information on biology, chemistry, geochemistry, and medical, social and political issues in order to evaluate mercury contamination for a single site or region (Veiga and Meech, 1996). low 0 -i 100 \ medium high 100 Technical Process is rejected • Process is accepted Figure 2.1 Alpha Factor for H g E X (Veiga and Meech, 1996) A s the behaviour o f workers depends on societal incentives and reactions, the definition of the level of acceptable mercury releases w i l l differ with the values o f the society. To map these differences, an alpha factor is calculated based on socio-political, technical and economic aspects of a society which relate to the acceptance or rejection o f mercury use in gold mining operations. A high alpha factor indicates acceptance of amalgamation practice and low control of mercury releases enforced by a society, which may be a country, a region or a city. For many regions in Africa, Latin America and Asia , a = 1; however, for Canada, where mercury is practically banned and well monitored by authorities, the alpha factor is much lower (0.1 or 0.01). For Canada 150 years ago, when the hazardous effects o f mercury were unknown and thousand of miners were colonizing the West, the alpha factor would have been much higher (about 10 or 100). Thus the 21 alpha factor can adapt H g E X ' s conclusions on high emission levels to different countries' or regions' conditions (See Figure 2.1). H g E X integrates common-sense knowledge with observations and field samples of variables that usually correlate positively with mercury pollution. B y accepting vague data and by estimating the importance of variables, H g E X enables the evaluation and recording of mercury use and its potential effects in situations where a lack of funds and equipment would otherwise prohibit study. 2.3 Results 2.3.1 Site Visits — Gold Processing Equipment and Methods in China This section describes the gold processing methods observed during the site visits. The primary methods observed include muller mills, ball mill/sluices, and dredges. Secondary processes observed include cyanidation, retorts, and dore purification. 2.3.1.1 Muller M i l l s Muller mills were found at sites in "Gold Mountain" and Shaanxi. Muller mills, also known as edge. mills or Chilean mills, simultaneously grind and amalgamate ore, kneading fine gold into the mercury. The mills usually consist of two or three large cast iron or stone wheels rotating on a central axis around a circular metal trough (see Figure 2.2). The wheels were approximately one metre in diameter. A l l muller mills observed were powered using electric motors, although it would be feasible to drive them with combustion engines or even animal power. A l l but the most remote areas of China are electrified, so electrical power is usually not a problem. Priester et al (1993) state that the mills require five to seven kilowatts of power to run, although a stronger motor is sometime required to start the wheels rolling. Ore is crushed to roughly minus one inch and then shovelled into the mi l l . A t least one kilogram of mercury is added per tonne of ore and ground together between the wheels and the trough. Lime is added to reduce mercury flouring and to raise the p H for cyanidation. Liberated gold contacts the mercury and forms amalgam, which can easily be collected by stopping the mi l l and cleaning the trough at least once a day. The 22 overflow fines are collected and leached with cyanide to further recover gold. The mills can process around one tonne of ore per hour, grinding the ore to approximately 80% mass passing 200 um. Figure 2.2 "Gold Mountain" Muller M i l l Locally manufactured muller mills are inexpensive, costing approximately US$1200. The muller wheels need to be replaced periodically and chips o f iron continuously fall off and need to be removed from the mercury, during cleaning, usually with magnets. A s the muller wheels each weigh at least a tonne, having local sources of production is practical to minimize transportation costs. In the "Go ld Mountain" region a muller mi l l manufacturer claimed to have sold over 500 mills locally. "Go ld Mountain" miners stated that they had taken the technology to areas in Inner Mongolia and to have exported the technology to areas as far as Myanmar, and that the technology was apparently introduced from Russia. In Shaanxi, a woman who used to mine explained that the muller mi l l technology was common in the area and had 23 been imported into the region by miners from Henan Province. Muller mills seem to be the technology of choice for A S G M in northern China. The ratio of mercury lost to gold recovered (Hgio s t:Aup roduced) is a common method for evaluating mercury losses from small-scale mining (Veiga, 1997; Hinton et al, 2003a). The ratio for muller mills varies greatly, as miners change the amount of mercury added based on their estimate of the gold grade and size distribution in their ore. Estimates of the Hgiost:Auproduced ratio run from about 10:1 to 50:1. This is explored in greater detail in Chapter 3. 2.3.1.2 Ball M i l l / S l u i c e Two operations in Jiangxi Province, two in "Gold Mountain", and one in Shaanxi Province were observed using a ball mill / sluice box method. The ore is crushed by a jaw crusher and then passes to a hopper feeding the ball mill. The electric-powered ball mills observed were around 1.5 to 2 metres in diameter and about three metres long. Oversized "pebbles" are collected from the ball mill discharge screen and returned to the feed. The ball mill discharges onto a sluice box with a mercury-coated plastic carpet or copper amalgamation plate before flowing to a spiral classifier that returns the coarse material to the ball mill (Figure 2.3). The fines from the classifier continue on to flotation and/or cyanidation processes. The carpet or plate is cleaned at least daily to collect the amalgam. These plants process in the range of 30 to 100 tonnes of ore per day. As night shifts were uncommon, this likely meant 10-hour days processing 3 to 10 tonnes per hour. Ball mills typically use around 12 k W per tonne of ore, varying greatly with the ore's hardness (Priester etal, 1993). Miners in Jiangxi stated that they bought their mills from equipment manufacturers in Shandong Province and that other operations in the area used the same technology. Two larger mills in "Gold Mountain" were observed using similar unit operations, but did not use mercury at the time of the visit. A larger mill temporarily closed in Shaanxi Province, which appeared to use this technology and this type of operation, is identical to the 72 operations described in He et al (2000). Thus, this method seems common for slightly 24 larger A S G M operations. These types of mills are built as a package by equipment manufacturers and installed with some technical instructions. Figure 2.3 Jiangxi B a l l - M i l l Mercury Sluice Circuit Jiangxi miners claimed to process 30 tonnes of ore per day and to lose one kilogram of mercury per day. Getting accurate gold grades from miners was difficult, but assuming a generous amalgamation recovery of 10 grams/tonne, this would lead to a Hgi0st:Auproduced ratio of 3.3:1. 2.3.1.3 Gold Dredges and Placer Mining Over 10% of China's gold reserves are in placer deposits (Zhou and Goldfarb, 2002). One gold dredge visited in Guangxi Province shared its section of river with at least four other gold dredges. The dredge consisted of a line of heavy iron buckets attached to a digging arm inclined between two pontoons, with a small superstructure holding everything together (Figure 2.4). 25 Figure 2.4 Guangxi Gravel/Gold Dredge Operation Each bucket could hold about two cubic feet of material, and the dredge probably moved less than 30 cubic yards per hour. The buckets discharged the river gravel onto a screen, separating the coarse stones from fine gravel. The fines passed over a simple sluice box with a plastic carpet to catch the gold. Both the fine and coarse material was deposited into small barges and then transported to shore. The sluice box was cleaned every day or two, and they usually recovered around one or two grams of gold with each cleaning, although they could recover up to six or seven grams in some richer areas of the river. However, these dredges were primarily recovering gravel for China's huge aggregates industry. The Guangxi dredge operation amalgamated and panned their sluice concentrates on the side of the river. From international experience, amalgamation of concentrates without retorts have Hgiost:Auproduced ratios of around 1:1 (Veiga, 1997). In 2002 a return trip to "Go ld Mountain" found a new gravel operation, using a front-end loader, screens, and a carpet sluice (Figure 2.5). The sluice was cleaned at least once a day and all o f the heavies retained were passed over a shaking table. They claimed not to 26 use mercury to recover gold from the concentrates; however, they did use mercury at their muller mill operation. Over the course of this research, dozens of similar dredges and gravel operations have been observed all over China. It would be logical to assume that many of these also attempt to recover gold. 2.3.1.4 Retorts Retorts were observed in Dexing and Leping in Jiangxi Province, and "Gold Mountain". The Dexing and Leping retorts consisted of two parts: a cast iron pot and a lid device. A metal tube, approximately three feet long, was attached to the top of the removable lid. Around a foot away from the lid, a water coolant tube encircled the lid tube. Amalgam is placed in the pot and the lid is sealed. Heat is applied to the pot and the mercury in the amalgam evaporates, and escapes the chamber through the tube. As the mercury gas passes through the water-cooled section of the tube, the vapour condenses and drips out the end of the tube, to be collected and recycled (Figure 2.6). In "Gold Mountain", miners used an innovative retorting method, heating amalgam in a bowl covered by a bucket sealed with water. This retort will be discussed in detail in Chapter 3. Figure 2.5 Gold Mountain" Gravel/Gold Operation 27 Figure 2.6 Jiangxi Province Retort Cooling Tube 2.3.1.5 Cyanidation & Zinc Strip Smelting A S G M leaching operations were observed in four processing centres in "Gold Mountain", in Inner Mongolia, and in Leping and Dexing in Jiangxi Province. In Bayuan in Shaanxi, miners sell their tailings to larger mills for cyanidation. Gold ore, often after amalgamation and/or flotation, is placed in vats in the range of 200 cubic feet. The tank is saturated with a leach solution, and the pregnant solution is drained out the bottom of the tank. The leaching time depends primarily on the type of gold ore; refractory gold ores could be leached for as long as a week. Cyanide concentrations vary widely with the type of ore. The unfiltered pregnant solution drains out the bottom of the tank and flows through locked metal boxes filled with fine zinc strips (Figure 2.7). Gold replaces the zinc through the Merrill-Crowe Process; periodically, the boxes are cleaned out, the zinc strips collected, and smelted to recover the gold. The solution is not aerated before cyanidation or de-aerated afterwards, leading to lower gold dissolution and recoveries. 28 Figure 2.7 "Go ld Mountain" Zinc Strip Boxes Slightly larger state-owned gold mines use activated charcoal in leach methods with stirred tanks to recover gold. These operations frequently cannot compete with the A S G M private processing operations. 2.3.1.6 Gold Shops Gold shops were only visited in Guangxi and "Gold Mountain", but can be found all over China. Inquartation, or 'quartering,' is undertaken to purify the gold. Silver is first added to ensure a gold-to-silver ratio of roughly 1:4. Then the silver and any remaining mercury are dissolved with nitric acid over a hot plate at low temperatures, usually out-of-doors with no fume hood, close to the miners' home. Nitric acid w i l l not dissolve the silver i f the silver ratio is too low. Gold is left as a brownish powder. To further purify the gold a cupellation process is used. Miners melt lead into the gold and place the gold-lead solid inside a concrete ball, which is then baked on a coal barbeque. The lead and other impurities are absorbed into the concrete leaving behind a relatively pure gold bead. Usually cupels are made of magnesite or bone ash; the use of 29 concrete for cupellation is uncommon. The gold is sold to local consumers or the government, usually at world market prices. Figure 2.8 Burning Dore in Longsheng 2.3.2 Interviews The results of the site visit interviews are integrated with the results of the site visit observations, in Section 2.3.1. In general, people interviewed were friendly and informative. More than a few of the site visits were followed up with a shared meal and a few drinks. These informal events often provided an excellent opportunity to gain the trust of the miners and to learn from them in a less threatening environment. During the initial site visits and interviews, no overt signs o f mercury related illnesses were observed, such as gingivitis and muscular tremors, often used as potential indicators of advanced mercury poisoning. Less pronounced symptoms such as fatigue, irritability, loss of memory, and depression would have been impossible to detect by an observer 30 with no medical training during short site visits and interviews. However, in the subsequent case study in "Go ld Mountain", one miner with severe mercury-related illness was interviewed, as described in Section 3.3.3.1. The results of the interviews with people from research institutes, government offices and embassies, formal and international mining companies, and multilateral agencies are integrated into Sections 2.3.3 and 2.4. Where possible, the interview results are corroborated with references. 2.3.3 HgEX Expert System H g E X was an invaluable tool during the initial research, as it provided detailed explanations of the primary factors of mercury releases from A S G M and of mercury bioaccumulation. It also provided an excellent set of questions and areas to investigate, which made the preliminary research far more effective than it would have been otherwise. In this section a determination of China's national alpha factor is presented, along with an evaluation of mercury contamination for the site of Bayuan, Shaanxi Province. 2.3.3.1 Alpha Factor Determination for China The national or regional alpha factor is one of the primary determinants of H g E X . The result of an alpha factor determination for China is shown below. H g E X questions are in italics and the certainty factor (CF) is noted in brackets at the end of the responses. "Yes" equals 100% C F and "no" equals 0%CF. Is it easy to evade laws that control Hg usage? Mercury laws were in place in China (Lin et al, 1997) and miners indicated they were aware of them in interviews. In Jiangxi Province and "Go ld Mountain" it seemed relatively easy to get a locally issued permit to use mercury, although higher levels of government did not recognize these permits. Mercury regulations were sometimes enforced. In Bayuan a month before the site visit in July 2000, the Shaanxi Provincial Environmental Agency fined and shut down all operations using mercury. Everywhere visited was cautious about provincial environmental agencies. (50% CF) 31 Is there any incentive provided by the government for informal mining operations? Interviews indicated there were not currently any incentives. In the 1980s to early 1990s the government did encourage A S G M (Andrews-Speed et al, 2003c), but since 1995-97 the government has cracked down heavily. However, local governments do encourage/work-with/own-a-stake-in some of the mines, as A S G M is frequently vital to providing employment and taxes to local economies. (10% CF) Frequently Hg pollution affects other social groups, such as fishermen, natives, urban people, etc. not directly involved with mining activities. Do the groups, which may be affected by Hg pollution, have political power? Downstream urban communities wield significant power in China, and are generally the driving force behind environmental concern. N o areas visited had significant wi ld fish stocks, but aquaculture was common and fishermen would have some power. "Natives," or minority groups in China, usually had some "autonomous" power over the counties they resided in, which would give them a significant amount of power. However, this would also mean they were probably a major force in local mining, although this does not seem to be the case in Tibet (World Tibet Network News, 1995) However, in poor regions local government would give precedence to A S G M . (90% CF) Many societies hold a mystique that gold mines are an easy way to get rich. Is this an important point in this society? Perhaps, however with so many sectors of the economy booming in China, this is only one of many options, and this is probably less of a factor than in other A S G M areas of the world. Miners interviewed considered mining more profitable than farming, but starting a business or moving to an urban area would be considered more profitable than mining. (50% CF) Is mining a traditional occupation in this society? Mining probably has a longer continuous history in China that anywhere else in the world, and is certainly is a traditional occupation. Historically mining, as with other 32 commercial activities, was looked down upon by Chinese society (Golas, 1999). Regardless, mining is new to some communities, and was new to the Bayuan site. (100% CF) Is the level of education high in this society? In the cities, the education level is high. In the areas where the mines are, the mine operators usually have about the equivalent of high school education, but much of this might have been received during the Cultural Revolution, when the education system was in turmoil. Virtually everybody would have been in school for between six and nine years. However, functional illiteracy is a big problem in rural China, mainly due to the difficulty of mastering Chinese writing. Officially the overall literacy rate in China is 86% (CIA, 2003); however, 90% of the illiterate 14% of the population is located in rural areas (People's Daily, 2002c). (80% CF) Is the interaction between miners and other educated people (university level) frequent? In the 1980s to early 1990s such interaction was more common (Zhang, 2002). In the early 2000s, much of the mining and process techniques came from equipment manufacturers, who presumably have access to university-educated people. The caution exhibited by A S G M operators toward environmental agencies indicates there was some interaction, and the environmental agencies presumably had some access to university-educated personnel. While some operations would interact, the vast majority would not interact frequently. (30% CF) Is the dominant media reliable and well informed? While the Chinese media is censored politically, it increasingly comments on environmental problems (People's Daily, 2000, 2002a, 2002b). The press tries to report on A S M disasters, but local operators and officials are often uncooperative. (40% CF) Are ecological groups active? There are many international groups with local cooperation on issues such as soil erosion in northern China, and urban environmental health issues. There are not any domestic 33 environmental N G O s with significant power (United States Embassy in Beijing, 2001). (20% CF) Is amalgamation the first option considered by miners to extract gold? Yes, all of the A S G M mines visited used amalgamation. Cyanidation is another common option, in addition to gravity and flotation. (100% CF) Is mercury readily available to miners? Yes, mercury was readily commercially available and inexpensive (US$ 6 - 8.50/kg). China has had a large mercury mining industry, which probably supplied most of the mercury. China also had A S M mercury mines, which may have been in the mercury supply chain. In addition, mercury is imported into China both directly and through Hong Kong (Maxson, 2004). (100% CF) Gold placers, colluvium, laterite deposits and abandoned tailings are usually neglected by companies due to low grade. Are miners currently working these types of deposits? Most A S G M operations visited were working small, high-grade quartz veins. N o miners were observed reprocessing tailings. Some gold dredges were observed, but only as part of aggregate operations. Reports from Canadian junior exploration companies (Southwestern Resources, 2004) indicate that A S G M miners often mine on potentially large deposits. (20% CF) Do miners make frequent improvements in the amalgamation process to increase gold recovery and reduce mercury emission? A l l operations would greatly vary the amount of mercury used depending on the quality of ore they were using. While miners visited tinkered significantly with their reagents, they tended to hold with the equipment available. (50% CF) Is it easy to introduce new gold extraction technologies? (Miners can assimilate easily?) The small-scale state owned mines did not seem very flexible, and in interviews managers stated they could not compete with the A S G M private operations, which were 34 generally run like small businesses. If a product could be shown to increase the revenue or reduce their costs, the private miners would be extremely interested. (50% CF) Are miners aware of the side effects of mercury? A l l o f the miners and people interviewed were aware mercury is dangerous. However, the specifics of mercury contamination were not necessarily well understood. (70% CF) Do the miners have access to any specialized technical support? The miners in Dexing were in contact with a company in Shandong Province that made mining equipment and provided some specialized technical training; in addition, they used government exploration teams to find their mines. Experienced miners from Henan Province trained the miners in Bayuan. However, these sorts of contacts were erratic and not available to all operations. (60% CF) Have the government or other research institutions come out with alternatives or solutions for mercury use? In the 1980s and 1990s, there was some assistance and research. The prevalence of retorts suggests some involvement. However, since the A S G M crackdown, it doesn't appear like there has been much interaction. (60% CF) Are miners currently working ores in which gold recovery from concentrate is low when amalgamation is used? A S M operators were extremely discreet with gold recovery figures. Most operations visited used multi-stage recovery, following amalgamation with flotation and/or leaching. Miners claimed to recover 30-70% of their gold with mercury methods. (20% CF) Is it easy to set up a monitoring programme to survey mercury pollution extent? To set up an effective, nationwide monitoring program would be prohibitively difficult and expensive due to the size of China. However, local or regional surveys are possible, in close cooperation with Chinese agencies or organizations (Lin et al, 1997; Da i et al, 2003). (30% CF) 35 Are there any examples of spectacular gold discoveries (or rushes)? Around 2001, gold was found in the Boka area in Yunnan Province. In the period of a couple of years over 50 tunnels were put into operation. Similar discoveries are probably made regularly, but are not significant in a country the size of China. In the early 1990s there was a gold rush in western China (Tse, 1996), and up to 10,000 miners rushed to the Nagqu area in Tibet in 1994 (World Tibet Network News, 1995). However, all the site visit mines were operated by local Chinese, and these gold rushes do not appear to have been on the scale of the gold rushes in Latin America (40% CF) . Is mining activity important for the region's economy? In most of the areas visited, mining was clearly the dominant industry. In China as a whole, it is quite important, but is superseded by agriculture, manufacturing and commerce (China State Statistical Bureau, 2000). (80% CF) Agriculture represents the main activity in some regions, but the quest for land leads individuals to see mining as an alternative. Is this the case for this society? A S G M is certainly seen as an alternative to the poor returns of agriculture in rugged, remote areas. However, better alternatives exist in China's booming coastal economy. Mine owners/managers are largely local business people, seeking a high return on their capital. Labourers often seek to supplement their agricultural income. (80% CF) Is the mercury price high compared with the international market (US$ 4/kg)? The price was as low as US$ 5.50/kg in 2001. (0% CF) Does mercury represent a high cost in the mining operation? Since the cost of mercury was so low and the ore was usually high-grade quartz veins, mercury was a minor cost (See section 3.3.1.6). (0% CF) Do the mining operations require expensive equipment? (High investment?) 36 Most of the mines had a significant investment in equipment (crushers, ball mills, muller mills, or dredges). However, these items are usually made locally for little cost, and some operations were low tech (Bayuan in particular). (30% CF) Is the region experiencing difficult economic times and is mining seen as a feasible economic alternative? China has boomed for the last 20 years; however, most of the development has occurred in the eastern coastal areas. A S G M areas are often far from the coast or large cities, and have had significantly less economic growth. Mining, along with migration to the cities, or small-scale rural industries is among the feasible alternatives. (50% C F ) Is the minimum wage low? In some of these areas, U S $ 300 or 400 would be an average annual wage. (100% CF) Is inflation low? In the early-to-mid 1990s, there was significant inflation, up to around 20%/year. However, since 1997, inflation has not been significant and some prices have even dropped. (100% CF) Low Acceptance 0.01 North America Today High Acceptance 100 China, Brazil Today North America in 1800s Figure 2.9 Alpha Factor Comparison These answers led to an Alpha Factor for China of 1.00, indicating that China has a moderate acceptance of amalgamation practices. A s shown in Figure 2.9, this factor is 37 similar to that of Brazi l in 1994 (Veiga and Meech, 1996). This Alpha Factor was used as input for the site evaluate for Bayuan described below. 2.3.3.2 Bayuan Site Contamination Bayuan, Shaanxi Province, is presented as a case study for a local diagnosis (Gunson et al, 2001). Southeast of X i ' a n , a major interior city and ancient capital of imperial China, lie the Qingling Mountains, home to hundreds of A S G M operations. A local Bayuan woman, who had once owned and managed an A S G M processing centre, provided most of the information about the area. One month before the Bayuan field study, the X i ' a n government environmental agency had shut down and fined a number of mines in the area due to non-compliance with environmental laws. It was assumed that the mines would soon reopen, as is frequently the case in China. This valley had four state owned mines and over ten privately owned mines. The private mines mainly used locally manufactured muller mills to amalgamate whole ore and used a sluice box to recover amalgam. A past A S G M operator stated that amalgam was frequently burned openly without retorts, and the mi l l tailings drained into pits. These tailings were sometimes sold to the state mines for reprocessing, but otherwise they were piled nearby, which allowed the runoff to drain into the local stream. The Qingling Mountains are full o f other small valleys similar to Bayuan. H g E X accumulated data input through its heuristic method, accepting uncertain or vague information. Using fuzzy logic and neural network techniques, H g E X output the following bioaccumulation assessment: This is a small site in which milling is the predominant extraction method. The amalgamation process is typically applied to whole ores, using muller pans. Amalgam is separated from heavy minerals in excavated pools or simply panned at creek margins. Amalgamation tailings are either discharged into the watercourse, left in pools or sold for reprocessing. Retorts are not used as the amalgam is burned directly in a pan. The sediment has a background level 38 inferred by HgEX at 0.042 ppm mercury.3 The water colour of the main drainage is clear and the biomass is low. Hot spots were visually identified, indicating a high mercury level. The contaminated sediment is rich in gravel and grey in colour. Near the mining activity, the watercourses are clear. Based on the above information, using a weighted inference equation, H g E X calculated a diagnosis by determining the following factors: • The local mercury releases were estimated to be very high, with a Degree of Belief -High Emission Factor (DoBHEF) of 90%. This factor is based on evidence about whether high releases of mercury are occurring locally, such as the use of muller mills; • The natural conditions were estimated to be slightly dangerous, with a Degree of Belief - Dangerous Environmental Factor (DoBDEF) of 39%. This factor was based on indicators such as water colour and the presence of local hotspots; • The adsorption process was slightly possible, with a Degree of Bel ief - Mercury Adsorption Factor ( D o B M A F ) of 20%, based on indicators such as suspended solids and sediment colour; and • The overall risk of bioaccumulation was determined to be high, with a Degree of Bel ief - Potential Bioaccumulation Risk of 100%, based on the equation D o B H E F + D o B D E F - D o B M A F . The summary generated by the input data was as follows: Mining activities are emitting very high mercury levels into the aquatic environment in which natural conditions are slightly dangerous for methylation and bioaccumulation. Mercury adsorption by sediments is slightly possible to control bioavailability. This situation leads us to believe that the risk of increasing mercury levels in the biota is high. 3 Please note that HgEX has not been validated for China. These results are estimates only. 39 Finally, the H g E X output stated that retorts should be introduced at local mine sites. 2.4 Discussion - ASGM in China The site visits and interviews, in conjunction with the H g E X expert program, conclusively determined that A S G M still took place in China in the early 2000s, and that mercury use was widespread and represented a significant health threat. The following discussion integrates the results of the interviews about the general state of A S G M in China with the information gathered from site visits and interviews, H g E X , and the findings of the literature review. This broad overview highlights important differences between A S G M in China and other nations and is necessary in order to understand the present state of A S G M in China and how the sector can become less damaging and more sustainable in the future. 2.4.1 Significance of ASM in China Unlike most developing countries, gold extraction is not the main A S M activity in China. China has the largest mining industry in the world, by number of mines, production, or employees. China's A S M miners make up the vast majorities of China's mines, approximately half of the employees, two fifths of the production, and around a quarter of the value of production (DFAIT, 2001). Min ing is a significant economic activity in all of China's provinces, municipalities, and territories, with the exception of Shanghai. It has been an important driver of China's booming economy and is one of the largest sources of pollution in China. China's A S M industry employs at least six mil l ion artisanal miners (Gunson and Veiga, 2004), composing about half of the artisanal miners in the world (Jennings, 1999) and constituting one of the largest sectors of the mining industry worldwide. They produce at least 11 per cent of the world's coal output, easily out-producing the entire coal industry of major producers like Australia or India. A s shown in Table 2.2, China is one of the largest producers of metals and minerals in the world, topping the world in antimony, coal, iron, lead, manganese, tin, tungsten and zinc (Billi ton, 1999); artisanal miners produce at least 30 per cent of each of these minerals (DFAIT, 2001). Table 2.2 does not 40 provide a complete accounting Chinese A S M , but highlights selected minerals. The chief economist of Rio Tinto, one of the largest mining companies in the world, states it "is difficult to overstate the importance of China to the future of the world mining and metals industries" (Humphreys, 2002). China is unique in the scale and diversity of its artisanal mining industry. Table 2.2 Chinese Productions of Several Major Minerals (after Bi l l i ton, 1999 and DFA] IT, 2001) Mineral China's China's % Produced Number of Product Production World by Chinese A S M in 1997 ('000 Ranking by A S M workers tonnes) Production Aluminium 8,000 10 75.63 23,625 Antimony 101 1 46.44 54,599 Coal 1,360,000 1 42.58 2,696,056 Copper, mined 440 8 9.08 54,865 G o l d 4 0.185 5 33.3 463,000 Iron Ore 249,000 1 32.13 259,704 Lead, mined 650 1 31.13 83,827 Manganese 2,300 1 65.77 74,170 Mercury 0.614 3 8.52 738 Molybdenum 30 2 15.46 11,364 Phosphate Rock 30,400 2 51.03 45,145 Salt 29,300 2 10.36 3,188 Tin , mined 56 1 44.48 31,906 Tungsten, mined 24 1 35.31 25,599 Zinc, mine 1,210 1 31.13 83,827 2.4.2 Scale of ASGM in China 2.4.2.1 Definition China has few completely illegal or unregistered miners, primarily because it is relatively easy to get some sort of permit from the local government. Zhong (1999) estimates that only 10% of A S M had no legal status in 1999. However, reliable statistics estimating the number of illegal miners are simply not available, given the nature of illegal mines and 4 The gold figures are based on the discussion in section 2.4.2.2. 41 the scale of A S G M over a vast area. Undoubtedly the number of illegal mines varies widely from region to region. Most artisanal and small-scale mining in China today can be categorized as Township and Village Enterprises (TVE) . The concept of the T V E is a flexible organizational model that was developed to handle the dismantling of the commune system as China began to implement economic reforms in the late 1970s. T V E s were intended to "promote economic growth and to absorb the surplus rural labour force and discourage excessive urban migration" (ILO, 1998). T V E s have since come to dominate China's rural economy, employing up to 120 mil l ion workers, and producing billions of dollars of goods including everything from toys to coal. They are also covered extensively in China's statistics, and thus present an easy target to use to define small mining. T V E s are hybrid institutions, commonly an ambiguous alliance between private entrepreneurs and local government officials acting as "owners". The entrepreneurs, or managers, negotiate contracts containing both implicit and explicit conditions for them to conduct their business with the relevant public officials. Managers can gain a great deal of flexibility in such contracts and the local officials secure revenue for their local government (or for themselves). Officials often also provide special access to inputs, public resources, tax breaks, or any number of other incentives to help their local T V E s . The local Peoples' Congress elects these officials, so many T V E s are inherently political institutions. The term T V E officially means an organization owned or financed by a township or village government. For the purpose of collecting statistics, the term has been expanded to include collectively and individually owned enterprises, including self-employed people, i f they are officially registered and are run by people from rural areas, defined as farmers, or peasants. A s T V E s are by definition in rural areas, T V E s once fell under the responsibility of the Ministry of Agriculture, but are now under the Ministry of Land and Resources. The distribution of T V E ownership varies widely between regions. In 42 practice, local governments interpret what constitutes a T V E based on local conditions (ILO, 1998). While the T V E category covers most A S M mines in China, there are exceptions. Some T V E mines are large enough, and perhaps sophisticated enough, to raise the question of whether they are actually artisanal or small-scale mines. There are privately owned small mines that may not be registered as T V E s , as well as an indeterminable number of outright illegal mines. Furthermore, a few of the state-owned mines, which are considered large-scale, are actually quite small and/or primitive. Mines owned by the federal, provincial, or prefecture (municipal) governments are often considered to be state-owned, or formal mines, whereas mines at a county, township or village level are generally T V E s . In addition, a range of state companies and agencies own A S M operations, including the army, the prison service and large-scale mining companies (Andrews-Speed et al, 2003b). Private mines, while financed by private individuals, often have at least some connections to local government. These mines are usually licensed, but only with locally issued licenses that are often just a formality. Anywhere from one to several businessmen w i l l often create a small formal company with ownership divided into shares based on the initial contributions. Profits are then split accordingly. These partnerships are sometimes referred to as collectives, but the labourers usually do not have a share in the partnership. The company w i l l use local prospectors or hire a professional geological team to find a deposit, or w i l l already have a deposit in mind. One or two of the businessmen w i l l manage the mine. In practice, these managers have considerable control of the mine finances and may often be seen as not splitting profits fairly. In the late 1990s, the central government turned over control of most of its artisanal mines to provincial and prefecture governments, in an effort to concentrate on larger mines. These state-owned small mines tend to follow the law much more closely than the other small mines, and often have a better safety and environmental record than others. 43 Worker cooperatives do not seem common. Equally absent are producer associations; individual small mines rarely formally cooperate with each other in associations, although the owners are often on good terms with the other mines in the area and occasionally discuss technical issues. 2.4.2.2 Scale of A S G M in China Figure 2.10 A S G M Areas Across China (after L i n et al, 1997) There are no published national figures estimating the numbers of small-scale gold miners in China or the amount of gold they produce. Estimates can only be inferred from other data. According to a D F A I T report (2001), in 1997, an estimated 3,000 A S M gold mines employed well over 150,000. In comparison, in 1994 the Chinese government estimated that 300,000 illegal entities were involved in A S G M (Tse, 1995). In another almost certain underestimate, Tse (2001) reports that in 2000, "about 1,000 gold mines were in China, most of which were small and processed ores with obsolete technology 44 that damaged the environment and wasted resources." Zhou et al (2002) clarify this number, stating that in addition to the 1,000 small gold mines, there are "numerous very small local mines (treating <25 t rock per day) that are not accounted for in government statistics." For context, in the North China Craton, with approximately 20% of China's gold reserves, around 900 gold deposits and occurrences are known (Hart et al, 2002). Site visits in the area would indicate that the vast majority of these would have at least one, and probably several, A S G M operations in production. A s seen in Figure 2.10, A S G M activities occur throughout China. Clearly, it is difficult to get accurate statistics on the number of A S G M operations or on the number of people who work at them. The problem relates back to the difficulty of estimating China's gold production. Unt i l 2001, China required that all gold produced be sold to the People's Bank of China (PBC) , the central bank, at a fixed rate, usually around 10% below the international price. The price of gold was changed to roughly match international prices in 1999 (Tse, 2000), and a free gold market was established in 2001. Unt i l 2001, China's gold production was listed as the amount of gold purchased by the People's Bank of China (Tse, 1997). However, most of the gold mined by illegal miners is not sold to the People's Bank, so the total gold production was significantly underestimated. Gold was sold illegally for several reasons: • Higher prices could be received (if only by smuggling); • M u c h of the gold was consumed locally for jewellery; and, • Chinese individuals could not legally own gold as an investment until 2002. A s a hedge against China's currency, wealthy people would purchase gold. A n indication of the extent of this illegal trade was seen in 1997, when for a short time China's official gold price was held above international prices, due to the plummeting of the international gold price. A s much as 300 tonnes of gold was smuggled into China from Hong Kong alone during this period (Tse, 1998). From 1990 to 2001, China's gold production has increased from roughly 100 tonnes/year to 185 tonnes/year (Tse, 1996, 2002). There is a discrepancy between Chinese official 4 5 gold production numbers and outside experts' numbers, although it has been shrinking in the past five years. In 1994, Chinese official figures state that 90.2 tonnes of gold were mined, and that 82 tonnes were purchased by the P B C . Independent experts estimate that around 132 tonnes of gold were produced (Tse, 1995). B y the government's own figures, around 10% of China's gold production was being sold illegally. However, extrapolated from the experts' estimate, around 40% of gold was being sold illegally. From figures available for 1994 and 1996 this adds up to 50 tonnes and 65 tonnes sold illegally respectively. A S G M gold production estimates rest on two main factors; how much of the gold sold legally in China is from A S G M and how much additional illegal production occurs. L i n et al (1997) estimate A S G M produced one-third of China's gold production of 105 tonnes for 1995; however, Tse (2000) reports that the 1995 gold production was 140 tonnes. L i n et a/'s (1997) assumption is probably an underestimate. Zhou et al (2002) state that "local governments and rural Chinese companies now play a more important role and are responsible for 45 and 38% of China's annual gold production, respectively." Probably none of these operations, accounting for a total of 83%, could reasonably be considered formal large-scale mines as understood in North America. The largest gold mine in China produces about 3 tonnes of gold per annum, far less than major international gold mines. China's gold production for 2001 was approximately 185 tonnes (Tse, 2002). Using a conservative estimate of one-third of production, this would lead to over 61 tonnes of gold produced by A S G M . Veiga (1997) noted a correlation between gold production and the number of gold miners in South America. He found that the number of miners per kilogram of gold mined ranged from 2 to 8. The more gold a country produced, the higher the ratio of miners required, perhaps because a higher number of gold miners were lured by gold rushes. In rural China, the cost of labour is very low, with mine labourers in the wealthier coastal provinces being paid from US$ 70 to 100 per month, and in poorer regions far less. Thus, the ratio of miners per kilogram of gold mined is unlikely to be at the lower end. If Brazil 's ratio of 7.5 is used, the number of artisanal and small-scale gold miners in China 46 would be as high as 463,000. This is not unreasonable, given the extremely high numbers of rural unemployment and underemployment in China. Generating rural employment is arguably the highest priority of the Chinese government (The Economist, 2002). This estimate can be verified using a simple model extrapolated from "Gold Mountain's" experience. There, one muller mill/cyanidation operation processes around 10 tonnes of ore, recovering around 90% of the 20 g/t feed ore, and employs around 10 people for at least 150 days a year. About 10 small mines feed the operation, each employing around 20 people. The 210 employees at the process centre and the mines thus produce approximately 27 kg of gold per annum. In other words, 7.8 miners are required to produce one kg of gold. This ratio leads to a national total of 474,000 A S G M miners, quite close to the estimate indicated above. While this model is grossly simplified, similar numbers were observed in other areas during the site visits throughout China. 2.4.3 Economic Impact of ASGM 2.4.3.1 National Impact Chinese A S M as a whole had an immense short-term positive impact on China's economy. The annual production of China's entire A S M industry's value is in the range of U S $ 26 bil l ion to $180 bil l ion (Ministry of Agriculture, 2000), significantly more than Canada's entire mining and mineral processing industry's output of U S D $ 12.5 bil l ion in 2001 ( N R C A N , 2002). 5 It has employed millions of surplus rural labourers, people who may otherwise have added further pressure to the tens of millions of workers migrating from rural China to its booming coastal cities. A S G M operations often mine small deposits that would not be economical for the large-scale mining sector to develop. However, A S G M can waste deposits that could be viable for more efficient large-scale mining. In addition, small-scale miners can encroach on the formal large-scale mines, and can cause great harm to the workings, by inadvertently flooding them. A S G M often high-grades a deposit, only taking the easy-to-reach, most 5 Canada's output excludes oil and natural gas, but includes coal, and is calculated at an exchange rate of 1.55 Canadian dollars to one US dollar, the Bank of Canada's exchange rate average for 2001. 47 valuable parts of the ore body. For example, in 1995 an estimated 30,000 "peasant prospectors" in the H o i X i i area of Qinghai Autonomous Region caused the government to close all gold mines in the area in an attempt to protect gold reserves (Reuters, 1995). •2.4.3.2 Local Impact Locally, A S G M mines often supply the bulk of tax revenue and employment for poor and remote counties. This revenue is largely used to pay for infrastructure, health and education, and reduces or eliminates the need for higher levels of government to provide support. The economic impact of A S G M is most felt in poverty-stricken, remote, agriculturally poor provinces like Guizhou. In contrast, in a booming coastal province such as Shandong, the importance of A S G M is more localized and relatively minor. 2.4.3.3 Impact on Formal Min ing Companies Most foreign and state-owned mining company officials met in the process of this research could give examples of problems with artisanal miners. It is a sensitive issue; few people want to go on record on the topic and there is very little documentation. A n indication of the animosity that exists between A S G M and larger mines can be seen from the complaints that arise. Chinese officials and state-owned companies, and Canadian junior exploration companies, frequently complain that A S G M destroys otherwise valuable deposits. A S G M often takes only 20-30% of the deposit, leaving the remaining ore less attractive to formal mining operations (United States Embassy in Beijing, 1996). In general, artisanal miners, despite warnings and evictions, often repeatedly illegally mine on property licensed to larger companies, usually high-grading deposits. This can make conventional mining more dangerous, less profitable and, in extreme cases, not viable. On the other hand, larger companies often become interested in a site because of the presence of the artisanal miners. Larger companies bring in police to force the small-scale miners off the property. Occasionally these removal operations can be quite volatile. Often the artisanal miners are migrant workers, and are forced to return to their home areas. Local miners might be given compensation i f a larger company decides to develop their property, and sometimes the larger company finds alternative areas for these miners 48 to work. These other areas may or may not be economically viable for the artisanal miners (Gunson and Yue, 2001). Southwestern Resources (SWG), a Canadian junior, has an especially interesting relationship with small-scale miners. In 2002 S W G formed a joint venture company with China Yunnan Province Nuclear Industry Team 209 (YPNI) to develop the Boka property (Southwestern Resources, 2004). Y P N I ' s name is a holdover from the planned economy days when exploration companies had mandates to explore for specific minerals to the exclusion of all others. Figure 2.11 A S G M Tunnels, Boka, China (Southwestern Resources, 2004) Y P N I had acquired a property called Boka, on the banks of the upper reaches of the Yangtze River, that the provincial exploration teams had indicated had gold, and started small-scale mining in early 2001. In 1999 or 1998, "i l legal" small-scale miners had already heard of this assessment and started to mine the area heavily, building more than 200 tunnels into the mineralized zone, as shown in Figure 2.11 (Southwestern Resources, 2004). With the signing of the joint venture, Y P N I had the "i l legal" miners evicted, 49 probably by local police forces. The "i l legal" miners had permits from local governments, but the local government did not have the authority to issue such permits (Paterson, 2003). Following the removal of the A S G M miners, the joint venture conducted extensive geological surveys of A S G M adits, or tunnels. Based on assays from samples collected from the tunnels, the joint venture announced the discovery of a massive gold deposit and S W G stock rose almost 12 times above its value at the start of the year. The joint venture undertook a large geological drilling program to delineate the deposit in 2003 (Southwestern Resources, 2004). N o compensation was offered to the small-scale miners, although several were hired to be on the local exploration team (Paterson, 2003). Y P N I did not stop small-scale mining operations until January 2004 (Southwestern Resources, 2004). 2.4.4 Legal Framework 2.4.4.1 Government Framework China's hierarchical government structure consists of 5 to 6 basic levels: National, Provincial, Municipal or Prefecture, County, Township, and Villages. While this may seem excessive, it is worth recognizing that many of the provinces are of similar population and size to major European nations. China is not a federal system - lower levels are directly subordinate to higher levels. In practice, however, these different levels of government often have conflicting interests and goals. Over the past few years China has also undergone massive government institutional changes (Gunson and Yue, 2001). After being shuffled from the Ministry of Agriculture, A S G M now falls under the auspices of the Ministry of Land and Resources. 2.4.4.2 Government Policy From the late 1970s to the mid-1990s, essentially all levels of government supported the rapid growth of artisanal and small-scale mining (Andrews-Speed et al, 2003b). For example, A S M coal mines were given various subsidies and tax allowances. While there was consensus on the need to increase production, there was no such consensus on anything else, such as health, safety and the environment. Production grew at a prodigious rate but legal, technical, environmental, and safety requirements seemed to be 50 ignored in most mining areas. Short-lived and half-hearted rectification campaigns failed to make any significant impact on the behaviour of miners or local governments (Andrews-Speed et al, 2003a). Since the late 1990s there has been little in the way of support for artisanal miners in China. L i n et al (1997) reported a crackdown on A S G M in 1996. Zhong (1999) outlined a series of government orders to curb small-scale mining that included: • The No . 6 Order of the Minister of the State Commission of Economy and Trade on the Catalog of the Existing Lagged Production, Techniques and Products that must be Banned; and, • Catalog of Curbing Unnecessary Repetition of Capital Construction Investment in the Industrial and Commercial Circles. The "Catalog of Curbing Unnecessary Repetition of Capital Construction Investment in the Industrial and Commercial Circles" banned approval of or investment in cyanide operations treating less than 10 tonnes per day of gold concentrate, pyrometallurgial projects treating less than 50 tonnes of gold concentrates per day, heap leaching projects treating less than 5000 tonnes per day, hard rock gold mining projects treating less than 25 tonnes per day and placer operations treating less than 200,000 cubic metres per day. A s it is difficult for many A S G M operations to achieve these tonnages, many are forced to become illegal. 2.4.4.3 Laws and Enforcement A S M in China is subject to a large and continuously growing and changing body of legal documents drawn up by various government agencies at all levels of government. This regulatory framework is characterized by "duplication, inconsistency and omission" and A S M operations would be hard-pressed to comply with them all (Andrews-Speed, 2003b). Furthermore, much of the framework is unsuited to A S M . A few examples of laws and regulations that apply to Chinese A S M include: 51 • Regulations for Registering to Explore For Mineral Resources Using the Block System, 1998 • Regulations for Registering to Mine Mineral Resources, 1998 • Regulations for Transferring Exploration Rights and Min ing Rights, 1998 • Construction Law of the People's Republic of China, 1997 • Law of the People's Republic of China on Safety in Mines, 1992 • Regulations on Prohibition of Chi ld Labour, 1991 • Regulations on Reporting and Handling Fatal Accidents of Workers and Employees in Enterprises, 1991 • Mineral Resources Law of the People's Republic of China, 1986. Enforcement can be both erratic and harsh. Government agencies wield significant power over mines and other small industrial operations and are often regarded as corrupt and insensitive to local needs. In August 1994, Shandong Province, China's largest gold producing province, reportedly closed down 641 illegal gold mines and 50 retailing and processing shops (Tse, 1995). On the border of Henan and Shanxi Provinces, 430 illegal gold mines were closed, employing 13,000 A S G M miners, and 9,000 pieces of equipment were confiscated. In Tongguan County, Shaanxi, 87 ore processing mills were shut down (Tse, 1997). Judging from China's record with A S M coal shutdowns, many of these operations probably restarted. China's environmental control agencies can occasionally be effective. In Beijing municipality, miners apparently use only gravity concentration for gold production. Over the 1990s, the spread of knowledge about mercury pollution and retorts may be due to these institutions. In spite of the array of laws effecting A S G M , operators are often not particularly concerned about legal consequences; they can usually bribe their way out of trouble at the local level. Managers interviewed in Jiangxi Province stated that it is easy to get all of the necessary local permits by hosting banquets for local government officials. In contrast, going through official routes is time consuming and costly. Enforcement of safety 52 regulations is lax, and in the event of casualties, management is only expected to pay the family around US$2,415 (Fong, 2001). In any case, the concept of a legal society and regulatory framework as understood in North America is rather new to the People's Republic of China; most of the laws are only a decade or so old and they often mean less and less the further one gets from Beijing. 2.4.4.4 Tax A s A S G M mines are essentially regulated by hundreds of small, local governments, which often own mines themselves, tax collection is anything but uniform. Rural tax collection is a contentious issue in China, characterized by violence and tax revolts, and unofficial taxes imposed by local governments (Pomfret, 2001). In theory, all mines are subjected to taxes, royalties, and other, often arbitrary, fees. Taxes include a 1.18% Resource Compensation Tax, a Resource Tax, and a Normal Business (or income) Tax. Royalties are collected locally for the central government, and about half of this money is redistributed at the county level. The received royalties are used for promoting exploration and technical/safety support to the mines as well as for local government administration. Counties may not receive disbursements in proportion to the royalties they collect; thus, there is not much incentive to collect these royalties, especially as the local governments often own the mines. Overall, the tax burden in rural China is relatively low, but ad-hoc fees imposed by local governments can be high, often reaching 20% of post-tax profits (The Economist, 2002). However, the overall tax burden of A S G M , including ad-hoc fees, is unknown and almost certainly varies significantly with location. 2.4.4.5 Illegal A S G M According to Tse (1996), most of the illegal A S G M operations are found in western China, in Gansu, Guizhou, Ningxia, Shaanxi, Sichuan, Qinghai, Xinjiang, Xizang, Yunnan. L i n et al (1997) reported that most of the alluvial gold deposits are mainly exploited by individuals and that around half of the hard rock gold operations are mined by individuals or collective units. Tse (1996) describes the illegal operations as such: 53 Scrambling to get rich, hundreds or perhaps thousands of gold hunters rushed to the west. Widespread illegal exploration and unregulated gold mining seriously damaged the environment in the region. Government authorities were concerned that the country's already degraded base of natural resources was also being depleted. Most of the mined gold was traded illegally. To keep things in perspective, most "i l legal" A S G M operations researched had local permits, paid local taxes, and boosted remote rural economies. They were known and, at some level, permitted to operate by at least county level government officials. The miners were primarily small entrepreneurs trying to escape the extreme poverty of farming a small plot of marginal land. They were illegal primarily because of the lack of effective laws and regulations suitable for A S G M . 2.4.5 Social Context: ASGM in the Community A S G M has a host of impacts on communities in China. Foremost is the employment and income A S G M brings to rural areas. The China State Statistical Bureau (2000) states that the average number of dependants per rural labourer is 1.54. Based on the estimate of 463,000 miners, this would lead to over 700,000 dependants. However, the ILO estimates that the number of people who depend on A S M , including a small multiplier effect, is approximately 7.5 dependants per miner (Jennings, 1999). Using this factor, the number of people dependant on A S G M in China would rise to 3,470,000 people. 2.4.5.1 A S G M Labourers Artisanal mine labourers in China are categorized as farmers (peasants), but usually spend most of their time and derive most of their income from mining. A l l rural workers, in theory, have some farmland available. The mines visited shut down for periods of approximately ten days around planting and harvest time. One problem that arises from these workers' status as farmers is that when the government or companies shut down artisanal mining operations, they do not concern themselves with the labourers, as they can "just return to their farms." However, many of these farms do not provide enough income to support a family, which is why many of the farmers begin mining in the first place. 54 Artisanal miners are occasionally migrants from poorer areas of China and bring all the problems usually associated with migrant workers, including prostitution and violence. This may be a particular problem with miners moving into ethnically distinct regions like Yunnan and Tibet (World Tibet Network News, 1995). 2.4.5.2 A S G M Income The typical salary for mine labourers in eastern China in 2001 was between six and eight hundred Renminbi ( R M B ) per month (1 U S $ = 8.28 R M B ) , or about 8400 R M B per annum (US$ 1014). Miners stated that this was roughly similar to other labour jobs, but that mines had paid much better even a few years previously. In comparison, the average Chinese mining engineer at a visited state-owned mine makes 27,000 R M B per annum (US$ 3260), and the average annual income per capita in rural China is 2210 R M B per annum (US$ 267), although this varies widely from region to region (China State Statistical Bureau, 2000). Mine owners are often quite wealthy, and are capable of investing hundreds of thousands of R M B in a new project. The owners are usually businessmen who may be involved in several types of industry. Their private businesses are often incredibly flexible, and demand quick payback of their investments. Arguably, the business skills learned by these A S G M mine managers and owners offer the best chance for sustainable development for these communities, as mines are closed down or run out of ore. 2.4.5.3 Women According to the "Regulations on Labour Protection for Female Workers and Employees" (State Council , 1988), "It is forbidden to assign female workers and employees to work in underground mines." Women are usually not directly involved with the ore extraction or mineral processing in Chinese A S G M , but they are often on mine sites assisting with cooking and other aspects of daily life. In "Go ld Mountain", some of the wives of A S G M owners were active in business affairs. The woman interviewed in Bayuan, Shaanxi, had been directly involved with processing gold ore, but she had insisted that her husband sell their processing centre and she bought a small store with the proceeds. She claimed she did not want to raise her children while operating a 55 mine. There may be hundreds of other such cases but women do not seem to be a significant source of labour. Dr. Peter Golas, from the University of Denver, spent considerable time traveling to various A S G M sites, while writing the "Science and Civilisation in China" history of mining. Golas (2003) concurred that women and children did not seem to be a large part of the labour force. In contrast, at a larger state-owned mine visited in Inner Mongolia about a third of the employees, including engineers, were women. 2.4.5.4 Children In all but the most impoverished and remote areas of China, children usually attend school. Legally, children are prohibited from working until age sixteen (State Council , 1991). Miners interviewed in "Gold Mountain" believed that some underage boys worked in coal mines in the next municipality. One report indicates that 10% of the artisanal gold miners in Nagqu, Tibet, were under the age of 14 (World Tibet Network News, 1995). However, in part because of the one-child policy, there are perhaps fewer children involved than in other countries. It is uncommon for rural families to have more than two or three children. Since such labour is illegal, there are no official statistics on its extent. Regardless, child labour does not seem to be prevalent in China's A S G M industry. 2.4.5.5 Labour Unions China has only one union; the state controlled and sanctioned A l l China Federation of Trade Unions and its branches. A n y unofficial unions, informal worker groups and attempts to form labour groups are illegal. The state union has little power to bring employers to account, and "given its status as a government sanctioned body in many cases, union leadership corresponds to factory management (China Labour Bulletin, 2003)." This was confirmed during site visits. Private A S G M operations visited did not have organized labour. 2.4.6 Health, Safety, and Environmental Impact The negative impacts of artisanal mining in China on health, safety, and the environment are huge and difficult to quantify. Many villages have found themselves enriched at the expense of their health and environment. A S G M is far less deadly than China's A S M coal industry, which kil ls 4,000-6,000 miners every year (State Bureau of Coal Industry, 56 2000). Regardless, A S G M certainly has accidents. In June 2002, an electrical fire ignited explosives in a Shanxi Province gold mine, ki l l ing 37 miners (People's Daily, 2002b). Miners had been ordered by management to continue working after a fire had broken out underground. The owners then tried to cover up the blast by secretly removing the bodies at night (People's Daily, 2002a). Table 2.3 lists a variety of health, safety, and environmental impacts arising from A S G M in China, in addition to mercury use. Table 2.3 Major health, safety, and environmental impacts of A S M in China Item Impacts Death and accident toll While underreported, fatalities and serious injuries occur regularly. Solid Waste A S M tailings and waste rock are arguably the largest source of industrial waste in rural China (Andrew-Speed et al, 2003b) Water While most gold dredges and placer operations are prohibited, many still existed as of 2002. Dredging for aggregates is widespread, and causes significant sedimentation problems. Cyanide Gold producers' cyanidation practices are often questionable and visibly cause significant problems downstream, where fish populations have sometimes disappeared. Cyanide spills have occurred. Hazardous Chemicals A wide variety of chemicals other than cyanide and mercury are often used by A S G M , including acids, flotation reagents, and p H modifiers. These chemicals are rarely closely accounted for or handled safely in a systematic method. A i r pollution Blasts, heavy equipment, and comminution account for significant amounts of dust and gas emissions. Silicosis and other lung infections appear to be widespread, exacerbated by heavy smoking. In a Jiangxi Province mine, silicosis led to the death of 100 miners, a quarter of the mine's workforce (Becker, 2004). Loss of vegetation Frequently nearby trees are cut for underground support (timbering), and vegetation is often cleared from around mine sites. This loss of vegetation can lead to erosion and loss of animal habitat. Water table Dewatering underground mines can lead to a drop in local water table, affecting agriculture. Subsidence Shallow, poorly or un-supported underground mining openings can subside, causing damage to surficial buildings or creating dangerous 'glory holes'. China has a special problem with the formation of sinkholes in karst areas due to mining (Zhou, 1997). Sound Like all mining, A S G M operations can be extremely noisy, due to blasts, heavy equipment, and crushing and grinding. Not only do these cause permanent hearing damage to employees, who rarely wear aural protection, but can affect nearby communities and homes. 57 The harsh conditions of A S G M can also lead to illness. The World Tibet Network News (1995) reported that artisanal gold miners in Tibet working 14 hour days and subsisting on steamed corn were falling i l l due to diseases such as emphysema. There have been at least three major cyanide spills in the past 10 years in China. In late 2001, in Henan Province, 11 tonnes of sodium cyanide leaked into the Luohe River due to a road accident involving a truck transporting cyanide to a local gold mine. Thousands of police, soldiers, and civilians were mobilized to contain the spill ( B B C , 2001). In September 2000, a transportation truck released 5.2 tonnes of liquid sodium cyanide into the Wuguan River in Shaanxi Province (People's Daily, 2000). Both accidents resulted in fish deaths, but no human casualties. However, a number of cyanide deaths resulted in an accident south of Kuanchung, Hebei Province, in 1995 (Mortensen, 2003). 2.4.7 Technical Assistance Programs 2.4.7.1 Technical Assistance From the late 1970s to the mid 1990s, plenty of government assistance was provided to A S M . To use coal as an example, state-owned coal companies and coal bureaus provided substantial technical assistance to small-scale coal miners in the late 1970s and early 1980s (Andrews-Speed et al, 2003b), in an effort to increase national production. The state gold bureaus and companies may have been asked to give A S G M similar assistance at some point. In the 1980s and early 1990s some universities and research institutions provided technical assistance to A S G M (Zhang, 2003). However, since the mid-1990s, there has been little technical assistance available. Equipment manufacturers w i l l often help with equipment installation and provide some initial instructions, after which the miners are left to their own devices. Some governmental agencies, such as the China Mining Association, have small-scale mining departments, but these are too small to offer much direct assistance. 2.4.7.2 Research Cooperation There are a few cases of research programs or cooperation with A S M in China. In 1993, the Scientific Committee on Problems of the Environment (SCOPE) initiated a program that studied mercury and A S G M . The Slovenian Institute Jozef Stefan cooperates with 58 the Institute of Geochemistry, at the Chinese Academy of Sciences, in Guiyang, Guizhou Province on mercury-related issues, but A S M is only incidental to their research. Dr. Andrews-Speed, of the University of Dundee, Scotland, researches coal A S M and energy policy in China. Dr. Peter Golas, of the University of Denver, Colorado, undertook substantial A S M research in the 1990s while working on his history of mining in China. 2.4.7.3 Domestic Financial Institutions Bank financing is theoretically available to A S G M , but none of the miners interviewed thought it was a serious possibility, mainly because too much collateral is required. The state-owned mines and TVEs have some access to government finance and would have better access to banks. The private miners studied were entirely financed with private money. In general, China's banking system is in crisis and is not in a position to lend money to small private operations (The Economist, 2002). 2.4.7.4 Non-Governmental Organizations Non-Governmental Organizations are restricted in China, and the regulatory procedures necessary to officially register are often insurmountable. Many of China's NGOs are actually set up by the government with the intention of gaining international funding and have little or no institutional freedom (United States Embassy in Beijing, 2000). In fact, in order to avoid bureaucratic difficulties, a few genuine NGOs have actually registered as businesses (United States Embassy in Beijing, 2000). At its root the Chinese government is suspicious of any non-profit organization that does not owe allegiance to the Communist Party. Existing NGOs tend to focus on educational issues and do not interact with artisanal miners. No international NGOs are known to be involved with A S M issues in China. 2.4.7.5 Multilateral Organizations Virtually none of the major multilateral organizations like the United Nations, the World Bank, or government development agencies like the Canadian International Development Agency (CIDA) have programs to work with Chinese A S M . The only exception seems to be the ILO in cooperation with the Gesellschaft fur Technische Zusammenarbeit (German 59 Development Agency), which helped the Changsha Safety Training Centre for Small & Medium-sized Coal Mines in Hunan province. 2.5 Conclusion The interviews, field studies, and use of the H g E X expert system undertaken in this study clearly indicate that A S G M is widespread in China, that mercury use is common and represents a potentially serious health threat. The literature review indicated that this was a woefully understudied area, with few articles of any variety on the topic. Every site and region visited showed indications of mercury use from A S G M , be it from muller mills, mercury sluices, or the amalgamation of concentrates. Many operations followed amalgamation of whole ore with cyanidation. H g E X provided a valuable tool to assist with the evaluation of mercury contamination from A S G M in China. The national alpha factor for China was determined to be 1.0, roughly similar to Brazi l , indicating that China is moderately accepting of amalgamation practices. In the discussion, what constitutes A S G M in China is defined and it is determined that roughly 463,000 A S G M miners produced around 61 tonnes of gold in 2001, in regions all over China. The number of these miners' dependants could total almost 3.5 mil l ion people. The discussion outlines the complex relationship that A S G M has with the formal and international mining industry and highlights the lack of formal or informal support enjoyed by A S G M . A S G M is tightly bound with Chinese society through a wide range of ties and stakeholders, often with conflicting interests. 60 3 Study Two: Artisanal and Small-scale Gold Miners in "Gold Mountain" 3.1 Introduction The purpose of this chapter is to develop a case study of the community of "Gold Mountain", in order to elaborate upon and verify the findings of Chapter Two. "Gold Mountain" is a pseudonym for a northern Chinese gold mining village located in the rugged terrain between the Gobi desert and the Ye l low River Plains. Geologically, "Gold Mountain" is located on the northern margin of the North China (or Sino-Korean) Craton, which underlies much of north-eastern China and contains hundreds of gold deposits, most containing less than 1 tonne of gold (Hart et al, 2002). The gold in these deposits occurs primarily as veins associated with Mesozoic intrusions into mainly Precambrian host rocks (Poulsen and Mortensen, 1993). These veins chiefly contain pyrite, gold, electrum, chalcopyrite, galena, pyrrhotite, magnetite, molybdenite, and trace amounts of telluride minerals (Zhou et al., 2002). "Go ld Mountain" has hosted A S G M operations for over 70 years and is home to roughly 10,000 people. In addition to gold mining, iron mines and stone and gravel quarries operate in the valley. Agriculture consists primarily of corn and chestnuts in addition to some tree farms. In the nearby reservoirs, fish farms raise carp. 3.2 "Gold Mountain" - Case Study Methodology This study describes the community of "Go ld Mountain", details its gold mining processes, describes a novel mercury retort used locally, and estimates mercury losses and the health and environmental impact due to A S G M . In order to build this case study, several site visits were made between 2000 and 2004. The following steps were taken: • Site visits were undertaken in order to understand in detail every step of their process and to determine the extent of A S G M in the community. Sites visited are noted in Table 2.1, Chapter Two; 61 • Interviews were conducted in order to gain a more in-depth understanding of community, the structure of A S G M in the village, and the health impact of amalgamation. People interviewed include mine and plant owners and managers, mine operators and labourers, retired miners, government officials, and local fishermen. During interviews, miners were observed for any signs of mercury-related illnesses, such as muscular tremors or gingivitis. • Representative samples o f approximately 200 grams where taken from intermediate and final tailings and assayed for gold and mercury back in Vancouver. Shovels full of tailings were taken from a number of locations in each tailings pile and then, using the cone and quarter method, the samples were reduced to approximately 200 grams. The samples were returned to Canada and sent for gold and mercury assay using I C P - M S ; and, • A n estimation of the total mercury releases from A S G M in the community was calculated on the basis of the interviews and samples collected. 3.3 "Gold Mountain" Case Study- Results 3.3.1 "Gold Mountain " Mining and Processing Practice In May 2001, there were six muller mills located in three different processing centres in and around "Go ld Mountain". These privately operated centres process ore from local gold mines at a set fee. There is a slightly larger state-owned gold processing operation in the valley, which only uses cyanidation. There are also several small, unlicensed and independent gold processing operations, which exclusively cyanide heap-leach low-grade ores. The ore is supplied to the centres by 40 to 50 small underground mines, mining gold-bearing veins in the area. Artisanal gold mining operations decreased significantly between 2000 and 2001. In 2000, there were 12 muller mills, another larger gold processing centre, which used leaching technology, and between 70 and 80 mines. L o w gold prices and the increasing difficulty of processing the ores are the main reasons for the shutdowns. The twisted iron remnants of discarded muller mills can be found on the outskirts of the village. However, 62 higher gold prices in 2002 have revitalized local interest in gold. This section details step-by-step the A S G M operations in "Go ld Mountain". 3.3.1.1 Min ing Practice Two gold mines were examined in "Go ld Mountain". The mines consisted Of tunnels about five to six feet tall. Stopes were drilled using pneumatic jacklegs and blasted with dynamite. Miners usually cut into the mountain on one level, then angle down steeply to a gold vein level. Once a vein is found, the miners follow it, blasting out larger openings i f the vein expands. The veins are typically one metre thick, slightly dipping, two or three metres wide, and extending several hundred metres. Gold grades usually range from 10-40 g/tonne. Ore is manually mucked, using handcarts, which are moved between levels by motor-driven winches. Waste rock is usually dumped a short distance from the mine's mouth. From the mine, the ore is taken several kilometres to the m i l l , usually by locally constructed two-cycle 10 tonne ore trucks. Miners often, but not always, wear safety helmets; accidents do occasionally occur. One mine visited employed about 20 people, and the other about 30. Labourers were paid U S $ 72 - 97 per month. Managers received a share of the profits. 3.3.1.2 Comminution and Amalgamation Two gold processing centres, operating a total of four muller mills, were studied. The first centre, operating one muller mi l l , employed about 10 people, including a cook, labourers, managers, and security guards. Labourers were paid around U S $ 72 per month. The second mi l l employed closer to 30 or 40 people. Miners would bring their ore to a mil l ing centre and pay a set fee per tonne of ore processed, in addition to paying for any reagents used. Miners would choose the quantity of reagents used in consultation with the millers. A s all o f the fees were fixed, the millers had little financial interest in the grade of the ore and were not responsible i f little gold was recovered. A representative of the mine would observe all o f the processing procedures. 63 A t the processing centre, ore would be stored, segregated according to ownership. In turn, each miner's ore is crushed using a jaw crusher. The ore is then shovel fed to a muller mi l l along with several kilograms of mercury. A steady stream of water flows into the mi l l , and lime is added to reduce mercury flouring. The ore is ground and put in contact with the mercury by three large iron wheels turning around a circular ironclad track. The fines overflow into a chute and are collected as the slurry passes into several settling tanks, before draining into a small tailings pond. The settled fines are later shovelled out for cyanidation. Random objects (like a hammer from a hammer mil l) are placed on the chute to be used as somewhat dubious obstacles to retain mercury. Most overflow material stopped by these objects tends to be washed into the settling tanks during the mi l l cleaning stage. The millers do not use riffles in the sluice box to catch fine gold or mercury, claiming they pose a security risk. Most of the particles seem to settle in the first few boxes. The final box, considered the mil l ' s property, takes days to f i l l . About 10 to 15 tonnes of ore can be processed per day, or around one tonne per hour. Figure 3.1 Cleaning the Concentrate From the Mul ler M i l l 64 A t least once per day, the muller mills are stopped and cleaned out. The ground ore and amalgam are shovelled into a hand basin, where the amalgam is liberated from coarse particles with a water hose (see Figure 3.1). Free iron from the mi l l is collected with a hand magnet. The excess ground material flows into the tanks with the fines. The amalgam is collected and placed in a thick cloth and manually squeezed, getting rid of the excess mercury. The amount of gold recovered by the mi l l varies with the ore, but miners claim it is frequently around 60% to 70%. The mills would operate around 10 to 12 hours per day. The muller mills are manufactured in a nearby town, and a visit to the manufacturer found a busy and relatively sophisticated operation. In addition to the muller mills, machine tools are used to manufacture jaw crushers and magnetic separators. The iron for the muller wheels is cast on site. The plant has been in operation for between 15 and 20 years and has sold about 500 gold muller mills to the surrounding region, selling mills for about 10,000 R M B or roughly US$1,200. There was also a smaller muller mi l l manufacturer in town. 3.3.1.3 Retorts "Go ld Mountain" processing centres use an uncommon and simple retort. The ball of amalgam from the muller mi l l cleaning is placed in a small pan on top of an electric hotplate, seated in a water-filled bowl, and then covered with a metal bucket (Figure 3.2; Figure 3.3). The bucket and the water form an air seal, creating a crude but effective retort. When heated, the mercury evaporates and the mercury vapours condense on the cover-bucket walls and drip into the water-filled bowl. After approximately an hour, the cover is removed and the owner of the dore takes it home for further purification. A s the temperature reached in the bucket is likely high, the mercury recovery is probably about 95% (Veiga, 2004b). The condensed mercury appears to coalesce (see Figure 3.4) and is periodically collected and recycled. This simple and cheap retort seems unique to "Gold Mountain". 65 bucket evaporated Hg condensed Hg metallic water electric element bucket Figure 3.2 "Gold Mountain" Retort Diagram Figure 3.3 "Go ld Mountain" Retort Photo Figure 3.4 Mercury Droplets from "Gold Mountain" Retort 3.3.1.4 Cyanidation and Zinc Smelting The fines from the settling tanks are shovelled out when the tanks are full and the sludge is placed in piles until the cyanidation tanks are available. Then the ore is shovelled into 10 tonne leach tanks, for vat leaching. About 5 kg of lime per tonne of ore are mixed with water or barren leach solution from the previous leach. Then, depending on whether the ore contains oxides or sulphides, 2.5 or 5 kg of sodium cyanide respectively are added to the solution. This solution percolates through the ore and then drains from the tank to a locked box full o f zinc strips, which collect the gold (Merrill-Crowe Process). N o de-aeration takes place, and cyanide is only added once at the start of the operation. The barren leachate is then re-circulated to the leach tank, with a retention time of usually seven days. If the ore grade is less than 3 or 4 g/t gold, determined visually from experience or simple panning, then the ore is ground using a hammer m i l l , then added directly to the leach tanks without amalgamation. The operators did not have any method of determining the cyanide or gold concentration in solution. 67 The final discharge is treated with powdered bleach (Ca(OCl)2) to remove the cyanide ions from solution. The primary reaction is: CN- + ocr = CNcr + cr It is not clear i f the amalgamation tailing discharge is treated. After the ore is leached, the zinc strips are scooped into a bag and the barren solution is drained off. The loaded strips are taken home, where a furnace w i l l already have been started. The furnace is basically a sturdy iron barrel filled first with wood and then topped off with 200 kg of coking coal (at $US60 per tonne). The wood is ignited first, and with the aid of a fan, the coke catches fire. Then the zinc strips are put in crucibles and placed in the furnace. After 30 to 60 minutes, several scoops of borax (at $US0.12 per kilogram) are added to the crucibles. After 15 minutes or so, the flux is poured into other crucibles and allowed to cool, with the precious metals settling to the bottom. After being quenched, the crucibles are emptied onto an iron plate and the gold and silver are knocked off the bottom of the slag. Overall, the miners claim a recovery rate of about 90%. N o effort is made to protect against fumes generated during the smelting of the zinc strips. While the leaching kinetics of mercury are slower than gold (Flynn, 1995), mercury would both move into the solution and plate on the zinc strips along with the gold. This mercury recovered on the strips would then be volatized in the smelting operation. 3.3.1.5 Gold Purification Miners typically need to purify their dore, as described in Section 2.3.1.6. 3.3.1.6 "Go ld Mountain" A S G M Cost Structure It is important to understand the economic reality of the miners in order to understand the dynamics of their mercury use and in order to offer realistic alternatives to amalgamation. The following is a breakdown of the costs and revenue of one typical miner. He claimed it cost him roughly US$12 to mine a tonne of ore and another US$5/t to transport the ore 68 to the processing centre. He paid the centre roughly US$1 7/t for processing and reagent fees, leaving him with a profit of US$42/t at a recovery rate of 90% with a grade of lOg/t. His cut-off grade is roughly 4 g/t, not including silver. These figures do not include taxes or local corruption ("rent seeking"). Since only six muller mills, each processing 10-15 tonnes of ore per day, serviced the 40 to 50 mines, the tonnages mined must be less than 3 tonnes of ore per day, excluding waste rock. A t a 90% recovery of lOg/t, this would mean a daily production of about 27g/day per mine, or 90g/day per mi l l . A s the grade can exceed 40 g/t, and the muller mills can process up to 15 tonnes per day, these estimates are conservative. Table 3.1 Typical Miner 's Revenue and Profits Per Tonne of Ore Ore Grade: 10 grams A u / tonne Local Gold Price, Summer 2001: $8.45/gram Item $US/tonne of ore Percentage of cost Total Revenue (90% recovery) ($US/tonne of ore) $76.09 Min ing Costs $12.08 35.7 Plant Fee $9.66 28.6 Mercury $0.96 2.8 Other Reagents $5.56 16.4 Cyanidation $0.72 2.1 Transportation $4.83 14.3 Total Costs ($US/ton of ore) $33.82 100.0 Profit (Total Revenue - Total Costs) ($US/tonne) $42.27 Mercury is included in the reagent fee, in addition to lime and any further reagents. Miners are charged for the amount of mercury not recovered after cleaning the muller mi l l . The mercury the miner is charged is for is thus the amount of mercury in the recovered amalgam (about 6 grams in this example) and the amount mercury reporting to the tailings (approximately 200 to 2500 ppm). Table 3.1 shows that mercury thus represents less than three percent of the processing costs. In Venezuela, where mercury costs close to five times as much (US$ 20-25), mercury represents around 4% of the processing cost, not including cyanidation (Veiga and Gunson, 2004). 69 3.3.2 "Gold Mountain " Mercury Releases and Sample Collection Through the site visits and interviews, several estimates of the amount of mercury lost by the muller m i l l amalgamation process were determined. These included verbal estimates, records from the processing plants, and assaying tailing samples. 3.3.2.1 M i l l Operator Estimates on Mercury Lost M i l l operators at one processing site in "Gold Mountain" estimated that they used 10 kg of mercury per tonne of ore and lost around 0.5 kg/t. A s the muller mills processed between 10 and 15 tonnes of ore per day, this would lead to mercury releases of 5 to 7.5 kg of mercury released daily for each of "Gold Mountain's" six mills. 3.3.2.2 M i l l Data on Mercury Lost The muller mi l l operators kept records for all o f the ore processed and the amount of reagents, including mercury, used at their plants; however, this data was a closely kept secret. Table 3.2 shows sample data on mercury losses, recorded and shared by one "Gold Mountain" processing centre. The "Mercury used in muller m i l l " column is the recorded mass of mercury added to the muller mi l l per tonne of ore. The "Mercury recovered from m i l l " column records the mass of the excess mercury recovered after squeezing the amalgam with cheesecloth. The "Mercury Lost" column is the difference between these two columns. A s can be seen, the mercury recovered by the retort is only a small fraction of the total mercury used. Table 3.2 "Gold Mountain" Processing Centre Data Ore Sample Mercury used in muller mill (kg/tonne) Mercury recovered from mill (kg/tonne) Mercury recovered from retort (kg/tonne)6 Total mercury lost (kg/tonne) Miner 1 1.77 1.60 0.01 0.16 Miner 2 2.38 2.18 0.01 0.19 6 Assuming that amalgamation achieving 50% recovery of the gold in an ore grading 20g/t ore, that the dore consisted of 40% mercury and that the retort recovered 95% of the mercury. 70 3.3.2.3 Sample Collection Data on Mercury Lost In order to assist with mercury emission estimates, a representative sample was collected from both muller mi l l tails and the cyanidation final tails at the centre. The muller mi l l tails had a mercury grade of 237 ppm, and the cyanidation tails had a grade of 108 ppm. The muller mi l l tails would indicate a loss of 0.237 kg/tonne of ore processed. These numbers are also in line with L i n et al (1997), who reported A S G M tailings in Dexing County averaged 100-300 ppm mercury. 3.3.2.4 A S G M Mercury Release Estimates Table 3.3 compares the mercury estimates from sections 3.3.2.1-3.3.2.3, and estimates how much mercury is lost per muller mi l l per day, as well as the total for all six mills in the valley, processing from 10 to 15 tonne/day. Table 3.3 Muller M i l l Mercury Loss Estimates For One Muller Mill Hg Loss 10 tonne/d ore 15 tonne/d ore (kg/tonne) kg of Hg/day kg of Hg/day Miner 1 0.167 1.7 2.5 Miner 2 0.2 2.0 3 Tails Grade 0.237 2.4 3.6 Mi l l e r Estimate 0.5 5 7.5 A s can be seen, the mi l l operator's estimate is far higher than the other two sources; it is also the least verifiable of the sources and is thus discounted as inaccurate. Based on the remaining two estimates, the total amount of mercury discharged per muller mi l l would range from 1.7 kg to 3.6 kg/day. While the muller mills may be able to process up to 15 tonnes/day, the average is probably closer to 10. The primary reason for this is that ore is not processed continuously; miners bring in lots often smaller than 10 tonnes and it takes time to switch from one miner's ore to the next. The best mercury loss estimate is from approximately 1.7 to 2.4 kg per mill/day. 71 Estimating the average gold grade as 20 g/t, and a muller mi l l gold recovery rate of 60%, the average ratio of mercury lost per gold recovered would be 14:1 to 20:1. This estimate is conservative, as the recovery rate is probably significantly lower than 60%. The lost mercury would report primarily to the tails where it would be subjected to cyanidation. A s can be seen from the final tailing mercury levels, much of the mercury is recovered during cyanidation and cemented on the zinc strips; however, this mercury is later lost during the smelting process. In addition, a significant portion of the mercury remains in the tailings, as seen in section 3.3.2.3. One processing centre, hosting a single muller mi l l , had an area of approximately 3000 metres squared. The amalgamation and cyanidation tailings were spread over an area of approximately 0.1 kilometres squared. Assuming that six mills operate for at least 150 days per annum, the total mercury discharged in the community would range from 1.5 to 2.2 tonnes of mercury per annum. Furthermore, for many years, the amounts may have been twice as much, since there used to be at least 12 muller mills. 3.3.3 ASGM Health and Environmental Impact "Go ld Mountain's" people and environment have endured artisanal gold mining operations for over 70 years. Gold mining, however, is far from being the only local source of pollution. Farming, with its extensive use of fertilizers on the hillsides, creates large erosion and runoff problems, and the widespread use of coal not only effects air quality but also contributes to acid rain. The two-cycle local haul trucks are also a major source of air and noise pollution. Iron mines and sand and gravel quarries and dimension stone quarries all lead to an assortment of health and environment problems. However, mercury and cyanide probably create the most serious local problems, and since there are relatively few sources, their effects could be mitigated. 3.3.3.1 Health Impact A s in the initial study on A S G M in China (Section 2.2.2), during site visits and interviews, miners were observed for overt indications of mercury related illnesses. 72 Unlike the initial study, one miller interviewed at a processing centre stated that he had been diagnosed with health problems due to mercury exposure. The miller frequently handled mercury during his workday, through the cleaning of the muller mi l l , the retorting of the amalgam, and the smelting of the cyanidation zinc strips. His illness had reached a point where he was having difficulties with muscle tremors, a symptom of serious mercury vapour poisoning. He travelled to Beijing to receive more sophisticated medical treatment than was available in "Go ld Mountain", where his hospital urine test showed high levels of mercury. He was treated, possibly with a chelating agent, and recovered enough to return to work. Miners interviewed reported that retorts were introduced around 1990, and that the use of retorts had dramatically reduced the number of acute mercury vapour poisonings in "Gold Mountain". 3.3.3.2 "Go ld Mountain" Aquaculture The significant mercury losses in the valley raise the possibility of bioaccumulation of mercury in fish in the two local reservoirs. Both reservoirs were visited in order to evaluate this risk. The reservoirs were constructed primarily to create a stable source of water for downstream agriculture. The larger reservoir had several floating fish pens located in deeper areas of the water. The smaller reservoir was stocked, without using pens. Three varieties of carp were farmed. The carp were fed pellets made from flour, bone, and fishmeal, processed elsewhere. N o hatcheries were observed locally - fish farmers probably bought fingerlings from nearby regions. Indigenous fish were virtually extinct. There was one local variety of carp, but it was carnivorous and local fishermen attempted to wipe out the species to protect their own stock. Anecdotally, the locals have stated that they frequently find deformed fish in the area. 73 The lack of indigenous fish may have been assisted by the A S G M operations. Around 1985 cyanide use became widespread in the area, and the poor treatment of cyanidation releases may explain the death of fish species in the streams and reservoirs. Much of China is heavily polluted due to a wide array of industrial activities; various other explanations are possible. 3 .4 Discussion Clearly "Gold Mountain" has significant mercury releases due to gold amalgamation, raising two key questions. First, what are the impacts of these releases and second, i f these releases are dangerous, how can they be reduced? 3.4.1 "Gold Mountain's" Processing Practice There are two main issues of concern with regard to "Gold Mountain's" amalgamation process. First is the use of muller mills to amalgamate all of the gold ore processed (whole ore amalgamation), not just a concentrate. The second concern is the use of cyanidation following whole ore amalgamation. Whole ore amalgamation by definition puts all of the ore in contact with mercury, vastly increasing the amount of material contaminated. The key to reducing mercury releases is to encourage the amalgamation of only gravity concentrates, as then only a small portion of the ore is brought into contact with mercury. The high levels of mercury releases demonstrate a need to determine if alternate technologies can be introduced into A S G M processing centres in China, further discussed in Chapters Four and Five. The practice of following amalgamation with cyanidation raised the question of whether cyanidation increases the mobility of mercury in the amalgamation tailings, further discussed in Chapter Four. 3.4.2 Health and Environmental Impact of Mercury Releases in "Gold Mountain " 3.4.2.1 Health Impact of Mercury Vapour Mercury vapour pollution continues to be a concern, as demonstrated by the mercury illness described in Section 3.3.3.1. It is not clear that miners always use retorts, or if 74 miners using the "Go ld Mountain" Retort wait until it cools before removing the l id. The impact of mercury vapour arising from the smelting of the gold-bearing zinc strips, or during purification of gold dore may be a significant factor. While it was not possible to conduct mercury analysis of locals in "Go ld Mountain", L i n et al (1997) undertook such studies in Jiangxi Province. They found urine samples from miners in Dexing averaged 64.73 ug/L and ranged from 3.6 to 539.6 ug/L, with 91% of the 185 men, women, and children surveyed reporting above 20 ug/L. These numbers are far above the 4 ug/L that the World Health Organization (1991) considers normal. However, there is no indication that the urine samples were collected over a 24-hour period, as is standard for urine analysis, or that urine creatinine levels were sampled (Wilhelm et al, 1996). Ideally for urine tests, urine should be collected over a 24-hour period in order to minimize the effect of dilution from drinking fluids. In order to mitigate the awkwardness of 24-hour urine collection, creatinine levels can be analyzed in addition to mercury. Emissions of creatinine, a breakdown product of creatine, fluctuate little over a 24-hour period, thus creatinine can be use to normalize the concentration of mercury emissions. Other non-intrusive procedures have been suggested as possible indicators of mercury poisoning. These include questionnaires, observing physical behaviours such as touching their finger to their nose, memory tests, or reproducing simple line drawings (Veiga and Baker, 2004). Care needs to be taken to avoid cultural bias with these sorts of tests. Tests such as these were not undertaken during this study, but are recommended for future work. 3.4.2.2 Bioaccumulation With at least 1.5 tonnes of mercury being released annually from " G o l d Mountain", mercury bioaccumulation in fish in the local reservoirs is a real potential. This potential effect appears to be minimized, as the fish farmed in the local reservoirs are fed food pellets made outside of the "Go ld Mountain" area. Fish are a luxury food item, hardly a staple, even for wealthier people. In addition, due to the "impoundment effect," bioaccumulation of mercury occurs widely in reservoirs with no obvious source of 75 mercury (Stokes and Wren, 1987). Thus, even i f bioaccumulation was demonstrated by testing fish samples from fish farms, it might not be attributable to A S G M , and fish consumption might be low enough not to lead to significant health problems due to methylmercury. Regardless, given the amount of mercury being released into the environment, it is likely that conditions favourable to bioaccumulation would occur somewhere downstream. 3 . 5 Conclusion The "Go ld Mountain" case study clearly verified that artisanal and small-scale gold mining in China releases significant amounts of mercury. A l l stages of the gold mining process were detailed, including a retort design that seems to be unique to "Go ld Mountain". A review of the cost structure for gold mining in "Go ld Mountain" found mercury to be only 2.8 % of the overall cost. M i l l operator interviews, records, and assays of m i l l tailings led to the conclusion that the ratio of mercury lost per unit of gold recovered for muller mills is from 14:1 to 20:1. A sample of mi l l tailings was found to have a mercury concentration of 237 ppm, and a sample of cyanidation tailings had a concentration of 108 ppm. The small community of "Gold Mountain" alone was estimated to release at least one and a half tonnes of mercury per annum. Mercury vapour was found to have caused at least one miner to become i l l , although the impact of methylmercury was less clear. It was recommended that non-intrusive procedures be used to highlight possible indicators of mercury poisoning. The case study demonstrated that key areas of concern are amalgamation of whole ore using the muller mi l l and the smelting of cyanidation zinc strip concentrates without fume hoods. 76 4 Study Three: "Gold Mountain" Alternative Technology Test Program 4.1 Introduction The high levels of mercury releases found in the "Gold Mountain" case study, and the associated health and environmental risks due to those releases, demonstrated a need to evaluate technology alternatives to amalgamation. In cooperation with a "Gold Mountain" processing centre, three samples were sent to U B C to undertake mineral processing test work. These three samples were not from the same ore body and are not uniform, thus provide insight into the variability of ore that the milling centre must process. This test program also provided an opportunity to verify the results of the initial research on A S G M in China and the "Gold Mountain" case study, and to examine the impact of cyanidation following amalgamation. This chapter describes the objective, methodology and results of test work undertaken on three "Gold Mountain" samples. 4.1.1 Sample GM1 - "Gold Mountain " Feed Ore The first sample consisted of 14.6 kg of feed ore crushed to roughly minus one inch. The feed sample was to come from one of the several mines that bring ore to be processed at the centre. The feed sample was oxidized (reddish ih colour), and contained some organic material and bits of plastic, probably from the shipping sack. 4.1.2 Sample GM2 - "Gold Mountain " Muller Mill Tails The intermediate sample of 11.6 kg was taken from the muller mill overflow, before cyanidation; thus the ore had been ground with mercury and lime. The sample was grey in colour and slightly damp with some course material. It also contained some organic material and bits of plastic. 4.1.3 Sample GM3 - "GoldMountain" Cyanidation Tails The final sample (13.9 kg) was the tails from the cyanidation of the muller mill overflow. The sample was light grey in colour and slightly damp with some course material, in addition to some organic material and bits of plastic. 77 4.2 Methodology The objectives of the test work were the following: • To verify the mercury losses reported in Chapter Three; • To evaluate the potential alternative technology to replace the current amalgamation/cyanidation techniques. A Knelson Concentrators lab-scale M D 3 centrifugal concentrator ( K C - M D 3 ) , in conjunction with cyanidation, was studied as the alternative technology; and, • To explore the effect of cyanidation of mercury tailings on mercury solubility. 4.2.1 Sample GM1 - "Gold Mountain " Feed The four main tests undertaken on the G M 1 sample are outlined in Figure 4.1. Sample GM1: Feed Test W o r k Homogenize and split 60g Sulfur, Copper, and Tellurium Assay 14,600 g sample (GM1-A) r (GM1-B) r (GM1-C) (GM1-D) 8012g Sample to Knelson Grind 1971g to 120 microns Grind 1985g to 78 microns Grind 2000g to 291 microns I f 4 Triple Pass through MD 3.5. Each pass followed by a grind Single Pass through Knelson MD 3 Single Pass through Knelson MD 3 Single Pass through Knelson MD 3 1 1 r • • 48 hour leach test 48 hour leach test 257 hour leach test r Conduct size analysis and submit samples for assay Conduct size analysis and submit samples for assay Submit samples for assay Figure 4.1 G M 1 Test Program 78 4.2.1.1 Sample Preparation and Characterization Once received in Vancouver, the G M 1 sample was cleaned of plastic and organic debris. The entire sample was crushed to -10 mesh (-1 mm) using a cone crusher. Using riffles, two representative 30-gram samples were taken for assay. One sample was sent to A C M E Analytical Laboratories to assay for copper, nickel, cobalt, iron, arsenic, antimony, mercury, and tellurium, using I C P - M S . The sample was assayed to determine the background level of mercury in the feed ore. The other substances were assayed to determine potential problems with gold recovery during cyanidation. The second sample was sent for a total sulphur level assay, in order to determine the sulphides content of the ore. 4.2.1.2 Gravity Recovery Test The first test is called a Gravity Recoverable Gold (GRG) test. It was conducted with a centrifugal concentrator and provides an indication of how much gold could be recovered using only gravity methods. The methodology for the first test ( G M 1 - A ) was as follows: 1) A n 8 kg sample was processed through a 3" Laboratory Knelson Concentrator ( K C - M D 3 ) at a fluidization water flow rate of approximately 3.5 litres/minute and a rotation rate of 60 times the force of gravity; 2) During the test, sub-samples of the tailings stream were collected for assays; 3) A t the end of the concentration stage, the concentrate was washed from the inner cone of the K C ; 4) The K C concentrate was panned to produce a pan concentrate and pan tailings (middlings) sample; 5) The tailings were then ground in a lab-scale rod m i l l and steps 2 to 4 were repeated, followed by another grind and a final third pass through the K C - M D 3 ; 6) The concentrate and tailings samples were labelled, dried, weighed and sent to an independent local lab for fire assay to extinction. 7 9 This methodology is based on the concept that progressive size reduction allows the determination of gold liberated at finer grinds, without over-grinding or smearing any coarse gold. 4.2.1.3 Gravi ty / Cyanidation Tests In order to evaluate using gravity concentration in conjunction with cyanidation to replace amalgamation followed by cyanidation, the following three tests were undertaken. The second and third tests ( G M 1 - B and G M 1 - C ) were run side-by-side to compare the effects of different grind sizes on recovery. 1) The first 2 kg sample was ground 18 minutes to 80% of the material passing 120 um, and the second 2 kg sample was ground 50 minutes to 80% passing 78 um, using a lab-scale rod mi l l ; 2) Each sample was run through a K C - M D 3 at a fluidization water flow rate of approximately 3.5 litres/minute and a rotation rate of 60 times the force of gravity; 3) A t the end of the concentration stage, the concentrate was washed from the inner cone of the K C ; 4) The K C concentrate was panned to produce a pan concentrate and pan tailings (middlings) sample; 5) The concentrates were labelled, dried, weighed and sent to an independent local lab for fire assay to extinction; and, 6) The tailings were dried and weighed, and then proceeded to cyanidation. Samples G M 1 - B and G M 1 - C were subjected to cyanidation in bottle roll tests, run as follows: 1) The K C - M D 3 tailings were slurried in a bottle to approximately 40% solids; 2) Lime was added to the slurry and the p H was adjusted to 10.5; 3) When the p H was stabilized, cyanide was added to give a concentration of 1 g/L sodium cyanide; 4) Small aliquots of slurry were removed at 24-hour intervals and the p H and cyanide concentrations determined; 80 5) Cyanide and lime were added as needed to maintain the initial p H and cyanide levels, and the aliquots were collected for gold analysis; and, 6) After 48 hours, the slurry was filtered, washed, and dried. The cyanidation residue was riffled and representative samples of approximately 500 grams were analyzed for size distribution and submitted for gold assay. The fourth test (GM1-D) was run as follows: 1) A 2 kg sample was ground at 65% solids for 3 minutes in a stainless steel rod m i l l ; 2) The sample was run through a K C - M D 3 at a fluidization water flow rate of approximately 0.8 litres/minute and a rotation rate of 60 times the force of gravity; 3) A t the end of the concentration stage, the concentrate was washed from the inner cone of the K C ; 4) The K C concentrate was panned to produce a pan concentrate and pan tailings (middlings) sample; 5) The concentrates were labelled, dried, weighed and fire assayed to extinction; and, 6) The tailings were dried and weighed, and then proceeded to cyanidation. Cyanidation proceeded using the bottle roll method, as follows: 1) A representative 480g sample of the K C - M D 3 tailings was slurried in a bottle to approximately 35% solids; 2) Lime was added to the slurry and the p H was adjusted to 10.5; 3) When the p H was stabilized, cyanide was added to give a concentration of 0.5 g/L sodium cyanide; 4) Small aliquots of slurry were removed at several intervals and the p H and cyanide concentrations determined; 5) Cyanide and lime were added as needed to maintain the initial p H and cyanide levels, and the aliquots were collected for gold analysis; and, 6) After 257 hours, the slurry was filtered, washed, and dried. The cyanidation residue was riffled and representative samples of approximately 500 grams were analyzed for size distribution and submitted for gold assay. 81 4.2.2 Sample GM2 - "Gold Mountain " Muller Mill Tails The test work undertaken on the GM2 sample is outlined in Figure 4.2. Sample GM2: Intermediate Test Work Homogenize and split 11,600g sample Size Analysis: 600 g T 60g Sulfur, Copper, and Tellurium Assay 4710g - 2 passes through MD3 Submit samples for assay Flask Shaking Test 3X50g Regrind sample Filter and Assay 3 r d Pass through MD3 Sample and Assay Figure 4.2 GM2 Test Program 4.2.2.1 Sample Characterization The GM2 sample was cleaned of plastic and organic debris, then screened at 1 mm, dried and weighed. The plus 1 mm material was assayed for gold and mercury. The sample was homogenized with a riffle, and then a dry 650g representative sample was screened with a Roto-Tap for 15 minutes to determine the size distribution. Representative 15-30g samples were pulverized, sealed, and sent for gold and mercury assays to A C M E Laboratories in Vancouver. Two additional representative 30-gram samples were taken for assay. One sample was sent to A C M E Analytical Laboratories to assay for copper, nickel, cobalt, iron, arsenic, 82 antimony, mercury, and tellurium, using I C P - M S , and the second sample was assayed for total sulphur. 4.2.2.2 Gravity Recovery Test Sample G M 2 was tested using a K C - M D 3 as follows: 1) A 5 kg sample was processed through a K C - M D 3 at a fluidization water flow rate of approximately 3.5 litres/minute and a rotation rate of 60 times the force of gravity; 2) A t the end of the concentration stage, the concentrate was washed from the inner cone of the K C and the K C concentrate was panned to produce a pan concentrate and pan tailings (middlings) sample; 3) The tailings were passed through the Knelson a second time, and during the test, sub-samples of the tailings stream were collected for assay; 4) Step 2 was repeated; 5) The tailings were ground for two minutes in a lab-scale rod mi l l ; 6) The tailings were passed a third time, and during the test, sub-samples of the tailings stream were collected for assay; 7) Step 2 was repeated; 8) The concentrate and tailings samples were labelled, dried, weighed and sent to an independent local lab for fire assay to extinction. 4.2.2.3 Mercury Solubility Test The combination of mercury use followed by cyanidation gives rise to concern, as residual mercury may be oxidized during the cyanidation process, allowing the mercury to be more easily methylated by bacterial processes. In order to compare the solubility of mercury in the tailings before and after cyanidation, a shake flask test was undertaken. This was completed as follows: 1) Two representative 50g samples of G M 2 were placed in a 250 m L Erlenmeyer flask with 125 m L of distilled water; 2) The flask was shaken at medium intensity on a shake table for 24 hours; and 83 3) The solution was then suction filtered through glass fibre filter paper and submitted for mercury assay using a P S Analytical Mil lennium Mer l in System. This method was adapted from Hinton (2002). 4.2.3 Sample GM3 - "GoldMountain" Cyanidation Tails The test work undertaken on the G M 3 sample is outlined in figure 4.3. Homogenize and split 13,900 g sample Size Analysis: 650 g T Submit samples for assay Sample GM3: Tailing Test Work 4570g - 2 passes through MD3 T 60g Sulfur, Copper, and Tellurium Assay Flask Shaking Test 3X50g Regrind sample T Filter and Assay 3 r d Pass through MD3 I Sample and Assay Figure 4.3 G M 3 Test Program 4.2.3.1 Sample Characterization The G M 3 sample was characterized in an identical manner to G M 2 , except the initial screen sample was approximately 650 grams, not 600 grams. 4.2.3.2 Gravity Recovery Test The G M 3 sample was tested in an identical manner to G M 2 . 84 4.2.3.3 Mercury Solubility Test The G M 3 sample was tested alongside the G M 2 sample, in an identical manner. 4.3 Results 4.3.1 Sample GM1 - "GoldMountain"Feed 4.3.1.1 Sample Characterization The assay results of the initial sample characterization for G M 1 are displayed in Table 4.1. The high copper and sulphur levels, and the presence of tellurium, indicate the presence of copper sulphides and gold tellurides, which would inhibit cyanidation. The mercury level of 788 ppb can be taken as an indication of the background level of mercury in the gold ores. A s this is far less than the mercury concentration in the muller mi l l or cyanidation tails, the background mercury levels are considered negligible. Table 4.1 G M 1 Sample Characterization Cu (ppm) Ni (ppm) Co (ppm) Fe (%) As (PPm) Sb (ppm) Hg (ppm) Te (ppm) Sulphur (%) 4418.2 205.8 37.2 12.1 10.5 1.4 0.788 13.7 6.08 4.3.1.2 Gravity Recovery Test Table 4.2 G M 1 - A Gravity Recoverable Gold Test Mass Assay Dist'n Size Dist'n (um) Product (g) (%) Au (g/t) (%) P80 = 475 Pan Cone. 8.7 0.11 2817 7.0 Middlings 84.1 1.05 70.40 1.7 Sample Tails 169.3 2.11 40.80 2.0 P80 = 236 Pan Cone. 17.8 0.22 2520 12.8 Middlings 67.7 0.84 95.60 1.8 Sample Tails 122.0 1.52 33.30 1.2 P80 = 117 Pan Cone. 27.1 0.34 1882 14.5 Middlings 69.8 0.87 127.00 2.5 Final Tails 7445.5 92.93 26.60 56.5 Totals (Head) 8012 100.0 43.76 100.0 Knelson Cone. 275.2 3.43 514.6 40.4 85 A s can be seen in Table 4.2, the grade of the ore sample is calculated to be 43.76 g/t gold in test G M 1 - A . The test indicated that around 40% of the gold could be recovered by gravity methods. The 80% passing particle size for the initial pass was 475 um, 236 um for the second pass, and 117 um for the third. The gold recovered increased with each grind, suggesting that most of the gold is quite fine. The gold recovered would be relatively easy to upgrade, as most of the recovered gold reported to the pan concentrate. 4.3.1.3 Gravity / Cyanidation Tests The three gravity / cyanidation tests results can be seen in Tables 4.3, 4.4, and 4.5. While not reflected in the gravity recovery stage, the increased grind time seems to have improved G M l - C ' s cyanidation recovery over G M 1 - B . A s can be seen, the G M 1 - B and G M 1 - C tests' total recoveries are significantly less than G M 1 - D , primarily due to the significantly shorter leach time. The long leach times used by the miners in "Gold Mountain" thus seems justified. Table 4.3 G M 1 -B Test Results G M 1 - B 120 um Product Mass (g) Mass (%) A u (g/t) Au Dist'n (%) Pan Con 7.09 0.36 701.30 23.0 Pan Middlings 3.38 0.17 46.30 0.3 C N Solution (48 hr) n/a n/a n/a 25.5 Tails 1974.63 99.47 22.57 51.2 Total 1985.10 100.00 43.33 100.0 Total Recovery 0.53 48.8 Table 4.4 G M 1 - C Test Results G M 1 - C 78 um Product Mass (g) Mass (%) Au (g/t) Au Dist'n (%) Pan Con 28.06 0.36 2959.90 22.9 Pan Middlings 5.14 0.17 229.70 0.9 C N Solution (48 hr) n/a n/a n/a 30.3 Tails 1937.60 99.47 21.26 45.9 Total 1971 100.00 46.09 100.0 Total Recovery 0.53 54.1 86 Table 4.5 GM1-D Test Results GM1-D 291 nm Product Mass (g) Mass (%) Au (g/t) Au Dist'n (%) Knelson Con 112 5.60 321.50 36.28 C N Solution (257 hr) n/a n/a N/a 52.18 Tails 1888 94.40 21.26 11.54 Total 2000 100.00 49.63 100.00 Total Recovery 5.60 88.46 The mass of the gravity concentrate for samples GM1-B and GM1-C are unusually low, as the KC-MD3 concentrates are usually around 100 to 120 g. The GM1-B and GM1-C tests were conduced on an older KC-MD3, which sometimes rejected lighter material in the concentrate bowl during the process of shutting down the machine for cleaning. While the gold recoveries do not seem to have been significantly affected, a higher mass pull would likely have made the overall recovery higher, although the grade would be correspondingly lower. too 90 80 70 60 50 40 30 20 10 0 GM1-D Leach Time • GMl-D (291 um) 50 100 150 200 Leach Time (Hrs) 250 300 Figure 4.4 G M l - D Leach Time 87 The ore has an extremely slow leaching rate, as shown in Figure 4.4. Figure 4.4 does not include the gold recovered in the aliquots for GM1-D, which boosts the overall cyanidation recovery of the KC-MD3 tailings by 25.9%. Table 4.6 GM1-B and GM1-C Cyanidation Tailings Analysis (GM1-B) P80 120 nm (GM1-C) P80 78 nm Size (urn) Au (ppm) Au Dist'n (%) Au (ppm) Au Dist'n (%) +150 28.50 4.5 5.01 1.2 +63 28.45 51.8 17.05 15.2 +45 21.41 17.0 25.13 31.0 +38 18.25 8.7 26.34 27.4 -38 15.24 18.0 19.50 25.2 Total 22.57 100.0 21.26 100.0 G M 1 Cyanidation Tails A u Distribution too 90 ~ 80 0 Au P80 - 60 ".m 9 / / 1 • Au 1*80 -102 um — • — Au Dist 120 um 4 i a LU LMSI io urn i— i 10 100 1000 Size (|im) Figure 4.5 GM1-B and GM1-C Cyanidation Tailings Analysis The effects of additional grinding on the ore are shown in Table 4.6 and Figure 4.5. Grinding sample GM1-C down to 78 um significantly increased the gold recovery in the +45 um range. The cyanidation tailings from the GM1-B sample, with the coarser grind, show that the ore retained 52% of its gold in the -150 um to + 63 um size fraction, where sample GM1-C retained only 15% of its gold in the same fraction. The overall recovery 88 o f G M 1 - C was 5 4 . 1 % versus 4 8 . 8 % for G M 1 - B , p r imar i ly due to the improved cyanidat ion recovery as a result o f increased gr inding. 4.3.2 Sample GM2 - "Gold Mountain " Muller Mill Tails 4.3.2.1 Sample Character izat ion Af te r p i c k i n g out the obv ious ly a l i en mater ial , the + l m m oversize was 204 g, w i t h 46.7 g/t go ld and 1231 p p m mercury. T h i s oversize material was assayed for g o l d and mercury and inc luded i n the size and grade dis t r ibut ion data. Howeve r , it was not inc luded i n any further test work , as it was a relat ively sma l l amount, at 1.75% o f the total mass, and it w o u l d have needed to be crushed before it cou ld be passed through the centrifugal concentrator. The P 8 0 for sample G M 2 was found to be 180 u m . The back calculated assay for go ld was 41.42 g/t. F o r mercury, the back calculated assay was 2350 p p m . The size analysis for G M 2 showed that 5 2 % o f the go ld and 5 3 % o f the mercury were found i n the - 5 3 u m size fraction, representing 2 6 % o f the total mass. GM2 Size, A u , and H g Dis t r ibu t ion o n 1 y u 7 n 'w' / u / * .2 60 -CO m SA • Size Dist — • — Au Dist 1 A H g Dist CM i n . ON 20 -10 • 0 -1 0 1( Size )0 1000 ((im) Figure 4.6 G M 2 Size , G o l d , and M e r c u r y Dis t r ibu t ions A s shown i n F igure 4.6, the go ld dis t r ibut ion c lose ly matched the mercury dis t r ibut ion. 89 Table 4.7 GM2 Sample Characterization Cu (ppm) N i (ppm) Co (ppm) Fe As (ppm) Sb (PPm) Hg (ppm) Te (ppm) Sulphur (%) 41065 117 32 7.09 7.7 43.6 2350 5.3 6.34 Table 4.7 shows that GM2 has a high copper content, at 4.1%, demonstrating that samples GM1, GM2, and GM3 are not from uniform ore bodies. The tellurium in the GM2 sample is significantly lower than that in the GM1 sample, which indicates that either amalgamation is recovering some tellurium with the gold, or that the sample simply has less tellurium than GM1. The high copper and high sulphur content may indicate that the copper is in the form of sulphides, such as chalcopyrite. 4.3.2.2 Gravity Recovery Test As can be seen in Table 4.8, the GM2 test results calculate the grade of the sample to be 38.1 g/t gold, lower than the sample characterization assay results. The test indicated that around 43% of the gold could be recovered by gravity methods, which is surprising considering this sample had already been amalgamated. Table 4.8 GM2 Gravity Test Results Mass Assay Units Dist'n Size Dist'n Product (g) (%) Au (g/t) A u (%) P80= 144 jim Pan Cone. 30.9 0.66 1447 948.7 24.9 Middlings 62.0 1.32 157 206.7 5.4 Pan Cone. 13.3 0.28 537 151.6 4.0 Middlings 68.8 1.46 75.2 109.8 2.9 Sample Tails 305.7 6.49 23.4 151.9 4.0 P80= 100 nm Pan Cone. 24.7 0.52 336 176.2 4.6 Middlings 58.7 1.25 49.8 62.1 1.6 Final Tails 4145.9 88.02 22.8 2006.9 52.6 Totals (Head) 4710 100.0 38.1 3813.9 100.0 KC-MD3 Cone. 258.4 5.49 302 43.4 The size distribution for the initial pass was 80% passing 144 um, lower than the initial size distribution, as the plus 1 mm material was not included. After the two minute grind the P80 was 100 um. Over 30% of the gold was recovered in the first stage, suggesting 90 that additional grinding may not be economical. In addition, the gold recovered in the first pass would be relatively easy to upgrade, as most of the recovered gold reported to the pan concentrate. 4.3.2.3 Mercury Solubility Test A s this test was primarily undertaken to compare G M 2 and G M 3 ' s mercury solubility, these results are examined together in section 4.3.3.3. 4.3.3 Sample GM3 - "Gold Mountain " Cyanidation Tails 4.3.3.1 Sample Characterization G M 3 Size, A u , a n d H g D i s t r i b u t i o n tnn _ on yu an r " S O */ ' 0 .5 60 -CO CO m en > • Size Dist —•— A u Dist A Hg Dist Cu "fl 40 • a ™ <u Pm i n 00 xU i n 1U o -I 0 1( r Size l )0 10 fims) Figure 4.7 G M 3 Size, Gold, and Mercury Distributions After picking out the obviously alien material, the + l m m oversize was 305 g, with 1.31 g/t gold and 82 ppm mercury. This oversize material was assayed for gold and mercury and included in the size and grade distribution data. It was also not included in any further test work, as it was a relatively small amount at 2.18% of the total mass, and it would have needed to be crushed as well . The P80 for sample G M 3 was found to be 215 um. The back calculated assay for gold was 1.58 g/t. For mercury, the back calculated assay was 283 ppm. The size analysis for 91 G M 2 showed that 36% of the gold and 48% of the mercury were found in the -53 nm size fraction, representing 23% of the total mass. A s can be seen in Figure 4.7, the gold distribution was not closely tied to the mercury distribution. Clearly cyanidation successfully recovers gold in preference over mercury. Table 4.9 G M 3 Sample Characterization Cu (ppm) Ni (ppm) Co (ppm) Fe (%) As (ppm) Sb (ppm) Hg (ppm) Te (ppm) Sulphur (%) 9356 87.7 17 4.48 13.8 63.04 283 4.02 3.24 Table 4.9 indicates that copper grade is still relatively high, at almost 1%. Except for the arsenic and antimony content, the other metal contents were significantly lower than in G M 2 , suggesting that the leaching process removes a lot of the other metals, slowing the extraction of gold. 4.3.3.2 Gravity Recovery Test Table 4.10 G M 3 Gravity Test Results Mass Assay Units Dist'n Size Dist'n Product (g) (%) Au (g/t) Au (%) Pan Cone. 25.1 0.51 10.8 5.5 5.5 Middlings 59.3 1.21 2.03 2.4 2.4 Pan Cone. 37.6 0.76 5.19 4.0 3.9 Middlings 48.4 0.98 2.28 2.2 2.2 Sample Tails 90.1 1.83 1.33 2.4 2.4 P80= 121 Jim Pan Cone. 42.1 0.86 10.7 9.2 9.1 Middlings 44.9 0.91 2.46 2.2 2.2 Final Tails 4570.5 92.93 0.78 72.5 72.1 Totals (Head) 4918 100.0 1.01 100.5 100.0 K C - M D 3 Cone. 257.4 5.23 4.89 25.4 A s can be seen in Table 4.10, the G M 3 test results calculate the grade of the sample to be 1.01 g/t, lower than the sample characterization assay results. The test indicated that around 25.4% o f the gold could be recovered by gravity methods, which would perhaps be uneconomical considering the low grade. The size distribution for the initial pass was not recorded, but after the two-minute grind the P80 was 121 um. The recovered gold 92 was widely distributed between the pan concentrates and middlings, and the concentrate grades low, suggesting that the gold may not be easy to upgrade. 4.3.3.3 Mercury Solubility Test The intermediate mercury tails from the muller mi l l ( G M 2 , 2224 ppm Hg) were compared to final tails following amalgamation and cyanidation ( G M 3 , 285 ppm Hg), using two duplicate samples (A and B) . The mean value of G M 2 ' s solution had a concentration of 15.34 ug/L mercury and G M 3 ' s solution mean had a concentration of 3.78 ug/L mercury. However, since G M 2 ' s overall mercury levels are an order of magnitude higher than G M 3 ' s , the ratio of mercury dissolved to the mercury present in the sample was compared. A s seen in Table 4.11, approximately twice the ratio of mercury was dissolved from the cyanidation tails sample G M 3 , compared to the intermediate sample: 17.2 x 10"6for G M 2 and 31.1 x 10"6for G M 3 . Table 4.11 Mercury Solubility GM2 (2224ppm Hg) GM3| [285ppm Hg) Product Oig/LHg) Ratio of Hg (\ig) in solution to Hg in tailings (u.g) (Hg/LHg) Ratio of Hg (ng) in solution to Hg in tailings (u.g) Sample A 14.30 16.1 x 10"b 5.14 45.1 x 10"° Sample B 16.37 18.4 x 10"6 2.41 21.1 x 10"6 Mean 15.34 17.2 x 10"" 3.78 31.1 x 10"b 4.4 Discussion 4.4.1 "Gold Mountain " Samples Table 4.12 Grade Differences Between Test Samples SAMPLES Cu (%) Fe (%) Te (ppm) Hg (ppm) G M 1 - Feed Sample 0.44 12.11 13.71 0.78 G M 2 - After Muller , Pre Cyanide 4.11 7.09 5.30 2224 G M 3 - Final Tails 0.94 4.48 4.02 285 The three " G o l d Mountain" samples are clearly not from the same ore deposit, as is demonstrated in Table 4.12. While this presents difficulties comparing the test work, it shows the diversity of ores that a mineral processing centre in " G o l d Mountain" receives. 93 It also demonstrates that any alternative technology introduced needs to be robust enough to handle a diversity of ores. The mercury levels in the samples are significantly higher than the initial samples collected during the case study research; 2350 ppm vs. 237 ppm for the muller mi l l intermediate samples (GM2) , and 283 ppm versus 108 ppm for the final cyanidation tails (GM3) . This verifies that the estimates of mercury releases made in section 3.3.2.4 are certainly conservative. Detailed studies of the mineralogy of samples G M 1 - G M 3 was not undertaken. While this may have provided further information on how to optimize the recovery of gold in these samples, the variability of the samples suggest that such an optimization might not be useful for the range of ores brought into a processing centre. The miners need inexpensive, simple, and robust technology capable of achieving good gold recoveries from a range of ores. 4.4.1.1 G M 1 The G M 1 test work verified that a significant portion of the ore is amenable to gravity separation, with a G R G value of 40%. The test work also showed that the long leach times used are indeed necessary, and that increased grinding would improve both gravity and cyanidation recoveries. The miners had indicated that some of the gold ores were especially refractory, and that they were interested in new technologies to improve recovery. It is quite possible that the G M 1 sample was one of these ores, due to the high tellurium grade and the high percentage of gold finer than 63 um. That after almost 11 days of cyanidation in a bottle rol l , the tailings grade was still 5.6 g/t gold, indicates that the miners may have had significant difficulties achieving high recoveries. This slow leach time may be due to the copper and tellurium in the ore. Miners may improve recovery by adding cyanide in stages. A s mentioned in section 3.3.1.4, miners only add cyanide and lime at the beginning of cyanidation. The miners could float and/or roast the ore to improve recoveries, but this would likely be beyond the miners' technical skills or financial capability. 94 4.4.1.2 G M 2 The high gold grade of G M 2 (38.1 g/t), and the exceptionally high mercury levels (2224 ppm) suggests the amalgamation process was not especially effective in this case. Usual whole ore amalgamation tails have mercury levels in the range of 200 to 500 ppm (Section 3.3.2.3; Veiga, 1997). Perhaps the feed ore had a particularly high grade (over 60 g/t gold), and the miners had some inkling of this, thus added significant amounts of mercury in an attempt to increase recovery. The tight correlation between the mercury and gold size distribution (Section 4.3.2.1) indicates that amalgamation in the muller mills may be extremely effective; potentially all o f the gold is essentially amalgamated. However, this sample is from the muller mi l l tailings; clearly the amalgam is not being effectively recovered from the mi l l . While "Gold Mountain" miners stated that amount of gold was in the range of 40-60%, it seems unlikely in this case. While the muller mi l l may be an efficient and inexpensive comminution and amalgamation device, it may not excel at gold recovery. This is corroborated by the high percentage of gold recovered by the K C - M D 3 in the first pass. One of the prime causes for this may be that muller mills can flour the mercury, grinding it fine enough that it becomes largely inactive due to impurities and its inherent surface tension. Another reason may be that cleaning the muller mills by only removing the material found at the bottom of the muller mi l l trough is ineffective. Perhaps special amalgamation plates (Veiga and Gunson, 2004) would effectively remove much of the amalgam in the overflow. Presumably much of this excess mercury would be recovered using cyanidation and volatized during the zinc strip smelting, to the hazard of the miners. 4.4.1.3 G M 3 Sample G M 3 demonstrates that on some ores, the "Go ld Mountain" processing centres can be quite efficient at gold recovery. The low gold grades (—lg/t) and the poor results on the gravity tests (25.4% after three passes) indicate that their process (or at least their cyanidation) was effective, even with a high copper grade to inhibit cyanidation, unless the initial grade of this ore was quite low. Reprocessing tailings may not be an attractive option in " G o l d Mountain". The lower grade of mercury in the cyanidation tails compared to the amalgamation tails (283 ppm versus 2224 ppm), suggests that much of 95 the mercury from the amalgamation process is recovered during cyanidation. This recovered mercury is subsequently volatized during the smelting of zinc strips, described in section 3.3.1.4. Obviously, uncapped tailings dumps often located beside streambeds, with mercury concentrations 283 ppm, are a significant risk to the local watershed. 4.4.2 Mercury Solubility The United States Environmental Protection Agency's ( U S E P A , 2002) guidelines state two figures for freshwater mercury contamination. Its Criteria Maximum Concentration ( C M C ) is an "estimate of the highest concentration of a material in surface water to which an aquatic community can be exposed briefly without resulting in an unacceptable effect." Its Criterion Continuous Concentration ( C C C ) is an "estimate of the highest concentration of a material in surface water to which an aquatic community can be exposed indefinitely without resulting in an unacceptable effect." The C M C and C C C for mercury in freshwater is 1.4 ug/L and 0.77ug/L respectively ( U S E P A , 2002). Metallic mercury is a relatively stable substance, so significant dissolution shows the presence of soluble species. Thus the G M 2 and G M 3 ' s dissolution concentrations of 15.34 and 3.78 ug/L mercury give rise to concern. The leaching kinetics of mercury during cyanidation are fast, but slower than those of gold. A s a result, only part of the mercury is leached, the rest remaining with the tailings. In gold leaching conditions, mercury forms [Hg(CN)4] ", through the following equation: H g 2 + + 4 N a C N = Hg(CN) 4 2 " + 4 N a 2 + [Hg(CN)4] 2 " can easily form complexes with other substances (Flynn, 1995). A s the tailings are disposed of beside the local stream and no effort is made to safely manage the tailings, rainwater could easily mobilize this mercury. Hinton et al (2003a) suggest a number of options for remedial solutions for mercury-contaminated sites. The difficulty is that funding probably would not be available for clean up. Foremost efforts should be put toward helping existing operations use less damaging methods. 96 4.5 Conclusion The test work completed verified that high levels of mercury were being released from "Gold Mountain's" processing centres. The muller mi l l tailings' mercury concentration was 2224 ppm, and the cyanidation tailings' concentration was 285 ppm mercury, from a background mercury concentration, as indicated by the feed ore, of only 0.78 ppm. The three samples were from different deposits and show the diversity of ores that mil l ing centres must process. Test work completed on the feed sample, grading about 43 g/t gold, indicated that around 40% of the gold was recoverable by gravity methods. A n eleven-day cyanidation bottle test showed that the leaching rate of the ore is extremely slow, perhaps due to the high levels of copper (0.44%) and tellurium (13.71 ppm) in the ore. Gravity methods could be introduced to replace whole ore amalgamation, although amalgamation of gravity concentrates may still be necessary to produce a saleable product for the miners. Increased comminution would help the miners achieve higher recoveries and potentially reduce cyanidation times. The muller mi l l tailings sample, with a gold grade of 38.1 g/t, indicated that amalgamation had not effectively recovered gold from the initial ore, especially as 43% of the gold in the tailings was recovered, 30% in the first pass without additional grinding. The cyanidation tailings sample, with a gold grade of 1.0 g/t, indicates that for this ore at least, the processing centre effectively recovered most of the gold. The mercury concentration indicates that much of the mercury from the muller m i l l is removed by cyanidation, to be volatilized during the zinc strip smelting process. The remaining mercury in the tailings also represents a significant risk to the local watershed. In addition, the mercury solubility tests indicate that cyanidation, following amalgamation, may increase the solubility of the mercury in the tailings. The miners should be encouraged to abandon their practice of whole ore amalgamation, although amalgamation of gravity concentrations may continue to be necessary. 97 5 Mercury Releases, Alternatives, and Implementation A s demonstrated in the previous three chapters, China's artisanal and small-scale gold mining industry releases significant amounts of mercury. This chapter brings together the results of the previous chapters to estimate the extent of mercury released by A S G M , discusses potential technology alternatives to amalgamation which could reduce those releases and examines some issues surrounding alternative technology and technology transfer. 5.1 China's Mercury Releases A S G M represents one of the main sources of mercury releases in China, but is by no means the only source. First, the releases due to A S G M are estimated, followed by a brief discussion of other sources of mercury. 5.1.1 Estimating China's ASGM Mercury Releases The following mercury emission estimate for A S G M is based on: • The estimate of China's annual gold production due to A S G M ( A S G M A u ) ; • Estimates of the methods of A S G M gold production; muller mills, ball mill/mercury sluice combinations, and placer methods; and, • Estimates of the ratio of mercury lost to gold recovered by these different methods. Thus, the annual mercury emission from Chinese A S G M production is estimated from the following equation: H g A u - A S G M A u * ( M M A u % * M M R + B S A u % * B S R + P M A u % * P M R ) Where: H g A u is the amount of mercury released per annum due to A S G M activities; A S G M A u is the amount of gold produced per annum by A S G M ; M M A u % is the percentage of gold produced by A S G M operations using muller mills (Section 2.3.1.1); M M R is the ratio of mercury lost to gold recovered for muller mills; B S A u % is the percentage of gold produced by A S G M operations the ball mill/sluice combination (Section 2.3.1.2); 98 B S R is the ratio of mercury lost to gold recovered for ball mill/sluice operations; P M A u % is the percentage of gold produced by A S G M operations exploiting placer deposits (Section 2.3.1.3); and P M R is the ratio of mercury lost to gold recovered from placer deposits. The annual A S G M gold production estimate ( A S G M A u ) is 61 tonnes per annum, based on the calculation in section 2.4.2.2. The estimates of the percentages of different A S G M gold production methods are the least certain of these estimates. Ba l l mi l l / sluice operations were found widely around China, as discussed in Chapter Two, and have the capacity to mi l l large amounts of ore. He et al (2000) demonstrated that this is the main amalgamation method for the smaller state-owned gold operations using amalgamation. There appeared to be large numbers of muller mills , but they were more localized, and thus are given a smaller share of total production. Placer gold makes up around 13% of China's total gold production and often producers are small-scale operations (Zhou et al, 2002). Estimates of the ratio of mercury lost to gold recovered by method are detailed in Table 5.1. The muller m i l l estimate ratio ( M M R ) Hgiost:Auproduced ranges from is 14:1 to 20:1, as demonstrated in section 3.3.2.4. The ball m i l l / sluice ratio (BSR) range was determined as follows. He et al (2000) provide data for 33 ball mi l l / sluice operations, and claim that they consumed mercury at a Hgi0st:AuProduced ratio of 6.35:1. On average, 75% of the mercury consumed was later recovered using retorts and active carbon filters, leaving a ratio of 1.59:1. While this is probably representative of the state-owned amalgamation mills, it may be an underestimate in the case of the smaller A S G M operations. Section 2.3.1.4 estimated the ratio as 3.33:1, based on interviews. However, mercury losses from a similar copper plate amalgamation process described by Veiga and Gunson (2004) revealed a Hgiost:Auproduced ratio of 1.5:1-2:1. A s He et al (2000), and Veiga and Gunson (2004) produced more verifiable results than the interviews could have done, the B S R used for the low range is He et al's average of 1.59:1 and the high range is 2:1. The ratio used for the placer operations (POR) is 1. A s no whole ore amalgamation placer 99 operations were observed in China, it was assumed that they only amalgamate concentrates. For all of these methods, the actual ratios vary widely with equipment quality, human ability and ore variation. Table 5.1 Mercury to Gold Loss Ratio for China Unit Abbreviation Mercury Lost: Gold Produced Source Muller M i l l s , China M M R 1 4 - 2 0 : 1 Section 3.3.2.4 Ba l l mi l l - Sluice, China B S R 3.33 : 1 Section 2.3.1.4 Bal l mi l l - Sluice, China B S R 1.59 : 1 He etal, 2000 Amalgamation Plate ~ B S R 1 .5 -2 : 1 Veiga and Gunson, 2004 Only Concentrate P M R 1 : 1 Veiga, 1997 Estimates of the total amount of mercury released due to A S G M in 1997 and 2001 are outlined in Table 5.2. For 2001, the low estimate is 237.6 tonnes of mercury per annum and the high estimate is 652.1 tonnes. Table 5.2 Chinese Annual Mercury Releases attributable to A S G M Factor 1997 20( )1 L o w High L o w High A S G M A u 58 87.5 61 92.5 M M A u % 20 30 20 30 M M R 14 20 14 20 B S A u % 50 45 50 45 B S R 1.59 2 1.59 2 P M A u % 30 15 30 15 P M R 1 1 1 1 H g A u 225.9 616.9 237.6 652.1 The ratio of mercury lost to gold recovered seems to be the standard method used for accounting for mercury losses (Veiga, 1997). However, it is an imperfect standard for two main reasons; it does not account for mercury lost when no gold is recovered, and it assumes that miners w i l l accurately report the amount of gold they are recovering. It is easier to learn how much mercury is being lost per day per tonne of ore processed than it is to discover its ratio to gold recovered, as the former does not require miners to be 100 honest in reporting their gold recovery and can be verified by assaying tailings. Ideally the amount of mercury that miners buy each month could be recorded. Table 5.3 L o w Estimate of A S G M Gold Production and Mercury Releases (2001) Method Au Production (tonnes/a) Au Production (%) Hg Releases (tonnes/a) Hg Releases (%) Muller mi l l 12.2 20 170.8 71.9 Ba l l mi l l / sluice 30.5 50 48.5 20.4 Placer methods 18.3 30 18.3 7.7 Total 61 100 237.6 100 A s can be seen in Table 5.3, by far the largest contributors to mercury releases are muller mills, with 20% of A S G M production and 72% of the mercury releases, even using the low end of the range of Hgi o s t:Aup roduced ratios. Muller mills are clearly the area where efforts at reducing mercury releases should be concentrated. 5.1.2 Other Sources of Mercury In addition to A S G M , China has several other sources of mercury releases. He et al (2000) described how China's gold mines release around 20 tonnes of mercury per annum. The mines covered in the article are most likely the state-owned mines that still use amalgamation, and generally would not overlap with the mercury releases described above. China's massive coal mining industry emitted around 303 tonnes of mercury from coal combustion in 1995 alone (Wang et al, 2000). Artisanal and small-scale coal mines contributed around 127 tonnes of mercury in 1997 (Gunson and Veiga, 2004). Feng and Hong (1999) determined that most of this mercury was associated with sulphides in the coal and could be removed by basic coal cleaning, at least in Guizhou Province. In 2001, China produced roughly 1089 mil l ion tonnes of coal, but due to the central government's policy of shutting down A S M coal mines, their share of production has dropped from 47% in 1996 to 23% in 2001 (Andrews-Speed et al, 2003b). Assuming an average 101 mercury content of 0.22 mg/kg in China's coal, A S M coal mines contributed around 56 tonnes of mercury and formal coal mines contributed 240 tonnes of mercury in 2001. A s detailed in Gunson and Veiga (2004), some 50 A S M mercury operations lose around 28 tonnes of mercury per annum (in addition to their mercury production). The formal mercury-mining sector would also contribute to mercury releases; however, apparently most of the state-owned mercury mines have now been shut down (Maxson, 2004). In 2001, China's state-owned mercury mining sector produced approximately 190 tonnes of mercury, down from 830 tonnes in 1997 (Tse, 2002). Assuming a 90% recovery rate of mercury, the state-owned mercury mines released around 19 tonnes of mercury in 2001. The mines' production should not be considered releases, since much of that mercury may make it to A S G M operations or mercury polluters and would thus lead to double counting. There is a wide variety of other sources of mercury pollution including: • Chlor-alkali plants (1400-2700 tonnes from 1952-2000, Q i et al, 2000); • Batteries • O i l product combustion; • Cement production; • Lead production; • Zinc production; • P ig iron and steel production; • Caustic soda production; • Waste disposal; and, • Deforestation (Pacyna and Pacyna, 2002). Maxson (2004) estimates that the total amount of mercury consumed in East As ia in 2000 was 1100 tonnes, including A S G M , not including combustion of fossil fuels. This figure almost certainly underestimates consumption from A S G M , especially in Indonesia (Veiga and Baker, 2004) and China. Regardless, much of the 1100 tonnes of mercury was probably consumed in China, and thus the number may provide an indication of mercury 102 releases from other sources. Furthermore, the wide array of additional sources for mercury makes it difficult to determine the exact source of ambient mercury in China. 5.1.3 Comparing Mercury Releases in China The magnitude of mercury sources from mining in China for 2001 is compared in Table 5.4 and Figure 5.1. Table 5.4 Chinese Sources of Mercury Releases Due to Min ing (2001) Source M e r c u r y Releases (tonnes) A S G M 238 State-owned Coal Mines 176 A S M Coal Mines 127 A S M Mercury Mines 28 State-owned Gold Mines 20 State-owned Mercury Mines 19 Total 601 State-owned State-owned Mines 39.9% Figure 5.1 China's Mercury Releases Due to Min ing 103 A S G M is the largest source of mercury in China, along with the state-owned coal mining sector. The rejection of whole ore amalgamation through the adoption of alternative technology could dramatically reduce mercury releases. However, the effective introduction of alternative technology is not a trivial matter. 5.2 Alternative Technology Introducing alternative technologies to economically marginal activities in developing countries is hardly a new concept. Governments, N G O s , aid organizations and others have been doing this work for decades, especially in agriculture. Past efforts have often failed unless they met the following criteria: • Economically beneficial - The technology must be inexpensive to operate and it must generate obvious financial benefits; • Simple - The technology must be easy to use and would ideally utilize readily available resources; and, • Expedient - The economic mineral must be efficiently recovered. More directly, any new technology must be "fast, easy, and cheap" (Hinton et al, 2003a). The first efforts to address these issues were termed "Appropriate Technology." Appropriate Technology is a curious term, with an ideological background and a politicized history. This section defines and explains "appropriate technology" and explains why "alternative technology" is used as a substitute in this thesis. Appropriate Technology refers to a philosophy of life as much as to technology. It has its ideological basis in the Gandhian Movement and Edward Schumacher's writings, such as "Small is Beautiful" (Schumacher, 1973). The term suggests that there are different technological options for solving problems and that some technologies may be more appropriate in some situations than others. The counterpart to this is that there is technology that is not "appropriate". 104 In the 1970s and early 1980s Appropriate Technology was in vogue, and a number of books and articles were published on the topic, each with their own definition of what exactly Appropriate Technology (AT) referred to. The list o f A T criteria in the "Appropriate Technology Sourcebook" by Darrow and Pam (1981), as listed below, is representative. Appropriate Technology Criteria: 1) L o w in capital costs; 2) Use local materials whenever possible; 3) Create jobs, employing local skills and labour; 4) Are small enough in scale to be affordable by a small group of farmers; 5) Can be understood, controlled and maintained by villagers wherever possible, without a high level of Western-style education; 6) Can be produced out of a small metal-working shop, i f not in a village itself; 7) Suppose that people can and w i l l work together to collectively bring improvements to their communities, recognizing that in most of the world important decisions are made by groups rather than by individuals; 8) Involve decentralized renewable energy sources, such as wind power, solar energy, water power, methane gas, animal power, and pedal-power (such as in that highly efficient machine, the bicycle); 9) Make technology understandable to the people who are using it and thus suggest ideas that could be used in further innovations; 10) Are flexible so that they can continues to be used or adapted to fit changing circumstances; 1 l ) D o not involve patents, royalties, consultant fees, import duties, shipping charges, or financial wizards; practical plans can be obtained free or at low cost and no further payments is involved. A s can be seen from the seventh to the eleventh criteria, appropriate technology involved far more than simply using a technology that is most appropriate for the given situation. A more succinct definition, from the U.S . Congress's Office of Technology Assessment, 105 defines A T as "being small scale, energy efficient, environmentally sound, labour-intensive, and controlled by the local community" (Hazeltine and B u l l , 1999). A T ' s popularity arose from the belief that in order to alleviate poverty, groups concerned with development should focus on providing simple technology rather than expensive modern factories, dams, highways, and airports. A T could provide "goods, services, and jobs that w i l l not be provided any other way. N o company or organization w i l l be able to invest enough in high-technology factories in developing countries to provide sufficient jobs" (Hazeltine and B u l l , 1999). In addition, A T was attractive to people in wealthy nations trying to simplify and lessen the environmental impact of their lifestyle. This apparently sensible approach to development was not without enemies. Leaders from developing nations resented the idea of trying to modernize using technologies and equipment often perceived as outdated and primitive. D i d A T actually work toward national development, or did it in fact discourage industrialization? While the ideology of A T became widely accepted in the 1970s and 1980s, it was often used as a way to justify exporting cheap and low technology goods from developed nations. The term quickly became overused, creating a dichotomy between "real appropriate technology", and "old fashioned/obsolete technology." Developing nations became increasingly frustrated with the term and the poor quality equipment they were receiving under the guise of A T from various development programs. Often knowledge transfer was eschewed in favour of simply sending cheap equipment, without building the capacity to maintain the equipment or supply spare parts. In effect, A T came to be associated with obsolete equipment. B y the early 1990s the term was badly enough regarded that the U N I D O eliminated the term in favour of terms like "sustainable technologies" (Veiga, 2003). In this thesis the term 'alternative technology' is used as a substitute, with the aim of capturing the flexibility and utility of appropriate technology without its stigma. "Alternative technology" is hardly a new term either. Jequier and Blanc (1983) defined it as follows: "the term used to describe new types of equipment or new organisational forms which represent a viable alternative to the existing 'main-stream' technologies of 106 today." Other terms with similar meanings include "environmentally sound and appropriate technology ( E S A T ) " for appropriate technology using renewable resources, and "intermediate technology," for a technology that stood halfway between traditional and modern technology. For the purposes of this work, "alternative technology" is: 1) Environmentally and/or socially beneficial - it must provide concrete improvements over the existing technology; 2) Economic - it must match the capital availability of those who w i l l own and work with it and provide a reasonable return on investment; 3) Safe, with no major health risks; 4) Simple - the technology can be understood, and is attainable by the users; 5) Easy to maintain - users need to be able to afford spare parts or be able to repair the machine on their own; 6) Flexible enough to be able to handle changing circumstances; "Sustainability" in A S G M is contradictory, as an ore deposit is inherently finite, and thus not sustainable. However, communities that host A S G M activities are potentially sustainable, and small-scale miner organizations (companies, cooperatives, or whatever form they take) are potentially sustainable sources of employment, materials, and income, i f they act responsibly enough to gain social licence to operate and are flexible enough to adjust to changing circumstances (Veiga et al, 2001). A n y technical alternative must be thoroughly examined, pre-tested, and appropriately modified i f it is to be successfully transferred to miners and lead to more environmentally sound and socio-economically sustainable A S G M activity. 5.2.1 Sustainable Alternative ASGM Technology The test work above demonstrates that it is possible to get similar gold recoveries from a Knelson Concentrator / cyanidation circuit as from a mercury muller m i l l / cyanidation circuit, with far less mercury releases. A centrifugal concentrator was initially chosen for the alternative technology, as the miners were interested and the technology is relatively 107 efficient, and thus would provide a good indication of the optimal recoveries possible using gravity. The miners were familiar with traditional gravity technologies such as sluice boxes, but were not interested in using them, as they felt their mercury methods were more advanced. Regardless, Knelson Concentrators may not necessarily be the most 'sustainable' alternate solution for A S G M miners using muller mi l l technology. A simpler gravity solution, such as a sluice box, j ig , or the Cleangold™ magnetic sluice, in combination with cyanidation, may work almost as well . A s demonstrated in Figure 5.2, both mercury and gravity circuits fulfill a similar role: removing coarser, more liberated gold as cheaply and easily as possible before the more costly and time-consuming process of cyanidation. 100000 I — — 1 10000 •—s «] a | 1000 J , i l w> a a 100 1 .a io I -I h 1 1 1 1 1 ' 1 Jigs Spiral Strokes and Amalgamation Centrifugal Shaking tables Cyanidation concentrators riffles Concentrators Figure 5.2 Gold Particle Size Recovery Range (after Yannopoulos, 1991 and Priester et al, 1993) A centrifugal concentrator is likely to achieve higher recoveries than an alternative gravity method, as the centrifugal concentrators are more effective at recovering fine gold. However, the cyanidation process would likely pick up this fine gold in the next 108 step. Since the A S G M cyanidation processes are not especially efficient, recovering the fine gold in a gravity step would probably increase the overall recovery. For A S G M operations, as opposed to formal operations, this would probably be a small increase in revenues and might not have a fast enough payback to justify the added cost of a centrifugal concentrator. One of the prime problems with centrifugal concentrators as alternative technology is that they are relatively complex to manufacture well . The main manufactures are located in Canada and thus concentrators are subject to import duties, and finding spare parts can be expensive and time-consuming. However, centrifugal concentrators tend to be robust, and many of the wear parts, like drive belts, could easily be replaced locally. In comparison, muller mills are simple, inexpensive and manufactured locally. A s mentioned in section 2.3.1.1, a muller m i l l with a capacity of 0.5 - 1.0 tonnes per hour costs around US$1,200, versus at least $US 10,000, not including duty or taxes, for a centrifugal concentrator with a similar capacity. A t least one Chinese company produces centrifuges; however, they are not widely accepted, are poorly manufactured, and are not as efficient as the main Canadian suppliers. Inexpensive, simpler centrifugal concentrators are also manufactured in Brazi l and Zimbabwe. One of the main reasons that centrifugal concentrators might be appropriate is perception. Chinese A S G M operators/owners are well aware of the alternate gravity methods mentioned above, but distain them and consciously choose mercury methods instead. Sluices/jigs/spirals are seen as too simple and susceptible to theft, as the high-grade gold concentrate may not be secure. Centrifugal concentrators can be easily locked during operation and may be less susceptible to theft. In addition, centrifugal centrifuges represent a new technology, and thus have the potential to renew interest in gravity methods. In other words, even i f centrifugal concentrators did not offer significantly better results than simpler gravity technologies, A S G M miners might be wil l ing to accept switching from mercury to gravity methods due the perception that centrifugal concentrators were a more modern, efficient technology, as well as reducing the need for mercury use. 109 Regardless, even operations that installed centrifugal concentrators would likely continue using mercury to some extent. The gravity concentrate is unlikely to be a high enough grade to proceed to direct smelting. The most efficient method of further concentration is to mix the gravity concentrate with mercury through barrel amalgamation. While this still involves mercury, using amalgamation on only the concentrates reduces by several orders of magnitude the amount of mercury released. Veiga (2004b) reports that barrel amalgamation can achieve recoveries of about 90%. A key part of making the centrifugal concentrator / cyanidation method more appropriate for Chinese A S G M is to try to integrate the muller mills as the primary comminution stage of the circuit, without whole ore amalgamation, instead of introducing ball mills. The miners are comfortable with the muller mi l l technology, and its inexpensiveness, local manufacture, and efficiency make it more sustainable than getting new ball mills manufactured in other areas of China. A key point in favour of muller mills is that the entire process is visible; unlike ball or hammer mills, at no point is the ore hidden. This has proven a key concern in implementing new technologies in Zimbabwe (Veiga, 2004b). It also reduces the capital costs of installing a new gravity circuit. A Belgian development assistance project in Chile implemented an identical flow sheet, replacing a Chilean muller mi l l amalgamation centre with a Knelson / cyanidation circuit, while keeping the muller m i l l for comminution, but not amalgamation (see figure 5.3) (Dandois, 2000). Stamp mills may be another alternative; however, they are not manufactured in China, and are not an accepted technology. The Cleangold magnetic sluice, which uses polymeric magnetic sheets, with the magnetic poles aliened normal to the direction of the flow, inserted into an aluminium sluice box, offers another potential alternative technology. Magnetite or iron fillings, forms a corduroy-like bed on the sluice floor, which appears effective at recovering fine gold. Preliminary studies indicate recoveries similar to centrifugal concentrators for some ores (Gunson, 2003), with concentrate grades reaching 2854 g/t (Veiga and Gunson, 2004). 110 While the sluices are not manufactured locally, they are inexpensive, simple, and easy to maintain. Equipment to process 0.5-1.0 tonnes per hour would cost around $US500. Figure 5.3 Muller M i l l / Centrifugal Concentrator, Chile (Davidson, 2001) Alternatives to reduce mercury releases are plenty, including gravity, flotation and cyanidation methods. However, any attempt to replace mercury use must understand that amalgamation is often an extremely effective method and can be difficult to improve on. One key understanding that seems to have arisen is the need not to focus on eliminating mercury use, but on eliminating the amalgamation of whole ore and to promote the use of retorts. Attempts to ban amalgamation may reduce releases significantly, but would force other operators underground and eliminate the possibility of introducing more sustainable approaches. A better alternative is to target practices that lead to the most releases, such as whole ore amalgamation in muller mills in particular, and share practical alternatives with artisanal and small-scale miners. i l l 5.3 Technology Transfer Transferring technology, even i f it is alternative technology, is a hugely complex task. Comparatively little resources have gone into A S G M technology transfer compared to agricultural or industrial technology transfer, as A S G M activities have often been marginalized, and involve fewer people with far less political clout. However, the issues that arise are quite similar. The basic tenet of technology transfer is that technology does exist, or technology can be readily developed, that is superior to the existing technologies being used by the people in the field. The challenge is how best to inform the end users that this technology exists, or can be developed, and to get them to accept the new technology and successfully implement it in their operations. Technology transfer immediately runs into a significant hurdle with its basic tenet; existing technologies have been honed for years (sometimes centuries) to local physical, economic and social conditions, and it can be difficult to prove or convince the users that the new technology is in fact better. One example is the quimbalete, a traditional Andean amalgamation gold-processing technology that can metallurgically outperform newer gravity and cyanidation techniques. The quimbalete, while simple and cheap, also loses significant amounts of mercury, and is extremely labour intensive. A n y attempts to replace quimbaletes must convince the users that the expense of new equipment and the potential loss of employment and tradition are outweighed by the improvements in their local health and environment and by increased productivity ( C A S M , 2002). A l l technologies have different costs and benefits and it is rare that the benefits of a new technology are so overwhelmingly positive that it w i l l be universally accepted. N e w technologies often come with hidden or delayed costs, such as social disruption, overuse of land, or long-term health effects. A n y truly effective alternative technologies have to be developed in close cooperation with artisanal and small-scale miners and researchers. Working in isolation and bringing packaged solutions to the miners is unlikely to be successful, regardless of how clever the research. It is well demonstrated that bottom-up measures, or programmes developed by 112 the people participating in them, tend to be most effective and enduring (Hinton et al, 2003a). A S G M encompasses an enormous diversity of operations and communities. A sustainable alternative technology solution for one area might be entirely inappropriate for another, even nearby, area. For example, a solution for a primarily agricultural community with part-time miners may be significantly different than one for a migrant gold rush community. Careful consideration of the local socio-economic conditions, as well as the technical problem, is paramount to a successful technology transfer. Technical assistance for A S G M should not simply be a downscale of formal mining technology. A truly sustainable alternative technology, i f it is effectively transferred, should be self-maintaining, and should encourage further innovation. Other A S G M operations should be able to see its effectiveness and work toward their own sustainable solutions. A n example of the difficulties involved with technology transfer can be seen from the international experience with A S G M processing centres. 5.3.1 AS GM Processing Centres Several previous models and projects, described in brief below, have aimed to reduce A S G M mercury releases through the creation of clean (or cleaner) processing centres. The thrust of the concept was that individual, very small-scale processing operations are extremely difficult to effectively monitor, implement safety or environmental controls, or upgrade technologically. Building one central ore-processing centre in an area, in theory, allowed the creation of a larger-scale, cleaner operation, although still small in comparison to the formal mining industry. A central plant could be monitored to ensure health, safety, and environmental concerns are addressed, and could take advantage of more expensive and complicated technologies, leading to higher recoveries. Miners would bring a concentrate or high-grade ore to the centre and either pay a fee for mil l ing or simply sell their ore to the centre. The more organized nature of processing centres could make them more attractive to investment, regulators, the legal system, and society, in essence allowing A S G M to become part o f the formal economy (Hinton et al, 2003a). In addition, the centres can act as distribution points for technical and safety information for A S G M . 113 Several processing centres or models have been conceived, including: • U N I D O ' s Unit o f Gold Extraction and Controlled Amalgamation ( U N E C A ) model, based on the Carhuachi Center in Venezuela. The U N E C A Center concept was to be implemented in approximately six months, use special amalgamation plates and/or a sodium chloride electrolytic leaching process, and be built for approximately US$250,000. Miners would be charged approximately US$1 per kilo of concentrate processed. However, centres were expected to run at a loss (Veiga and Beinhoff, 1997) and no U N E C A centres have been built. • Intermediate Technology Development Group (ITDG), in cooperation with the National Miners' Association of Zimbabwe ( N M A Z ) , built the Shamva Min ing Center in Zimbabwe, in 1989. Acting as a central mi l l for approximately 200 miners, the centre process used a jaw crusher followed by a ball m i l l and a centrifugal concentrator. The concentrate was rolled in an amalgamation barrel and then retorted. The tails of both the centrifuge and amalgamation barrel were subjected to cyanidation and the pregnant leach solution was precipitated with zinc shavings. In addition, the Center provided drilling and blasting services, ore transportation, technical support, and information on legal issues, geology and metallurgy and the selection and purchase of equipment. Local miners were responsible for their own ore throughout the mil l ing process, supervised by a plant manager. Miners paid - U S $ 6 per hour (~US$4/t) to use the plant and take away the gold recovered (on average ~5g/t). Tailings were reprocessed with cyanide by the plant to generate revenue (Hinton et al, 2003b; Simpson, 2004). Friction between N M A Z and I T D G caused the subsequent breakdown of project (Simpson, 2004). • The Agence de Cooperation au Developpement par les Sciences Et les Techniques, a Belgian development agency, set up an assistance project in Chile that replaced a Chilean muller mi l l operation with a centrifugal concentrator / flotation circuit, using the muller mills as grinding mills in the late 1990s. The project cost US$182,557 (Dandois, 2000), and has since closed. 114 Several issues can commonly arise from central processing centres. While locally effective in reducing mercury releases, these models were never cheap enough to become generally accepted among small-scale gold miners. Many of these centres still used mercury in the process, albeit losing far less mercury than the operations they replaced. Since these centres were static, they could only impact a small number of miners. They also competed against local processing operations, and they were often not profitable. For processing centres to genuinely appeal to small-scale miners, they need to have incentives to change their existing methods. Without maintaining at least the existing profits, it would be hard to convince miners of the utility in switching. Perhaps most importantly, all o f the above examples required significant capital investment and expert inputs from outside organizations. That all three were not imitated and are all closed is the best indication of the limits of this model. The Global Mercury Project (GMP) , which has projects in Brazi l , Indonesia, Laos, Sudan, Tanzania, and Zimbabwe is exploring a different model. Instead of static centres, the G M P would build Transportable Demonstration Units, or T D U (Veiga, 2004b). These units would demonstrate a variety of technical options for gold concentration, amalgamation and retorting to miners; the miners themselves would be able to select what is affordable, appropriate and durable for their own operations. Being transportable, the T D U would be able to visit several areas and educate a far wider range of miners than was possible with the static processing centres. Local processing centres would feel less threatened, as the T D U would not be a competing processing operation. Problems of land tenure, services, mineral rights, etc., involved with processing centres would be avoided and the T D U would be able to easily change and adapt new pieces of equipment used for demonstration. While the T D U would be dependant on aid money, it avoids many of the short falls of central processing centres and may be a more successful model for technology transfer to A S G M . However, for a large country such as China, for an idea like the T D U to be widely implemented would require the close cooperation of government at all levels. 115 5.4 A S G M and Policy Despite the serious negative impacts of amalgamation and the presence of real technical alternatives, few governments have implemented effective A S G M policies with any measure of sustained success. Andrews-Speed et al (2003c) identifies three main barriers to the effective management and regulation of A S M : The first is a combination of political opposition, vested interests and ineffective government. A policy for small-scale mining can only be implemented effectively if the interests of most or all relevant parties are adequately addressed. The second impediment is the inadequate or inappropriate nature of national regulatory regimes for small-scale mines, which includes both the system of laws and regulations and the institutional structure. The third obstacle is a lack of financial resources to implement policies. This reflects the situation in China. Politically, many figures in higher levels of government simply see A S G M as a problem, and want to shut it down rather than create a viable framework within which it can exist. On the other hand, local governments, who often depend on A S G M tax revenue and employment, are cautious about cooperating with any efforts that may potentially disrupt their income. Concerning regulations, the effort to simply ban mercury led to mines either continuing amalgamation illegally or convincing local governments to ignore the mercury ban and issue local permits. The policy and regulatory regime simply do not encourage honesty or the responsible use of mercury. The lack of financial resources to successfully implement policies is a combination of the sheer scale of China and the lack of political w i l l to use tax dollars to address the issue. A n y effort to address the problems of amalgamation in A S G M must recognize these barriers and work toward overcoming them. 116 6 Limitations and Bias The results of this thesis were restricted by several limitations. These include: • Language and Culture: The author, while generally functional in Chinese, is hardly fluent. The interviews from the first trip to "Go ld Mountain" and the trips to Guangxi and Inner Mongolia were somewhat limited due to the lack of an interpreter. However, the miners seemed more at ease without the presence of interpreters, as an interpreter indicates wealth and/or government connections. This ease can be demonstrated by the fact that the closest and most fruitful relationship developed over the course of this research was with the "Gold Mountain" miners, where the initial relationship and trust was built without the presence of an interpreter. Interviews, by necessity, were often informal and incomplete. • Access to Government Information: The Chinese government probably has various reports on A S G M and mercury that would not be available to the author. • Government: This research was not undertaken through official government or academic channels, primarily since such approval, i f available, would have been costly and time consuming. Since amalgamation is largely illegal, miners were often cautious about discussing their practices, and site visit locations and interviewees often were obscured. • Medical Training: A lack of medical expertise available restricted the ability to effectively judge the health impact of mercury use by A S G M miners. • Equipment: The financial and geographical logistics of the site studies were such that it was impossible to take equipment for onsite mercury or gold analyses, or to keep biological samples for later testing. • H g E X : H g E X was validated for the Amazon and for northern Canada. It may not be relevant in the sites examined in China. • Test Samples: The samples sent for test work were clearly from different ore bodies, which greatly restricted the ability to compare between results. In addition, the "Go ld Mountain" miners collected the samples, making it difficult to verify that the samples were representative. 117 A simple method of categorizing systematic research errors is to classify them in two general classes: selection and observation bias (Hennekens and Buring, 1987). Due in part to the above limitations the following bias are noted: • Selection Bias o Site Vis i t Regions: Site visits were chosen to represent several different regions of China, however, a limitation was that all o f the sites had to be reasonably accessible by public transportation. A l l sites visited were less than one day's travel from an urban center, a necessity as many rural village hotels are barred from accepting foreigners. This excluded A S G M sites in the most remote areas or in government restricted areas. It may be reasonable to assume that the more remote areas practice less safe processing methods than in areas relatively accessible to urban commercial and educational centers. o Site Visits and Interviews: Mines and processing centers could only be studied i f the miners were wil l ing to cooperate with the research. Outright illegal miners had little incentive to agree to site visits or interviews, and these miners' operations may have been less safe than operations wil l ing to cooperate with the author. o Test Work Samples: The test work indicates that the samples my not have been representative. The high gold grade and slow leaching nature of the feed sample, and the high gold grade and mercury content o f the muller mi l l tailing sample, indicate that they may have come from ores the miners were having difficulty processing. These samples may have exaggerated both the gold grades and the difficulty of processing of ores in "Gold Mountain". • Observation Bias o Interviewer Bias (A) , Observing for Indications of Mercury Poisoning: While the thesis indicates a low number of mercury poisonings, this could be due more to a lack of medical knowledge than an actual absence of illness. Systematic testing by skilled personnel could reveal far more 118 extensive mercury related health impacts. Only the most extreme cases could be observed. o Interviewer Bias (B), Observing for Mercury Bioaccumulation: The lack of mercury analysis equipment meant it was impossible to demonstrate bioaccumulation in local aquaculture. Without verification, the bias was that bioaccumulation was not a serious factor. o Interviewer Bias (C), Research Cooperation with Miners: Working with A S G M miners directly, instead of working through official channels, tended to bias the author in favour of the miners and local communities. M u c h of the research depended entirely on the good w i l l of local government officials and miners. Many interview questions dealt with sensitive issues for the miners and the answers were not necessarily honest, but were still valuable. o Recall Bias, Miner Estimates: The lack of assay equipment (or even scales) meant that miners often needed to be taken at their word for their estimates of grades and recoveries. Based on some discrepancies outlined in Chapters Three and Four, it is assumed that miners exaggerate their recovery rates and mislead with regard to their gold grades and mercury use. o Misclassification Bias, Mercury Source: A s mentioned in Chapter Five, China hosts many industries that emit or release mercury into the environment. A n y i l l effects due to mercury may not necessarily be due to A S G M . These biases would have been difficult to overcome during the course of this research, primarily due to the lack of initial study on the topic and due to limited funds. However, additional research could easily overcome these limitations. 119 7 Areas for Additional Research Several gaps in knowledge or practice were identified, including: • Great uncertainty about the actual gold production of, income generated by, and number of people employed in A S G M in China. • A lack of understanding regarding many details about the legal status and tax structure of A S G M in China. • A n incomplete picture of the role of women and children in A S G M in China. • Uncertainty regarding the risk of mercury vapour poisoning due to the smelting of zinc strips following amalgamation and cyanidation. • Uncertainty about the effect of cyanidation on amalgamation tailings. This preliminary work w i l l be followed up by further research at the University of British Columbia under the direction of Dr. Marcello Veiga. • Uncertainty about the amount of mercury that reports to the zinc strips versus the final tailings following the cyanidation of amalgamation tailings. • A need for case studies to verify the feasibility of replacing whole ore amalgamation of hard rock ores with gravity concentration. • A poor understanding of the effectiveness of muller mills for comminution versus more standard methods of hammer, ball, or rod mills. • A gap in the understanding of mercury releases from industrial sources in China. Toward the end of addressing many of these gaps, the following recommendations are suggested: • A thorough socio-economic baseline study of at least one A S G M community should be undertaken, using a method such as the Department for International Development's (DFID) Sustainable Livelihoods approach (DFID, 2001). • A proper environmental impact study of the behaviour of mercury in the environment due to A S G M should be undertaken in at least one area. • A comprehensive epidemiological study regarding exposure to mercury and the resulting health impacts on the A S G M population should be completed in at least one community. 120 • Following the three studies listed above, the data obtained should be used to validate H g E X for China. • A n organization should be built to coordinate these and more studies and to share the information learned with miners, researchers, policy makers, and the international community. To this end, the author has been instrumental in assisting in the creation of an organization called the Communities and Small-Scale Min ing Regional Network in China (CASM-China) . C A S M - C h i n a aims to bring together a wide range of people involved in artisanal and small-scale mining in China, within the context of poverty alleviation and sustainable communities (See Appendix I for further details). In addition, one of the initial aims of this research was to discover i f Chinese methods of dealing with these environmental and health problems could be applicable to other parts of the world, or i f techniques developed outside of the P R C could be used by the Chinese. The "Go ld Mountain" retort has already been successfully introduced into Colombia (Sandoval, 2003), Venezuela (Veiga and Gunson, 2004), and Laos (Veiga, 2004a), and has been shared with the international community through two papers (Hinton et al, 2003; Gunson and Veiga, 2004). Centrifugal concentrators were introduced to the A S G M community of "Go ld Mountain" in 2002. China also provides numerous examples of integrating A S M with a diversified rural economy; "Gold Mountain" meshes specialty agricultural crops, grain crops, tree farming, small-scale manufacturing, gold mining, iron mining, sand and gravel operations, and dimension stone quarries. C A S M - C h i n a can help to further facilitate the exchange of best practices around the world. 121 8 Conclusion The interviews, field studies, and test work undertaken for this thesis lead to the following main results and conclusions: • A S G M in China employs 463,000 A S G M miners producing around 61 tonnes of gold, in regions all over China, with almost 3.5 mil l ion dependants. A S G M released 237 to 652 tonnes of mercury in 2001 and is easily one of the top sources of mercury releases in China. A t the low range of the estimate, A S G M contributes about 40% of the mercury releases due to mining, and whole ore amalgamation in muller mills may be responsible for up to 70% of these mercury releases. A t every site and region visited, A S G M operations used mercury, often followed by cyanidation. The H g E X expert system tool was used to demonstrate the potential impact of mercury contamination from A S G M and to indicate that China has a moderate acceptance of amalgamation practices. • The "Go ld Mountain" case study verified that A S G M in China could release significant amounts of mercury, with "Go ld Mountain" alone releasing at least one and a half tonnes of mercury per annum. Mercury use represented only 2.8 % of the overall gold production cost, and the ratio of mercury lost per unit of gold recovered for muller mills is from 14:1 to 20:1. The "Gold Mountain" retort was described, representing a simple and inexpensive technology that has the potential to reduce mercury exposure to miners around the world. • The "Go ld Mountain test program further verified the high levels of mercury being released from "Gold Mountain's" processing centres, with the muller m i l l tailings mercury concentration reaching 2224 ppm, and the cyanidation tailings reaching 285 ppm mercury, from a background mercury concentration, as indicated by the feed ore, of only 0.78 ppm. Test work completed on the feed sample, grading about 43 g/t gold, indicated that around 40% of the gold was recoverable by gravity methods and that the leaching rate of the ore is extremely slow, perhaps due to the high levels of copper (0.44%) and tellurium (13.71 ppm). The test work demonstrated that gravity methods could be introduced to replace whole ore amalgamation, and that increased comminution would help the miners achieve higher recoveries, although amalgamation of gravity concentrates may still be necessary to produce a saleable product for the miners. The muller mi l l 122 tailings sample, with a gold grade of 38.1 g/t, indicated that amalgamation had not effectively recovered gold from the initial ore, especially as 43% of the gold in the tailings was recovered, 30% in the first pass without additional grinding. • The health impacts of mercury releases are of key concern. Mercury vapour pollution, despite the use of retorts, is of primary concern, as demonstrated by a miller who stated that he had been diagnosed with health problems due to mercury exposure. The lower mercury concentration in the cyanidation tailings indicated that much of the mercury from the muller mi l l is removed by cyanidation, to be volatilized during the zinc strip smelting process. The potential for mercury bioaccumulation exists and should be further investigated; the mercury solubility tests indicate that cyanidation, following whole ore amalgamation, may increase the solubility of the mercury in the tailings. • A n y attempt to address mercury pollution from A S G M must come to terms with the complex socio-economic and political framework in which A S G M in China operates. A S G M operations provide desperately needed employment and tax revenue in remote areas with few alternative economic activities but essentially lacks any formal or informal support from domestic or international organizations. • Effective policies must be implemented in China to encourage multi-stakeholder cooperation, and laws and regulations with which A S G M is reasonably able to comply must also be implemented. Building an effective policy toward minimizing mercury releases from A S G M would require significant political support at high levels, effective cooperation with local governments, and genuine dialogue between artisanal and small-scale miners, local communities and governments, and experts. A n effective policy would require a multi-pronged approach to the issue, involving regulatory initiatives, organization, education and training, and financial support. Ultimately this thesis found A S G M and the use of amalgamation to recover gold widespread throughout China and that it was the cause of significant environmental and health problems. Existing alternative technology could dramatically reduce the impact of A S G M and significantly reduce mercury releases. Furthermore, the practices of whole ore amalgamation and of following amalgamation with cyanidation should be eliminated. 123 Bibliography Andrews-Speed, P., A . Zamora, C. Rogers, L . Shen, S. Cao, and M . Yang. 2002. A framework for policy formulation for small-scale mines: the case of coal in China. Natural Resources Forum v. 26: 43-52. Andrews-Speed, P., M . Yang, L . Shen, and X . Cao. , 2003a. Thirty years of government policy toward township and village coal mines in China. University of Dundee, Scotland. Andrews-Speed, P., G . M a , X . Shi, and B . Shao. 2003b. The Impact of, and Responses to, the Closure of Small-Scale Coal Mines in China: A Preliminary Account. In The Socioeconomic Impacts of Artisanal and Small-Scale Mining in Developing Countries. G . Hilson, ed., A . A . Balkema Publishers, Rotterdam, 2003b. Andrews-Speed, P., M . Yang, L . Shen, and S. Cao. 2003c. The regulation of China's township and village coal mines: a study of complexity and ineffectiveness. Journal of Cleaner Production 11: 185-196. Atiyah, J., D . Leffman, and S. Lewis. 1997. China: The Rough Guide. Rough Guides Ltd., London, 1008p. Baker, R. 2002. Fish Mercury Database Summary 2001 - British Columbia. B C Hydro Report, Vancouver, British Columbia. Becker, J. 2004. China's growing pains. National Geographic v. 205 (3). 68-95. Bil l i ton. 1999. Minerals Companion. Min ing Journal Books Ltd, Kent. British Columbia - Ministry of Energy, Mines and Petroleum Resources ( B C - M E M P R ) . 1992. Health, Safety and Reclamation Code for Mines in British Columbia. Resource Management Branch, Victoria. British Broadcasting Corporation (BBC) . 2001. Chinese troops battle cyanide spill November 5. http://news.bbc.co.uk, Accessed February 11, 2004. Carmouze, J., M . Lucotte, A . Boudou. 2001. Mercury in the Amazon: Importance of Human and Environment, Health Hazards. Synthesis and Recommendations. IRD Editions. Institut de Recherche Pour le Devloppement, Paris. Central Intelligence Agency (CIA). 2004. China. The World Factbook. http://www.cia.gov/cia/publications/factbook/geos/ch.html. Accessed February 9, 2004. China Labour Bulletin. 2003. Child Labour in China: Causes and solutions, www.china-labour.org.hk, Accessed February 11, 2004. China State Statistical Bureau. 2000. China Statistical Yearbook. Beijing. 124 Communities and Small-Scale Min ing ( C A S M ) . 2002. Personal Communications. September 27. Small-scale miner, Santa Filomena, Peru. Dahlberg H . 1997. The Transformation from Artisanal to Small Scale Gold Mining: A n Outline for Sustainable Development. In Introducing New Technologies for Abatement of Global Mercury Pollution Derived From Artisanal Gold Mining, U N I D O , Vienna. Dai , Q., X . Feng, G . Qui , and H . Jiang. 2003. Mercury contaminations from gold mining using amalgamation technique in Xiaoqinling Region, Shanxi Province, P R China. J. Phys. IVFrance, v 107: 345-348. Dandois, P. 2000. Amelioration Des conditions De Travail et de Vie Des Mineurs Artisanaux de la Troisieme Region DAtacama (Chili) (in French). I D I C T E C A . C . D . S . T . , Universidad De Atacma, Chile, in French, 57p. Davidson, J. 2001. Photo of Chilean Muller M i l l with Knelson Concentrator, Chile. Davidson, J. 2003. Personal Communication, July. C A S M Secretariat. C A S M , Washington, D C . Darrow, K . and R. Pam. 1981. Appropriate Technology Sourcebook, Volume One, Volunteers in Asia , Stanford, 318p. Department of Foreign Affairs and International Trade (DFAIT) . 2001. The Mining Industry in China. FaxLink no. 43115, Ottawa, 76. Department for International Development (DFID). 2001. Sustainability Livelihoods Guidance Sheets, www.livelihoods.org, Accessed March 31,2004. Feng, X . , and Y . Hong. 1999. Modes of occurrence of mercury in coals from Guizhou, People's Republic of China. Fuel v. 78: 1181-1188. Flynn, C. 1995. Cyanide Chemistry-Precious Metals Processing and Waste Treatment. United States Bureau of Mines Information Circular, IC9429. Forstner U . and G . Wittman. 1979. Metal Pollution in the Aquatic Environment. Springer-Verlag, Berlin, 486p. Fong, T. 2001. Mainland's 'biggest lie' unravels to expose mine culprits. South China Morning Post, www.scmp.com, Accessed August 13, 2001. Golas, P. 1999. Joseph Needham Science and Civilisation in China. Vol. 5, Part XIII. Cambridge University Press, Cambridge, 538p. 125 Golas, P. 2003. Personal Communication. February 3, Professor, University of Denver, Colorado. Grandjean, P., R. White, A . Nielsen, D . Cleary, and E . Santos. 1999. Methylmercury Neurotoxicity in Amazonian Children Downstream from Gold Mining . Environmental Health Perspectives, V.107:587-91. Gunson A . , M . Veiga, and J. Meech. 2001. H g E X : Assessing the Risk of Mercury Pollution from Artisanal Gold Mines. In Proceedings from the 29th International Symposium on Computer Applications in the Minerals Industries (APCOM 2001) Beijing, 745-750. Gunson A . , and J. Yue. 2001. Artisanal Mining in The People's Republic of China. 74, Mining , Minerals, and Sustainable Development, London, 19. Gunson, A . and M . Veiga. 2004. Mercury and Artisanal Min ing in China. Accepted for publication, Environmental Practice, v. 6 (2). Gunson, A . 2003. Cleangold Test Program, Unpublished, The University of British Columbia, Vancouver, 18p. Hacon, S. 1990. Mercury Contamination in Brazil. Technical Report, F I N E P -Financiadora de Estudos e Projetos, Rio de Janeiro, Brazi l , 85p. Hakanson, L . , A. Nilsson, and T Andersson. 1988. Mercury in Fish in Swedish Lakes. Environmental Pollution, v. 49: 145-162. Harada, M . 1978. Methyl Mercury Poisoning due to Environmental Contamination (Minamata Disease). In Toxicity of Heavy Metals in the Environment, Part 1, Oehme F., ed., Marcel Dekker Inc, N e w York, 970p. Harada, M . , J. Nakanishi, E . Yasoda, M . Pinheiro, T. Oikawa, G . Guimaraes, B . Cardoso, T. K izak i , and H . Ohno. 2001. Mercury Pollution in the Tapajos River Basin, Amazon: Mercury Level of Head Hair and Health Effects. Environment International, v.27: 285-290. Hart, C , R. Goldfarb, Y . Qui, L . Snee, L . Mi l le r , M . Mil le r . 2002. Gold deposits of the northern margin of the North China Craton: multiple late Paleozoic-Mesozoic mineralizing events. Mineralium Deposita v. 37 (3-4): 326-351. Hazeltine, B . and C. B u l l . 1999. Appropriate Technology: Tools, Choices, and Implications, Academic Press, San Diego. He, Y . , Y . L i n , and W . L i . 2000. Mercury consumption and contamination of environment in gold mining in China. Journal of environmental Sciences, v. 12 (May Supplement): 39-44. 126 Hennekens, C . and J. Buring. 1987. Epidemiology in Medicine. Little, Brown and Company. Boston. 383p. Hentschel T., F . Hruschka, M . Priester. 2002. Global Report on Artisanal & Small-Scale Mining. Mining , Minerals, and Sustainable Development ( M M S D ) Report N o . 70, IIED and W B C S D , London, 67p. Hinton, J. 2002. Earthworms as a Bioindicator of Mercury Pollution in an Artisanal Gold Mining Community, Cachoeira do Piria, Brazil. Masters Thesis, The University of British Columbia, Vancouver. Hinton, J.J., M . M . Veiga, and A . T . Veiga. 2003a. Clean Artisanal Mining , a Utopian Approach? Journal of Cleaner Production, 11: 99-115. Hinton J. , M . Veiga, and C. Beinhoff. 2003b. Women and Artisanal Min ing : Gender Roles and the Road Ahead. In: The Socioeconomic Impacts of Artisanal and Small-Scale Mining in Developing Countries, G . Hilson, A . A . Balkema Publishers, Rotterdam, 739p. Horvat, M . 2002. Mercury as a global pollutant. Analytical and Bioanalytical Chemistry v. 374: 981-982. Horvat, M . , N . Nolde, V . Fajon, V . Jereb, M . Logar, R. Jacimovic, L . Qu, D . Drobne, J. Faganeli. 2000. Mercury and other trace elements in Guizhou Province, China: Preliminary assessment of mercury pollution in the areas of Wanshan and Quingzhen [sic] Results from field sampling in September, 1999. IJS-DP-8294. Humphreys, D . 2002. China Mining and Metals: The Waking Giant. Paper presented at the Canadian Institute of Min ing and Metallurgy, Annual General Meeting, Vancouver. Hylander L . and M . M e i l i . 2003. 500 years of mercury production: global annual inventory by region until 2000 and associated emissions. The Science of the Total Environment, v. 304: 13-27. International Labour Organization (ILO). 1998. China: Promoting safety and health in township and village enterprises. ILO East As ia Multidisciplinary Advisory Team. Bangkok, 55p. Jennings, N . 1999. Social and Labour Issues in Small-scale Mining. Report for discussion at the Tripartite Meeting on Social and Labour Issues in Small-scale Mines 17-22 May, 1999. International Labour Office. Geneva. Jequier, N . and G . Blanc. 1983. The World Of Appropriate Technology: A Quantitative Analysis. O E C D , Paris, 21 Op. 127 Jernelov A . , L . Landner, and T. Larsson. 1975. Swedish Perspectives on Mercury Pollution. Journal of the Water Pollution Control Federation v. 47 (4): 810-822. Keating, M . , K . Mahaffey, R. Schoeny, G . Rice, O. Bullock, R. Ambrose, J. Swartout, J. Nichols. 1997. Mercury Study: Report to Congress. United States Environmental Protection Agency, A i r , EPA-452/R-97-003. Lacerda L . 1997. Global Mercury Emissions from Gold and Silver Mining . Water, Air and Soil Pollution, v. 97: 209-221. L i n Y . H . , M . X . Guo, and W . M . Gan. 1997. Mercury Pollution From Small Gold Mines in China. Water, Air and Soil Pollution 97: 233-239. L u J. Y . , W . Schroeder, L . Barrie, A . Steffen, H . Welch, K . Martin, L . Lockhart, R. Hunt, G . Boi la , and A . Richter. 2001. Magnification of atmospheric mercury deposition to polar regions in springtime: the link to tropospheric ozone depletion chemistry. Geophysical Research Letters, v. 28 (n. 17): 3219-3222. Ma lm, O. 1991. Human and Environmental Mercury Contamination in Gold Garimpo Region of Madeira River, Amazon. (PhD Dissertation) Federal University of Rio de Janeiro - Biophysics Institute Carlos Chagas Filho, Rio de Janeiro, in Portuguese. M a l m O., W . Pfeiffer, C. Souza, and R. Reuther. 1990. Mercury Pollution Due to Gold Min ing in the Madeira River Basin, Brazi l . Ambio, v. 19 (n. 1): 11-15. Maxson, P. 2004. Mercury flows in Europe and the world: The impact of decommissioned chlor-alkali plants. European Commission, Directorate General for Environment, Brussels, 88p. Meech, J., M . Veiga, and D . Tromans. 1998. Reactivity of Mercury from Gold Mining Activities in Darkwater Ecosystems. Ambio , v.27 (n.2): 92-98. Mergler, D . 2002. Review of Neurobehavioral Deficits and River Fish Consumption from the Tapajos (Brazil) and St. Lawrence (Canada). Environmental Toxicology and Pharmacology, v. 12: 93-99. Merriam-Webster. 2004. muller. http://www.m-w.com, Accessed March 21. Ministry of Agriculture. 2000. China Township and Village Enterprises Yearbook. Beijing. Mitra, S. 1986. Mercury in the Ecosystem. Trans Tech Publications, The Netherlands. Mortensen, J. 2003. Personal Correspondence. Professor. Department of Earth and Ocean Sciences, The University of British Columbia, Vancouver. 128 National Institute for Minamata Disease (NIMD) . 2001. In the Hope of avoiding Repetition of a Tragedy of Minamata Disease. Report o f the Social Scientific Study-Group on Minamata Disease, N I M D , Minamata, Japan, 140p. Natural Resources Canada ( N R C A N ) . 2002. Canadian Minerals Yearbook 2001. Catalogue no. M38-5/50E-IN. http://vvww.nrcan.gc.ca/mms/cmy/2001CMY_e.html, Accessed February 10, 2004. Noetstaller, R. 1995. Historical Perspective and Key Issues of Artisanal Mining. Paper presented at the International Round Table on Artisanal Min ing . The World Bank. Washington, D C . Pacyna E . and J. Pacyna. 2002. Global Emission of Mercury from Anthropogenic Sources in 1995. Water, Air, and Soil Pollution, v. 137: 149-165. Paterson, J. 2003. Personal communication. 8 A p r i l , President, Southwestern Resources, C I M Annual Meeting, Montreal. People's Daily. 2000. Toxic Leak Contained After Spill, October 13. http://english.peopledaily.com.cn, Accessed February 11, 2004. People's Daily. 2002a. Gold Mine Blast in North China Leaves 46 Dead, June. http://english.peopledaily.com.cn, Accessed February 11, 2004. People's Daily. 2002b. Probe of Fatal Gold Mine Blast Continues, No More Bodies Found, July, http://english.peopledaily.com.cn, Accessed February 11, 2004. People's Daily. 2002c. China Approves New Plan for Illiteracy Elimination Work, Sept 9. http://english.peopledaily.com.cn, Accessed February 9, 2004. Poulsen, K , and J. Mortensen. 1993. Observations on the gold deposits of Eastern Hebei Province, China. In Current Research, Part D, Geological Survey of Canada, Paper 93-1D: 183-190. Pomfret, J. 2001. Seeds of Revolt in Rural China: Farmers' Heroes give a voice to besieged taxpayers, May 8. The Washington Post. http://www.hartford-hwp.com/archives/55/334.html, Accessed March 19, 2004. Priester M . , T. Hentschel, and B . Benthin. 1993. Tools for Mining: Techniques and Processes for Small Scale Mining. Gate / G T Z , Berlin, 537p. Pryor, E . 1965. Mineral Processing. Elsevier, N e w York, 488p. Q i , X . , Y . L i n , J. Chen, and Y . Ye . 2000. A n Evaluation of Mercury Emissions From the Chlor-Alkal i Industry in China. Journal of environmental Sciences, v. 12 (May Supplement): 24-30. 129 Reuters. 1995. Flood of prospectors prompts China gold mine close, Oct 14. http://www.tibet.ca/wtnarchive/1995/10/14_2.html, Accessed February 10, 2004 Sandoval, M.C. 2003. Personal Communication, March. Graduate Student. The University of British Columbia, Vancouver. Schumacher, E. 1973. Small is beautiful: a study of economics as if people mattered Blond and Briggs, London, 28 8p. Simpson, J. 2004. The Shamva Mining Centre, Zimbabwe, http://www.livelihoodtechnology.org/home.asp?id=csShamval, Accessed March 1. Southwestern Resources. 2004. Boka Gold Project, www.swgold.com. Accessed February 11,2004. State Bureau of Coal Industry. 2000. China Statistical Yearbook of Coal Industry - 1999 (in Chinese). Coal Industry Publishing House, Beijing. State Council. 1988. Regulations on Labour Protection for Female Workers and Employees. Beijing. State Council. 1991. Regulations on Prohibition of Child Labour. Beijing. Stokes, P. and C. Wren. 1987. Bioaccumulation of Mercury by Aquatic Biota in Hydroelectric Reservoirs. Chapter 16, SCOPE 31. John Wiley, New York, 255-277. Stopford, W. 1979. Industrial Exposure to Mercury. In The Biogeochemistry of Mercury in the Environment. J. Nriagu, ed., Elsevier/North-Holland Biomedical Press, Amsterdam, 367-397. Suzuki, T. 1979. Dose-effect and Dose-response Relationships of Mercury and Its Derivatives. In The Biogeochemistry of Mercury in the Environment, J. Nriagu, ed., Elsevier/North- Holland Biomedical Press, Amsterdam, 299-431. The Economist. 2002. No Rural Idyll. In: A Survey of China. 363 (8277): 8-9. Trakhtenberg, I. M . 1974. Chronic Effects of Mercury on Organisms. Translation by The John E. Fogarty International Center For Advanced Study in the Health Sciences, US Department of Health, Education, and Welfare, Washington, D H E W Pub. No. (NIH) 74-473, 333p. Tse, P.K. 1995. The Mineral Industry of China (1994). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 18p. 130 Tse, P.K. 1996. The Mineral Industry of China (1995). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 17p. Tse, P.K. 1997. The Mineral Industry of China (1996). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 19p. Tse, P.K. 1998. The Mineral Industry of China (1997). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 30p. Tse, P.K. 2000. The Mineral Industry of China (1999). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 23p. Tse, P.K. 2001. The Mineral Industry of China (2000). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 20p. Tse, P.K. 2002. The Mineral Industry of China (2001). In: Minerals Yearbook (Volume III - Area Reports: International - Mineral Industries of Asia and the Pacific). USGS, 22p. United States Embassy in Beijing. 1996. Gold Mining In China: Taming The Wild West. http://vvfww.usembassy-china.org.cn/sandt/goldw.htm, Accessed May 26, 2002. United States Embassy in Beijing. 2001. Birth Of An NGO? Development Of Grassroots Organizations In The Land Of Big Brother, http://www.usembassy-china.org.cn, Accessed July, 27 2001. United States Environmental Protection Agency (USEPA). 2002. Water Quality Criteria. Office of Water, Office of Science and Technology (4304T). EPA-822-R-02-047. Veiga, M. 1997. Introducing New Technologies for Abatement of Global Mercury Pollution in Latin America. UNIDO/UBC/CETEM, Rio de Janeiro, 94p. Veiga, M. 2000. Personal communication, Associate Professor, The University of British Columbia, Vancouver. Veiga, M. 2003. Personal Communication. February 27. Small-Scale Mining Expert. UNIDO, Vienna. Veiga, M. 2004a. Personal Communication. January 7. Small-Scale Mining Expert. UNIDO, Vienna. 131 Veiga, M. 2004b. Equipment Specification for the Demonstration Units in Zimbabwe Project EG/GLO/01/G34, UNIDO, Vienna. Veiga, M. 2004c. Personal Communication. March 5. Small-Scale Mining Expert. UNIDO, Vienna. Veiga M. and R. Baker. 2004. Protocols for Environmental & Health Assessment of Mercury Released by Artisanal and Small-Scale Gold Miners (ASM). Project EG/GLO/01/G34 (Global Mercury Project), UNIDO, Vienna, 156p. Veiga, M. and C. Beinhoff. 1997. UNECA Centers, a Way to Reduce Mercury Emissions from Artisanal Gold Mining and Provide Badly Needed Training. Industry and Environment, United Nations Environment Programme, v. 20 (4): 49-51. Veiga, M. and A. Gunson. 2004. Technical Aspects of Mercury Pollution Caused by Artisanal Gold Miners in Block B, El Callao, Bolivar State, Venezuela. Project XP/VEN/C04, UNIDO, Vienna. Veiga M. and J. Meech. 1992. Expert System for Risk Assessment of Mercury Discharge from Gold Mining Operations. Proc. Int. Symp. on Al in Material Process. 31st Annual Conf. of Metallurgists. ofCIM, 107-118. Veiga, M., and J. Meech. 1995. HgEX - A Heuristic System on Mercury Pollution in the Amazon. Water, Air and Soil Pollution, v. 80: 123-132. Veiga, M., and J. Meech. 1996. HgEx: An Expert System for Risk Assessment of Mercury, University of British Columbia. Software. Veiga, M., M. Scoble, and M. McAllister. 2001. Mining with communities Natural Resource Forum v. 25: 191-202. Wang Q., W. Shen, and Z. Ma. 2000. Estimation of Mercury Emission from Coal Combustion in China. Environmental Science & Technology v. 34 (13): 2711-2713. Wilhelm, M., F. Muller, and H. Idel. 1996. Biological Monitoring of Mercury Vapour Exposure by Scalp Hair Analysis in Comparison to Blood and Urine. Toxicology Letters v.88 (1-3): 221-226. Williamson, D. 1986. Mercury in Water and Sediments in the Churchill and Nelson Rivers, Manitoba, Canada. In Technical Appendices to the Summary Report of the Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion, v. 1. World Health Organization (WHO). 1991. Guidelines for Methylmercury in Fish (CAC GL 7 - 1991). http://www.who.int/fsf/Codex/methylmercury.htm. Accessed November 17, 2003. 132 World Tibet Network News. 1995. Modern-day Ulegal gold rush threatens Tibet environment. http://www.tibet.ca/wtnarchive/1995/5/9_l.html, Accessed February 11, 2004. Yannopoulos, J. 1991. The Extractive Metallurgy of Gold. VanNorstrand Reinhold, New York, 28 lp. Zhang, Jim, 2002. Personal communication. 7 October. Knelson Concentrators Representative in China. Beijing. , Zhong, Z. 1999. Small-scale Mining Activities in China - Contributions, Problems and Policy Options. Unpublished Presentation. The Small-and Medium-Scale Mining Workshop of the 3 r d Environmental Cooperation Workshop on Sustainable Development of Mining Activities, Cairns, Austrlia, 20p. Zhou, T., and R. Goldfard. 2002. Tectonics and metallogeny of gold deposits in China. Mineralium Deposita v. 37: 247-248. Zhou, T., R. Goldfard, G. Phillips. 2002. Tectonics and distribution of gold deposits in China - an overview. Mineralium Deposita v. 37: 249-282 Zhou, W. 1997. The formation of sinkholes in karst mining areas in China and some methods of prevention. Environmental Geology v. 31 (1/2): 50-58. 133 Appendix I C A S M - C h i n a Inaugural Meeting Report Inaugural Meeting Report To the Communities and Small-Scale Mining ( C A S M ) Secretariat regarding the Communities and Small-Scale Mining Regional Network in China (CASM-China ) Inaugural Meeting Shen Le i A J Gunson Phil ip Andrews-Speed January 2004 135 Forward The Communities and Small-Scale Min ing Regional Network in China (CASM-China) held its inaugural meeting in Beijing from January 5-7, 2004. C A S M - C h i n a aimed to bring together a wide range of people involved in artisanal and small-scale mining ( A S M ) in China, within the context of poverty alleviation and sustainable communities. Opinion within the country is sharply divided between those who see A S M activity as harmful and unnecessary and those who see it as essential for local development. This meeting broke new ground by including discussions that encompassed a full range of A S M issues. During the course of the meeting 43 participants from 24 agencies met together and discussed how an effective network w i l l be organized and what goals can be achieved in China's complex and often sensitive A S M sector. These agencies included: • Seven departments in the Ministry of Land and Mineral Resources (China Mining Association, Department of Mineral Exploitation, Senior Consultancy Center, Information Center, Center of Land and Mineral Legal Affairs, Academy of China Geological Environmental Monitoring, and the Academy of Land Resources Economics); • The State Safety Production Supervisory Bureau; • Two institutes in the Chinese Academy of Sciences (the Institute of Geographic Sciences and Natural Resources Research and the Center for Eco-Environment), • One institute in the Chinese Academy of Social Sciences (the Institute of Quantitative Economy and Technical Economy); • Four universities (China Renmin University, Shanxi University of Finance and Economy, Chongqing University, Liaoning University of Technology at Fuxin); • Three provincial agencies (Bureaus of Land and Mineral Administration of Yunnan, Chongqing, Shanxi), and; • Two local agencies (Shahe City in Hebei province and Shanghang county in Fujian province). Representatives from the Canadian Embassy, Rio Tinto, and B H P B i l l i t i o n attended the second reception dinner, as well as Dr. Philip Andrews-Speed, from the University of Dundee and A J Gunson, from the University of British Columbia. In addition to summarizing the meeting, this report details C A S M - C h i n a ' s organization, budget, goals and work plan for 2004. The total expense for hosting the meeting was under budget at U S D $7,786.29. The meeting marks a promising start for the network, and was well received by the participants, who were eager for the opportunity to discuss the role of A S M in China. 136 CASM-China Background Following C A S M ' s September 2003 Annual Meeting in Elmina, Ghana, Shen Le i , A J Gunson, and Philip Andrews-Speed submitted a proposal to create C A S M - C h i n a . In November 2003 C A S M accepted the proposal and in December 2003 provided C A S M -China with the terms of reference and the initial payments. Formal planning for the inaugural meeting began in November 2003 under direction of the organizing committee, composed of the initial proponents. Meeting Agenda There were three main goals for C A S M - C h i n a ' s inaugural meeting: • To bring together and form a network of a wide range of A S M researchers, experts and government officials; • To design a practical organizational structure for C A S M - C h i n a ; and • To build an initial work plan for C A S M - C h i n a . The agenda was set by the organizing committee and was intended to provide plenty of time for the participants to meet each other and discuss informally the issues raised during the meeting. Table 1: Program of Events CASM-China Inaugural Meeting Program of Events 5 t h January 2004 14:00 Registration 16:00 Introductions 18:30 Welcome Dinner 6 t h January 2004 08:30 A n overview of A S M in China (SL) 09:30 A n overview of A S M and the international experience (PAS) 10.30 A n introduction to C A S M and the concept of sustainable A S M communities (AJG) 11:30 Questions and discussion 12:00 Lunch 14:00 Inaugural Business Meeting for C A S M - C h i n a 18.30 Dinner with international stakeholders 7 t h January 2004 8:30 Final wrap-up session 11:30 Closing 137 Meeting Summary Monday, January 5, 2004 The meeting began at 4 :00PM at the Beijing Foreign Experts Building, in northern Beijing. After a brief introduction by Shen Le i , all o f the participants introduced themselves in turn, explained their experience with A S M , and described what they expected to get out of the meeting. A s many of the participants were not familiar with each other, it was important to provide adequate time to get to know one another. This format is unusual for China, where often only the highest-ranking people present are given time to speak. Following introductions, the participants met informally over dinner. After dinner, the organizers met with participants with three Chinese A S M officials with some authority to discuss the potential roles of C A S M - C h i n a Tuesday, January 6, 2004 A t 8:30 A M the meeting reconvened for three presentations. Shen Le i presented on the overall state of A S M in China. He reviewed the diversity and scale of Chinese A S M , the recent history and government policies toward A S M , and the positive and negative impacts of A S M in China. Philip Andrews-Speed discussed the overall state of A S M around the world. He reviewed changes in the international community's approach toward A S M and key current questions facing policy makers and development agencies with regard to A S M . He touched on characteristics distinguishing A S M from formal mining, the costs and benefits of A S M , A S M government policy internationally and the difficulties of effective A S M policies, and methods by which A S M policies could be improved. A J Gunson presented on C A S M ' s structure, vision, role, and activities. He stressed the importance of using A S M as a tool for poverty alleviation on the road toward more sustainable communities. He then discussed the potential for C A S M to cooperate with C A S M - C h i n a and a possible structure and role for C A S M - C h i n a . Following an informal lunch, the meeting resumed with a discussion of the costs and benefits of A S M in China. Participants noted that the A S M debate in China encompassed a wide range of topics including safety and health, environmental protection, industrial distribution, and rural poverty reduction. They also commented that the Chinese media and government usually only focused on the negative aspects of A S M in China. With a view to the broader issues, participants discussed current A S M policies in China and the potential for change. One main issue discussed was the legal definition of small-scale mining in China and its legal implications regarding mineral resource law. In particular, participants were concerned about creating a clear boundary between formal large-scale mining and A S M . Participants recognized that a new national policy toward A S M is vitally necessary. Some participants argued that in the future the central government cannot continue to close small-scale mines without providing any compensation for the 138 miners. In addition, the relationship of poverty reduction with A S M and health, safety and environment training were discussed. After the discussion the participants reconvened for a more formal dinner with representatives from the Canadian Embassy, BHPBi l l i t on , and Rio Tinto. The international participants expressed great interest in the situation of A S M in China and looked forward to cooperating more with C A S M - C h i n a in the future. Following the meal, the organizers met with four A S M officials from rural communities to discuss the role and structure of C A S M - C h i n a and potential for future C A S M - C h i n a activities and projects. Wednesday, January 7, 2004 A t 8 :30AM the meeting continued with a discussion of the future structure and role of C A S M - C h i n a . Participants fully recognized the importance and necessity of creating a multidisciplinary network approach to A S M in China. The importance of formalizing C A S M - C h i n a and registering officially was stressed. Registering C A S M - C h i n a independently was perceived as time-consuming and difficult. Not registering in some capacity is illegal. It was decided that the best option was to give C A S M - C h i n a a home within the China Min ing Association, which has a A S M department. The importance was stressed of using C A S M - C h i n a ' s webpage for sharing both Chinese and international reports and papers on A S M . A n online discussion forum was also deemed important. Participants recommended that C A S M - C h i n a cooperate with the China Min ing Association and a county level government to host a local meeting with a wide range of stakeholders to discuss the role of A S M and the potential for using A S M as a vehicle toward poverty reduction and more sustainable communities. Participants also decided to both widen the C A S M - C h i n a network's membership nationally and deepen C A S M - C h i n a locally. Shen Le i thanked the participants for their involvement and hard work. Philip Andrews-Speed thanked Shen Le i and his assistance for organizing the meeting and making it a success before closing the meeting. 139 Key Meeting Achievements Over the course of the CASM-China Inaugural Meeting the following main results were achieved: • The participants acknowledge as timely and of key importance the establishment of the CASM-China Regional Network. All participants fully recognized the necessity of further study on the issues of small-scale mining in China; while China's small-scale mining faces great challenges, it holds huge potential. • Participants recommend that CASM-China become a legal organization under the China Mining Association (CMA). The C M A has broad and direct links with local small-scale miners and is closely involved with the administration and research on A S M in China. Further discussions with the C M A are required to formalize the details. • Fu Mingke, previous Director General of Department of Ministry Exploitation in Ministry of Geology and Mineral Resources (now renamed the Ministry of Land and Resources), recommended that Shen Lei be appointed as the Director of the Small Scale Mines Committee in the China Mining Association (CMA) after CASM-China is linked with the CMA. • The Senior Consultancy Center and the Center of Land and Mineral Legal Affairs, both in the Department of Land and Resources, also expressed interest in hosting CASM-China. • While it was recognized that Chinese government officials and experts had made many great efforts to study and resolve A S M issues and done lots of studies on China's small-scale mines, there is still much work to be done. Participants recognized the importance of cooperating with international organizations such as the World Bank, UNIDO, and the ILO, both in order to share China's experience with other nations and to learn from the international experience. • Participants recognized that of the many issues challenging Chinese A S M , the issues of regularization, environmental protection, health and safety and poverty alleviation should be given priority in future action plans. CASM-China Work Plan 2004 Based on the outcome of the Inaugural Meeting, a work plan and budget for the rest of 2004 is shown in Table 2 on the next page. This work plan includes; formalizing C A S M -China, developing the Chinese-language web-based knowledge center, hosting local stakeholder meetings, and formalizing a national policy paper. In early February 2004, CASM-China will initiate the work plan, starting with the formalization of CASM-China under the China Mining Association. 140 Table 2: CASM-China Work Plan and Budget 2004 Activity Details Budget Projected (USD) Actual (USD) Inaugural Meeting Held January 5-7, 2004 7,786.29 CASM-China Formalization Set up CASM-China as a department under the China Mining Association (CMA) In-kind contribution from the C M A Webpage Development • Translate and make available key Chinese and international A S M legal and policy documents 1000.00 • Setup a discussion forum to share ideas and link agencies and individuals together 500.00 Local C A S M -China Meeting • Co-host two pilot multi-stakeholder meetings in A S M counties with the C M A 3,000.00 + local county and C M A • Report meeting results back to C A S M -China Expand Network Nationally Meeting participants are to reach out to their counterparts in A S M areas unrepresented at the meeting In-kind contribution from participants Expand Network Locally Rural meeting participants are to reach out in their communities to inform and involve other A S M stake-holders In-kind contribution from participants C A S M Annual Meeting Send at least one CASM-China representative to the 2004 C A S M Annual Meeting to report on CASM-China's activities 1,500.00 National Policy Paper Write a policy paper on A S M , poverty reduction, and sustainable communities to be presented to the Chinese State Council 500.00 Projects • Approach the China Science Foundation for cooperation studying the impact on communities of A S M coal mine closure and the potential for alternative sources of income 100.00 • Approach the ILO and the State Safety Production Supervising Bureau for cooperation on a pilot mine safety project 100.00 • Approach UNIDO and the Chinese Academy of Science for cooperation on a pilot mercury reduction project 100.00 CASM-China 2005 Meeting Begin planning and fund raising for the second annual meeting of CASM-China 300.00 Transaction Costs International money transfer fees, photocopying costs, and phone fees. 213.71 Total 7,213.71 7,786.29 141 Figure 1: CASM-China Inaugural Meeting Summary C A S M - C h i n a ' s Inaugural Meeting marks a promising start for the organization. Participants from around China were successfully brought together to discuss A S M in the context of poverty reduction and sustainability. A plan to formalize C A S M - C h i n a under the China Min ing Association was developed. As detailed in the work plan, much work remains to be done, including registering the organization, developing the webpage, holding a pilot local A S M stakeholder meeting, expanding and deepening the network and investigating potential projects. 142 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0081074/manifest

Comment

Related Items