Utilizing waste energy from a submerged arc furnace at a ferrochrome smelter plant

(1)

Utilizing waste energy from a submerged

arc furnace at a ferrochrome smelter plant

R Murray

22115153

Dissertation submitted in fulfilment of the requirements for the

degree

Magister

in

Electrical and Electronic Engineering

at the

Potchefstroom Campus of the North-West University

Supervisor:

Prof JA de Kock

(2)

Declaration

I, Ruan Murray, hereby declare that the thesis entitled “Utilizing waste energy from a submerged arc furnace at a ferrochrome smelter plant” is my own original work and has

not already been submitted to any other university or institution for examination.

R Murray

Student number: 22115153

(3)

Acknowledgements

I would like to thank the following people for their contributions, guidance, time and understanding before and during the course of this project:

• Prof. Jan de Kock, my supervisor, for his guidance, support and understanding. • My mother, Gretha Murray, for her love and support.

• My family, for their love and support.

• My friends, for their understanding, assistance and support. • Dave Kenchington, for his constant willingness to help.

• The Boshoek Smelter for their support, assistance and for allowing me to do this study.

• Peter Haley and the Namakwa Sands Smelter for their support, time and willingness to help.

• Tinus du Plessis, Garth La Fleur and the Meyerton Smelter for their time and willingness to help.

• Paul de Mattos and Ruud van Groenewoud for their assistance. • BBEnergy for their support.

• My deceased father, Rickes Murray, for always being with me in spirit.

This work was supported in part by the National Research Foundation, Grant 2015-87567.

"Don’t ask yourself what the world needs. Ask yourself what makes you come alive and then go do that. Because what the world needs is people who have come alive."

(4)

Abstract

In South Africa the current electrical energy crisis and the possibility of carbon tax forces ferrochrome (FeCr) smelters to improve the efficiency of their plants and to reduce carbon emissions at the same time. This can be done by utilizing the waste energy available at the smelter plant. This dissertation presents a techno-economic evaluation of the utilization of furnace off-gas, rich in carbon monoxide (CO), by using available technologies to convert the waste off-gas energy into electrical energy, which is referred to as co-generation in this dissertation.

The results from two case studies on existing co-generation plants are used to determine the efficiency of the technologies considered (gas engines, gas turbines and a gas-fired boiler). A single FeCr smelter in South Africa is selected as the reference plant and the off-gas it produces is analysed. The results of both the energy available in the FeCr furnace off-gas and the efficiency of the co-generation plants considered are used to determine the potential of the various systems to generate electrical energy.

The established potential of the considered co-generation plants is used to conduct an economic evaluation on both the existing co-generation plants (case studies) and the FeCr smelter co-generation plants considered.

The capital cost, operating cost, net present value (NPV), internal rate of return (IRR), simple payback period (SPP) and discounted payback period (DPP) are determined for each project. The operating cost and capital cost of the proposed projects are established using information from the case studies, technology suppliers and manufacturers. A sensitivity analysis is also conducted for each project and the influence of the variation of certain parameters on the NPV and IRR is established.

The results show that a gas engine co-generation plant, with an efficiency of 28%, could save a FeCr smelter 12% to 14% on its annual electrical energy consumption (purchased from Eskom). This would reduce the carbon emissions of the smelter plant by between 10% and 12%. The co-generation plant considered would have a capital cost of around R491 million in 2015 and an annual operating cost of around R31.3 million. Economic analyses show that the project would have an NPV of -R25 million, after 15 years of operation, and an IRR of 9.2%. This results in an SPP of 8.65 years and a DPP of 16.51 years. A gas engine co-generation plant would generate electrical energy at a cost of around R0.32/kWh in 2016.

In this dissertation, the potential of a co-generation plant is evaluated for implementation at a FeCr smelter. The most feasible system is identified using the results from the economic and technical evaluation (potential for generating electrical energy and reducing carbon emissions).

Keywords: Carbon emissions, co-generation, electrical energy, ferrochrome smelter plant, gas-fired boiler, gas engines, gas turbines, off-gas, waste energy.

(5)

Opsomming

In Suid-Afrika forseer die huidige elektriese-energiekrisis en die moontlikheid van kool-stofbelasting ferrochroomsmelters om die effektiwiteit van die aanlegte verbeter en terself-dertyd die koolstofdioksiedvrystelling te verminder. Dit kan gedoen word deur gebruik te maak van die onbenutte energie wat beskikbaar is by die aanleg. Hierdie verhandeling bied ’n tegno-ekonomiese evaluering van die benutting van FeCr-oond-afgas, wat ryk is aan koolstofmonoksied (CO), deur die gebruik van beskikbare tegnologieë om die onbenutte energy (afgas) te omskep in elektriese energie - waarna in hierdie verhandeling as ko-generasie verwys word.

Die resultate van twee gevallestudies met bestaande ko-generasieaanlegte word gebruik om die effektiwiteit van sekere tegnologieë (gas-enjins, gas-aangedrewe ketel met ’n stoomturbine en gas turbines) te bepaal. ’n Enkele FeCr-smelter in Suid-Afrika is gekies as die verwysingsaan- leg en die af-gas wat dit produseer word ontleed. Die resultate van die beskikbare energie in die FeCr-oond-afgas en die rendement van die ko-generasieaan-legte word gebruik om die potensiaal in elektriese energie-opwekking te bepaal.

Die potensiaal van die ko-generasieaanlegte word gebruik om ’n ekonomiese evaluering uit te voer op die bestaande ko-generasieaanlegte (gevallestudies), asook die ko-generasie-aanlegte wat oorweeg word vir die FeCr-smelter.

Die kapitaalkoste, bedryfskoste, netto huidige waarde (NPV), interne opbrengskoers (IRR), eenvoudige terugbetalingstydperk (SPP) en afslagterugbetalingstydperk (DPP) van elke projek word bereken. Die bedryfskoste en kapitaalkoste van die voorgestelde projekte is bepaal met behulp van gevallestudies, asook die tegnologieverskaffers en -vervaardigers. ’n Sensitiwiteitsontleding word ook gedoen vir elke projek en die invloed van die verander-ing in sekere waardes op die NPV en IRR word bepaal.

Die resultate toon dat ’n gas-enjinaanleg, met ’n rendement van 28%, ’n FeCr-smelter tussen 12% en 14% aan jaarlikse elektriese energieverbruik kan spaar. Dit sal ook die koolstofvrystelling van die smelteraanleg met tussen 10% en 12% verminder. Hierdie gas-enjinaanleg het ’n projekkoste van sowat R491 miljoen in 2015 en ’n jaarlikse bedryfs-koste van sowat R31.3 miljoen. Die ekonomiese analise toon dat die projek ’n NPV van -R25 miljoen, na 15 jaar, en ’n IRR van 9.2% sal hê. Dit lei tot ’n SPP van 8.65 jaar en ’n DPP van 16.51 jaar. ’n Gas-enjinko-generasieaanleg sal elektriese energie opwek teen ’n koste van sowat R0.32/kWh in 2016.

In hierdie verhandeling word die potensiaal van die implementering van ’n ko-generasie-aanleg by ’n ferrochroomsmelterko-generasie-aanleg geëvalueer. Die beste stelsel word geïdentifiseer met behulp van die ekonomiese en tegniese evaluering (potensiële elektriese energie wat gegenereer kan word en verminderde koolstofvrystellings).

Sleutelwoorde: Afgas, elektriese energie, ferrochroomsmelteraanleg, gas-aangedrewe ketel, gas-enjins, gasturbines, koolstofvrystelling, ko-generasie, onbenutte energie.

(6)

Contributions of this Study

This study contributed the following research output:

• A conference paper (published in proceedings) and presentation at the Industrial and Commercial Use of Energy (ICUE) conference in Cape Town (South Africa) on 19 August 2015, entitled: Potential in utilising off-gas at a ferrochrome smelter with gas engines by R Murray and JA de Kock. This paper received the “Highly Awarded Paper” award at the conference.

• A conference paper (published in proceedings) and presentation at the Southern African Universities Power Engineering Conference (SAUPEC) on 27 January 2016, entitled: Potential in utilising ferrochrome furnace off-gas in South Africa: A techno-economic studyby R Murray and JA de Kock.

• An article published in the January 2016 issue of the WattNow magazine (the SAIEE magazine), pages 26 to 32, entitled: Potential in utilising furnace off-gas in South Africaby R Murray and JA de Kock.

(7)

List of Figures

1.1 Energy resources used in South Africa [1] . . . 1

2.1 Components of a basic electric arc furnace [8] . . . 6

2.2 The cycle for steel melting in an electric arc furnace [9] . . . 7

2.3 Basic submerged arc furnace [12] . . . 9

2.4 Process flow of ferrochrome smelter plants in South Africa [13] . . . 11

2.5 Outokumpu ferrochrome process [14] . . . 12

2.6 Material balance for the production of one tonne of ferrochrome [15] . . . 13

2.7 Wankel combustion engine - operating cycle [21] . . . 17

2.8 Spark and compression ignition internal combustion engines (four-stroke) [22] . . . 17

2.9 Basic steam engine cylinder operation [26] . . . 19

2.10 Basic configurations of the Stirling engine [28] . . . 20

2.11 Basic components of the GE-Jenbacher gas engine [33] . . . 21

2.12 Gas engine energy balance [33] . . . 22

2.13 Dual-fuel operation [36] . . . 23

2.14 Block diagram of a combined heat and power steam turbine system [39] . 25 2.15 Diagram for electrical energy generation using a gas turbine [39] . . . 26

2.16 Basic boiler diagram [42] . . . 28

2.17 Boiler flue gas waste heat recovery [42] . . . 28

2.18 Simplified water tube boiler [45] . . . 31

2.19 Conventional boiler turbine-generator [46] . . . 32

2.20 Internal combustion gas engine-generator [47] . . . 32

2.21 Conventional engine-generator set with an additional boiler [46] . . . 32

2.22 Combined cycle gas turbine [46] . . . 33

2.23 Gas turbine combined heat and power system (regenerative cycle) [48] . . 34

2.24 Gas turbine-generator combined heat, power and cooling system [48] . . 34

2.25 Gas engine-generators in a combined heat and power system [50] . . . 35

2.26 Gas engine-generator in a combined heat, power and cooling system [51] 35 2.27 Main components of the organic Rankine cycle [54] . . . 36

2.28 Organic Rankine cycle unit [56] . . . 37

4.1 FeCr smelter process flow [62] . . . 43

4.2 FeCr smelter off-gas flow diagram . . . 43

4.3 Outokumpu process - FeCr smelter . . . 44

4.4 FeCr smelter off-gas composition histograms - Feb. and Mar. 2015 . . . . 45

4.5 FeCr smelter daily average gas composition - Feb. and Mar. 2015 . . . . 46

4.6 FeCr smelter off-gas users - Feb. and Mar. 2015 . . . 47

4.7 FeCr smelter daily off-gas usage - Feb. and Mar. 2015 . . . 47

4.8 FeCr smelter off-gas usage - Feb. and Mar. 2015 . . . 48

4.9 FeCr smelter off-gas produced - 2014 . . . 49

4.10 FeCr smelter off-gas gravimetric calorific values - Feb. and Mar. 2015 . . 49

(11)

LIST OF FIGURES LIST OF FIGURES

4.12 FeCr smelter calorific value histograms - Feb. and Mar. 2015 . . . 50

4.13 FeCr smelter daily energy availability - Feb. and Mar. 2015 . . . 51

4.14 FeCr smelter energy summary - 2014 . . . 52

4.15 FeCr smelter electrical energy consumption - 2014 . . . 55

4.16 FeCr smelter estimated CO2emissions - 2014 . . . 55

5.1 Basic DC arc furnace [64] . . . 57

5.2 Ilmenite smelter off-gas flow diagram (excluding co-generation plant) [65] 58 5.3 Ilmenite smelter off-gas flow diagram (including co-generation plant) [65] 59 5.4 Ilmenite smelter co-generation plant layout . . . 60

5.5 Ilmenite smelter co-generation plant single line diagram . . . 61

5.6 Ilmenite smelter daily average histograms for CO and H2- 2014 . . . 62

5.7 Ilmenite smelter off-gas daily average CO and H2content - 2014 . . . 63

5.8 Ilmenite smelter off-gas distribution between gas users - 2014 . . . 64

5.9 Off-gas produced by ilmenite furnaces and utilization thereof - 2014 . . . 64

5.10 Ilmenite smelter daily average off-gas calorific value in 2014 - Gravimetric 65 5.11 Ilmenite smelter daily average off-gas calorific value in 2014 - Volumetric 66 5.12 Ilmenite smelter off-gas calorific value histograms - 2014 . . . 66

5.13 Energy available in ilmenite furnace off-gas - 2014 . . . 67

5.14 Energy generated and exported by co-generation plant at the ilmenite smelter - 2014 . . . 68

5.15 Ilmenite smelter electrical energy consumed, exported and purchased - 2014 69 5.16 Percentage electrical energy saved by the ilmenite smelter - 2014 . . . 69

5.17 Ilmenite smelter co-generation plant and off-gas utilization . . . 70

5.18 Ilmenite smelter co-generation plant utilization - Pie chart . . . 70

5.19 Off-gas energy supplied to ilmenite smelter co-generation plant and electrical energy exported - 2014 . . . 71

5.20 Ilmenite smelter gas engine and co-generation plant efficiencies - 2014 . . 72

5.21 Ilmenite smelter carbon emission forecast and reduction - 2014 . . . 73

5.22 Percentage reduction in CO2emissions at ilmenite smelter - 2014 . . . . 73

5.23 FeMn smelter off-gas flow diagram (including co-generation plant) . . . . 77

5.24 FeMn co-generation plant layout . . . 77

5.25 FeMn smelter co-generation plant single line diagram . . . 79

5.26 FeMn smelter off-gas CO and H2content - 2014 . . . 80

5.27 FeMn smelter daily average off-gas composition - 2014 . . . 81

5.28 FeMn smelter off-gas distribution between gas users - 2014 . . . 82

5.29 Off-gas produced by FeMn furnaces and utilization thereof - 2014 . . . . 82

5.30 FeMn smelter daily average off-gas calorific value in 2014 - Gravimetric . 83 5.31 FeMn smelter daily average off-gas calorific value in 2014 - Volumetric . 83 5.32 FeMn smelter off-gas calorific value histograms - 2014 . . . 84

5.33 Energy available in FeMn furnace off-gas - 2014 . . . 85

5.34 FeMn smelter electrical energy consumed, exported and purchased - 2014 86 5.35 Percentage electrical energy saved by FeMn smelter - 2014 . . . 86

5.36 FeMn smelter co-generation plant and off-gas utilization . . . 87

5.37 FeMn smelter co-generation plant utilization - Pie chart . . . 87

5.38 Off-gas energy supplied to FeMn smelter co-generation plant and electrical energy exported - 2014 . . . 88

(12)

LIST OF FIGURES LIST OF FIGURES

5.40 FeMn smelter carbon emission forecast and reduction - 2014 . . . 89

5.41 Percentage reduction in CO2emissions at FeMn smelter - 2014 . . . 90

6.1 Proposed gas engine co-generation plant layout . . . 93

6.2 Proposed gas-fired boiler co-generation plant layout . . . 94

6.3 Proposed gas turbine co-generation plant layout . . . 95

7.1 High-level layout of economic analysis model . . . 98

7.2 Original co-generation project generation cost forecast (2014) - Ilmenite smelter [67] . . . 100

7.3 Co-generation plant cash flow - Ilmenite smelter . . . 101

7.4 Electrical energy cost forecast - Ilmenite smelter . . . 102

7.5 NPV sensitivity of ilmenite smelter co-generation project - Scenario 1 . . 103

7.6 IRR sensitivity of ilmenite smelter co-generation project - Scenario 1 . . . 104

7.7 Ilmenite smelter co-generation project sensitivity to electrical energy cost - Scenario 1 . . . 105

7.8 Ilmenite smelter co-generation project sensitivity to carbon tax and carbon credit . . . 106

7.9 Co-generation plant cash flow - FeMn smelter . . . 108

7.10 Electrical energy cost forecast - FeMn smelter . . . 110

7.11 NPV sensitivity of FeMn smelter co-generation project - Scenario 1 . . . 111

7.12 IRR sensitivity of FeMn smelter co-generation project - Scenario 1 . . . . 112

7.13 FeMn smelter cogeneration project sensitivity to electrical energy cost -Scenario 1 . . . 113

7.14 FeMn smelter co-generation project sensitivity to carbon tax and carbon credit . . . 113

7.15 Gas engine co-generation plant cash flow - FeCr smelter . . . 115

7.16 Gas engine co-generation plant electrical energy cost forecast - FeCr smelter116 7.17 Gas-fired boiler co-generation plant cash flow - FeCr smelter . . . 118

7.18 Gas-fired boiler co-generation plant electrical energy cost forecast - FeCr smelter . . . 119

7.19 Gas turbine co-generation plant cash flow - FeCr smelter . . . 121

7.20 Gas turbine co-generation plant electrical energy cost forecast - FeCr smelter122 7.21 Electrical energy cost forecast summary for considered co-generation plants - FeCr smelter . . . 125

B.1 NPV sensitivity of scenarios 2, 3, and 4 - Ilmenite smelter . . . 134

B.2 IRR sensitivity of scenarios 2, 3, and 4 - Ilmenite smelter . . . 135

B.3 NPV sensitivity of scenarios 2, 3, and 4 - FeMn smelter . . . 136

B.4 IRR sensitivity of scenarios 2, 3, and 4 - FeMn smelter . . . 137

B.5 Sensitivity to electrical energy cost(scenarios 2, 3 and 4) - Ilmenite smelter 138 B.6 Sensitivity to electrical energy cost(scenarios 2, 3 and 4) - FeMn smelter . 139 C.1 NPV sensitivity of FeCr smelter co-generation plants - Scenario 1 . . . . 140

(13)

List of Tables

2.1 Common products produced by a submerged arc furnace [10] . . . 8

2.2 Typical composition of ferrochrome furnace off-gas [15] . . . 13

2.3 Categories of particles [17] . . . 14

2.4 Fuel properties [18] . . . 15

2.5 Classification of steam turbines [21] . . . 24

4.1 Ferrochrome smelter site details . . . 42

4.2 Plant measured ferrochrome furnace off-gas composition - Feb. and Mar. 2015 . . . 46

4.3 FeCr smelter off-gas energy summary - 2014 . . . 52

4.4 FeCr furnace off-gas characteristics - Independent measurements . . . 53

4.5 FeCr smelter off-gas base value summary . . . 54

5.1 Purpose of major components in the gas trains of the ilmenite smelter co-generation plant . . . 59

5.2 Ilmenite smelter off-gas composition summary - 2014 . . . 63

5.3 Off-gas volumes produced and utilized at the ilmenite smelter - 2014 . . . 65

5.4 Ilmenite smelter off-gas calorific value - 2014 . . . 67

5.5 Ilmenite smelter co-generation plant energy efficiency summary - 2014 . . 72

5.6 Extra gas engine technical comparison - Ilmenite smelter . . . 74

5.7 Purpose of FeMn smelter co-generation plant components . . . 78

5.8 FeMn smelter off-gas composition summary - 2014 . . . 81

5.9 Off-gas volumes produced and utilized at the FeMn smelter - 2014 . . . . 83

5.10 FeMn smelter off-gas calorific value - 2014 . . . 84

5.11 FeMn smelter co-generation plant energy efficiency summary - 2014 . . . 89

6.1 Ferrochrome smelter base values . . . 92

6.2 Technical summary of FeCr smelter co-generation plants considered . . . 96

7.1 Economic evaluation scenarios considered . . . 97

7.2 Actual co-generation plant capital cost (2012) - Ilmenite smelter . . . 99

7.3 Annual estimated increase in co-generation plant capital cost - Ilmenite smelter . . . 99

7.4 Actual co-generation plant operating cost (2014) - Ilmenite smelter . . . . 100

7.5 NPV base values for each scenario - Ilmenite smelter co-generation plant 104 7.6 IRR base values for each scenario - Ilmenite smelter co-generation plant . 104 7.7 Extra gas engine economic comparison - Ilmenite smelter . . . 106

7.8 Actual co-generation plant capital cost (1996) - FeMn smelter . . . 107

7.9 Annual estimated increase in co-generation capital cost - FeMn smelter . 107 7.10 Actual co-generation plant operating cost (2014) - FeMn smelter . . . 108

7.11 NPV base values for each scenario - FeMn smelter co-generation plant . . 111

7.12 IRR base values for each scenario - FeMn smelter co-generation plant . . 112 7.13 Estimated capital cost of a gas engine co-generation plant - FeCr smelter . 114 7.14 Estimated operating cost of a gas engine co-generation plant - FeCr smelter115

(14)

LIST OF TABLES LIST OF TABLES

7.15 Estimated capital cost of a gas-fired boiler co-generation plant - FeCr smelter . . . 117 7.16 Estimated operating cost of a gas-fired boiler co-generation plant - FeCr

smelter . . . 117 7.17 Estimated project cost of a gas turbine co-generation plant - FeCr smelter 120 7.18 Estimated operating cost of a gas turbine co-generation plant - FeCr smelter121 7.19 NPV sensitivity of each co-generation plant (Scenario 1) - FeCr smelter . 123 7.20 IRR sensitivity of each co-generation plant (Scenario 1) - FeCr smelter . . 123 7.21 Existing co-generation plants economic evaluation summary (case studies) 124 7.22 FeCr smelter considered co-generation plants economic evaluation summary124

(15)

List of Abbreviations

AC Alternating current BFG Blast furnace gas

BL Black liquor CCGT Combined cycle gas turbine CEF Carbon emission factor CH4 Methane

CHP Combined heat and power CHPC Combined heat power and cooling

CI Compression ignition CO Carbon monoxide

CO2 Carbon dioxide CO2e Carbon dioxide equivalent

COG Coke oven gas CV Calorific value

DC Direct current DPP Discounted payback period EAF Electric arc furnace EC External combustion

EE Electrical energy ERR External rate of return Eskom Electricity supply commission

of South Africa EURO Single European union currency

Fe Iron FeCr Ferrochrome

FeMn Ferromanganese FO Fuel oil

GCP Gas-cleaning plant H2 Hydrogen

HHV Higher heating value IC Internal combustion IRR Internal rate of return LHV Lower heating value MARR Minimum acceptable rate of

return N2 Nitrogen

NG Natural gas NPV Net present value

O2 Oxygen ORC Organic Rankine cycle

pa Per annum PF Power factor

SA South Africa SAF Submerged arc furnace

SH Super heater SI Spark ignition

(16)

List of Units

c South African cent oC degrees centigrade c/kWh cent per kilowatt-hour GWh gigawatt-hour

h hour J joule

kg kilogram kg/Nm3 kilogram per normal cubic metre

kg/s kilogram per second kNm3 kilo-normal cubic metre

kPa kilopascal kt kilotonne (metric)

kV kilovolt kWh kilowatt-hour

kWh/kg kilowatt-hour per

kilogram kWh/Nm

3 kilowatt-hour per normal

cubic metre

m metre m3 cubic metre

mbar millibar mg milligram

mg/s milligram per second MJ megajoule

MJ/kg megajoule per kilogram MJ/Nm3 megajoule per normal cubic metre

MNm3 mega-normal cubic metre MVA megavolt-ampere

MW megawatt MWh megawatt-hour

Nm3 normal cubic metre Nm3/h normal cubic metre per hour

Nm3/h/MW normal cubic metre per

hour per megawatt R South African rand R/kWh rand per kilowatt-hour rpm revolutions per minute

R/t rand per metric tonne s seconds

(17)

1. Introduction

In order to understand the reasons for a study like this, a quick background is given in the first chapter of this dissertation on CO₂taxes in South Africa and the need for utilization of waste energy at smelter plants. The problem statement, scope and objectives of the research project are also given.

1.1 Background

Climate change is a great risk for humankind, and one way of reducing this risk is to reduce carbon dioxide (CO2) emissions, which constitute about 80% of the total emission

of greenhouse gases in South Africa [1]. South Africa has one of the most carbon-intensive economies in the world [2]. One cause of this is the mining industry, which plays a major part in the South African economy. In addition, most electrical energy used in the country is generated by coal-fired power stations operated by Eskom.

Eskom is a state-owned enterprise, whichgeneratesalmost all of South Africa’s electrical energy (95%) [1], with only two per cent of the electrical energy being produced by private companies. This makes the mining industry very dependent on the power supplied by Eskom. The energy resource used to produce 86% of the electrical energy in South Africa is coal, as seen in figure 1.1 [1], where CO2 emissions are reflected. Eskom is currently

operating at almost full capacity with a reserve capacity of less than 10%, making it difficult for it to supply enough power during peak power usage times. This was one of the reasons why South Africa experienced a shortage of power in 2008 and 2014/15.

Figure 1.1: Energy resources used in South Africa [1]

The previous South African Minister of Finance, Pravin Gordhan, announced that carbon tax would be introduced as from the first of January 2015 [3]. However, it was recently announced that the payment of carbon tax in South Africa had been postponed to 2016 [4]. The rate for this tax is set at R120/t for the CO2that is emitted, although companies will

only pay tax on between 20% and 40% of their total emissions, resulting in an equated estimated rate of between R24/t and R48/t. This will be implemented in two five-year

(18)

1.2. PROBLEM STATEMENT CHAPTER 1. INTRODUCTION

phases and the price will be increased annually by 10% during the first phase [5]. The introduction of carbon tax will have a major impact on the mining industries as well as on Eskom and is likely to cause electrical energy to become even more expensive.

With the introduction of carbon tax and the power supply from Eskom being regarded as unreliable by some people, businesses in the mining sector need to consider implementing technology to reduce their electrical energy consumption and their carbon emissions. This needs to be done in the near future if a company is to be profitable, as carbon taxes and the price of electrical energy could increase.

When considering a smelter plant in South Africa and the whole process that is followed to create the final product, there are various areas where waste energy can be utilized. One option is to use the waste gases (off-gas) produced by the furnaces, at the smelters, to generate electrical energy. The thermal energy (heat), as well as the gas itself, can be used for power generation purposes or for supplying heat to other processes or buildings.

1.2 Problem statement

In this dissertation the potential of using waste energy, in the form of off-gas, is considered for a ferrochrome (FeCr) smelter plant. Energy in the form of off-gas and heat can easily be wasted at smelting plants (furnaces) in South Africa. In South Africa’s mining industry a business now more than ever needs to implement forms of waste energy utilization in order to be profitable, with carbon tax becoming a reality, as well as for the sake of the environment.

One business that wants to startinvestigating ways of reducing its carbon emissions and use waste energy, at its furnaces, is a FeCr smelter at Boshoek (South Africa). This smelter plant has two submerged arc furnaces (SAFs), of 55 MVA each, that are used in the smelting process. Data from the Boshoek FeCr smelter, regarded as the reference plant, are gathered and analysed in this dissertation.

There are a number of ways in which electrical energy could be generated and energy recycled using waste resources. When considering a FeCr smelter, electrical energy could be generated using the off-gas produced during the smelting process, which could also reduce carbon emissions. Another option could be to recycle the waste thermal energy (heat) or to use this energy to generate electrical energy.

A few systems like this already exist, where electrical energy is generated from off-gas or waste heat. Without thorough knowledge of the products and systems that are available, one can easily implement an inefficient system. Therefore, in order to identify the most efficient and feasible system for a specific smelter (in this case FeCr) the available designs, products and systems need to be thoroughly evaluated.

Two case studies are conducted using data from two smelter plants, which already use their furnace off-gas for electrical energy generation. The first is an ilmenite smelter with two DC-arc furnaces where titania (TiO2) and iron (Fe) are produced. They use

the off-gas by generating electrical energy with gas engine-generators. The second is a ferromanganese smelter with four AC-arc furnaces, which uses the furnace off-gas by

(19)

1.3. PROJECT SCOPE CHAPTER 1. INTRODUCTION

generating electrical energy with a gas-fired boiler and steam turbine-generator.

In the case studies of the two smelter plants, a technical and an economic evaluation of the installed co-generation plants are conducted. The results are used to determine the potential of similar plants at the FeCr smelter. Each of the plants considered is also technically and economically evaluated.

In this document, an evaluation is done of the possible designs, products and systems that could be used at the Boshoek smelter, near Rustenburg in South Africa. The data from their two furnaces are used to determine the amount of waste energy (heat and/or off-gas), as well as the form and composition thereof. By using this information and doing research on the possible systems that can be used, the most feasible system for utilizing the waste energy at the furnaces can be presented.

1.3 Project scope

This dissertation entails the research of possible systems for utilizing waste energy, at the submerged arc furnaces of the Boshoek FeCr smelter plant. The off-gas that is produced is considered and options are identified for generating electrical energy, possibly reducing carbon emissions and increasing the overall efficiency of energy conversion at the submerged electric arc furnaces (EAFs). The research is only done for FeCr furnaces, but could aid in solving the same problem at other smelters. In the final chapter of this dissertation a feasible system for using the waste energy (off-gas), and through that increasing the overall efficiency, is presented.

1.4 Project objectives

The aim of this research project is to:

• Gather information on the FeCr smelter (Boshoek) operation and the waste energy available, as well as the composition thereof.

– Describe the operation of the Boshoek smelter. – Define the waste energy available at the furnace. – Accumulate available information on the waste energy.

∗ Composition of the off-gas. ∗ Produced volume of the off-gas. ∗ Energy available in the off-gas. – Define constraints on the system.

(20)

1.5. OVERVIEW CHAPTER 1. INTRODUCTION

• Collect information on possible options for generating electrical energy from the waste gases and heat at a FeCr smelter.

– Gather information on existing systems (solutions). – Determine possible layout of feasible plants or systems. – Define and describe the layout of each system.

• Evaluate the different options.

– Determine total amount of electrical energy generated. – Assess system (plant) efficiency.

– Determine total reduction of carbon emissions. – Conduct economic analyses.

• Recommend a system that is most suitable for ferrochrome submerged arc furnaces. – Base recommendations on:

∗ Economic analyses.

∗ Total reduction in carbon emissions.

∗ Total electrical energy that could be generated. ∗ Ease of implementation.

1.5 Overview

In summary, this dissertation considers a FeCr smelter as the base plant for energy recovery and/or utilization. Existing plants that already utilize the off-gas of their furnaces are considered as case studies and evaluated. From these existing plant results and other available technologies a few plants are defined and evaluated for implementation at the FeCr smelter. An economic analysis is conducted on each plant considered and the results are used to determine the most feasible option for utilizing the waste energy, off-gas, at the FeCr smelter plant.

(21)

2. Literature Study

A literature study is conducted using the problem and objectives of this dissertation, as discussed in the previous chapter. In the literature study different technologies and designs are discussed, the main goal being to shed light on the wide variety of prime movers, furnaces and systems available for off-gas utilization. Some furnaces, different types of engines, turbines and boilers are briefly looked at and their efficiencies, implementation, advantages and disadvantages are considered. Economic evaluation methods, calorific value (CV) calculations, gas-cleaning techniques and a basic description of carbon tax and carbon emissions are also discussed in this chapter.

2.1 Electric arc furnace

Many types of furnaces are used for a wide variety of applications. In this section, the operation and basic layout of electric arc furnaces (EAFs) are discussed, as these are the furnaces that will be focused on in this study. Two types of furnaces are considered: the open electric arc furnace and the closed submerged arc furnace.

2.1.1 AC open electric arc furnace

Electric arc furnaces are primarily used in the steelmaking industry, as scrap smelters, and in foundry operations. Over the past few years the use of EAFs for steel production has grown perceptibly.

The main reason for this is that using the EAF results in lower cost steel production, when compared to a similar blast furnace [6]. The electric energy required, 500 kWh, to create one ton (0.907 metric tonne) of liquid steel in an EAF is only a small amount of the 4.7 MWh to 5.8 MWh required when using a blast furnace or a basic oxygen furnace [7]. Construction

An EAF has many components, as seen in figure 2.1. These components are responsible for several functions, such as the containment of the molten steel and scrap, movement of structural pieces, supporting the supply of electrical power to the furnace and the auxiliary process equipment (usually located on and around the furnace) [8].

The EAF has a cylindrical shape and a removable roof, vertical cylindrical sidewalls and a bottom designed to contain and withstand the thermal energy generated during the smelting process. The side wall and roof of the furnace usually consist of water-cooled panels. The centre of the roof, where the electrodes enter the furnace (roof delta), is made from refractory (material that retains its strength at high temperatures) and can be water-cooled. The bottom of the furnace is made from a steel shell and several layers of refractory [6].

(22)

2.1. ELECTRIC ARC FURNACE CHAPTER 2. LITERATURE STUDY

which can move vertically. These electrodes are connected to a transformer (which supplies the electrical power). This transformer is usually located as close to the furnace as possible in order to reduce the losses due to the high currents employed during furnace operation.

A spout is built into the furnace, as well as a working door, and most EAFs can be tilted on rockers for tapping purposes. Many of these furnaces are also equipped with oxy-fuel burners, installed in the roof and sidewall, which are used to ensure uniform meltdown of materials by heating up cold spots in the furnace [9].

Figure 2.1: Components of a basic electric arc furnace [8]

Operation

At the start of the steel-melting cycle the roof and electrodes are raised and moved to one side while the furnace is charged (loaded) with steel scrap, which is dropped into the furnace by a clamshell bucket (two to three buckets of scrap steel [7]). After this, the roof is replaced and the electrodes are lowered and current is supplied to the electrodes. The

(23)

electrodes bore through the material in order to form liquid metal, while the scrap protects the lining of the furnace from the heat of the electric arc during the meltdown period. The arc is lengthened eventually by increasing the voltage to maximum power.

Finally, when almost all the metal has melted, the arc is shortened to reduce radiation heat losses and to avoid damage to the refractory of the furnace and then oxygen is injected to oxidise the carbon in the steel [9]. When the smelting and refining of the steel are completed the heat (liquid metal) is tapped into a ladle for casting. In figure 2.2 the cycle for steel melting, as briefly discussed above, is shown.

Figure 2.2: The cycle for steel melting in an electric arc furnace [9]

2.1.2 AC closed submerged electric arc furnace

Submerged arc furnaces (SAFs) are different from other furnaces because they can be used in a continuous process or a batch process. This furnace is called an SAF because the electrodes are usually buried deep in the furnace and the reaction takes place close to the tip of the electrodes [10]. This furnace is usually used to reduce minerals and raw materials to metals or metal compounds.

(24)

SAFs are the heart of ferro-alloy production plants and are widely used to produce a variety of metal products. Table 2.1 shows a few common products that are produced using an SAF.

Table 2.1: Common products produced by a submerged arc furnace [10] Products: Silicon metal Ferrosilicon Ferromanganese Silico-manganese Ferrochromium Calcium carbide Ferroboron

Phosphorous and ferrophosphorous Vanadium

Calcium silicon Ferronickel

Construction

The basic construction of a submerged arc furnace is closely related to that of the electric arc furnace. The SAF also has a cylindrical shape and the casing of the furnace is made of sheet steel. The lower part is lined with hard and strongly calcined carbon blocks (for high-temperature working conditions) and the upper part of the furnace is lined with fire-bricks [11]. These furnaces usually have a side-wall water-cooling system, while the bottom of the furnace is cooled with air ventilation and the roof is constructed from refractory and water-cooled panels for further cooling.

The roof of the furnace has three holes for the three (can also be six) electrodes, a few feeding chutes for feeding the raw materials into the furnace and then also a pipeline (also water-cooled) that regulates the flow of the off-gas out of the furnace. The electrodes used in an SAF are either self-baking electrodes or pre-baked electrodes and the electrodes are regularly extended by new pieces. These electrodes are formed using a green carbon paste and welding steel casings around the paste. The electrodes are semi-automatically slipped into the furnace bath and the slipping is controlled by a hydraulic system. The devices used to control and regulate the movement (slipping and holding) of the electrodes into the furnace are located all along the constructed electrode column [12].

The electric power is supplied by the furnace transformer to the electrodes via a high current line (bus tubes), which are also water-cooled and connected to the electrodes via contact clamps. In figure 2.3 a basic SAF is shown.

(25)

Figure 2.3: Basic submerged arc furnace [12]

Operation

As mentioned, the melting of material in a submerged arc furnace can either be a batch process or a continuous process. Charge mix is periodically loaded into the top of the furnace and melted to produce the required molten metal. During the operation of an SAF, the raw materials are fed into the furnace through the feeding chutes, see figure 2.3. The electrodes are used to heat up and eventually melt the materials. As the carbon is heated in the furnace it becomes a conductor (resistance decreases) and creates a conducting path between the electrodes [10]. The secondary voltage used for SAFs are lower and the currents higher than those used for the basic electric arc furnace. This is because of the lower resistance of the charge mix in the SAF, which also means that the electrodes used have a larger diameter.

As the materials melt inside the furnace the liquids, the slag and metal, are tapped through the tap hole. Not all the material inside the furnace is melted when the tapping process starts, and the tap hole is opened (usually drilled) and closed with a mud-gun right after the tapping process has been completed. The metal and the slag are separated during this process.

In the operation of an SAF, three basic processes can be used. The first process is the slagless process, where the arc occurs between the electrodes and the metal bath. In the second, the slag process, a coke bed is formed under the electrodes, which floats on top of the slag. The current passes through the coke bed as well as the slag before reaching the molten metal. Arcing is usually minimized in this process [10]. Lastly, in the slag resistance process the heat is generated by the current that moves through the slag. During this process no coke bed forms and no arcing occurs.

(26)

2.2. FERROCHROME SMELTER PLANT CHAPTER 2. LITERATURE STUDY

2.2 Ferrochrome smelter plant

Smelter plants in South Africa produce various ferro-alloys using mostly electric arc furnaces. In this dissertation, a ferrochrome smelter plant is considered, which is why this section will focus on the basic layout and operation of a ferrochrome smelter plant. In the basic operation of a FeCr smelter, a number of processes occur. Figure 2.4 shows the basic operational flow of most FeCr smelters in South Africa.

In this operation, the raw materials are received and stored on piles. The fine ore material is sent to the sinter plant where pellets are formed. These pellets, as well as the coarse raw material, are used in the batching process where they are weighed and the desired mass is moved to the furnace structure. The materials from the sinter plant (pellets) then move to the pre-heater, where thermal energy is used to heat the pellets in order to reduce the amount of electrical energy used when melting them in the furnace. The furnace is charged with the raw materials, which include the ore, reductants (char, coke, anthracite and coal), fluxes (quartz, limestone, magnesite and dolomite) and the pellets from the sinter plant [13]. The grade of the metal to be made determines the combination in which the raw materials are fed to the furnaces.

The smelting occurs in the furnace and the off-gas that is produced is extracted through an off-gas pipeline, which is situated on top of the furnace roof. This off-gas flows to the scrubbers, where the gas temperature is reduced and particulate matter is removed from the gas. Some of the off-gas (usually CO) is then used for preheating and sintering purposes and the rest of the off-gas is flared off into the atmosphere [13].

After the materials in the furnace have melted, the tapping process starts. In this process, the materials inside the furnace are tapped through the tap hole. As the furnace contents flow out of the furnace the metal (FeCr) is separated from the slag. The metal and slag are cooled, after which the slag is sent to the recovery plant to make sure that no FeCr is lost. The metal (FeCr) is crushed and transported to the customer.

In figure 2.5 the basic ferrochrome production process is shown. In this figure, two furnaces are shown on the layout of the plant. The water plant is also included in this figure and the water is used at the furnace for various cooling purposes. This is a closed-loop system. Gas-cleaning systems, used to remove particulate matter from the off-gas produced, are also shown in the layout.

More detail regarding the composition of the furnace off-gas produced and particulate matter content is discussed in the following section. The methods for removing these particulates will follow after that.

(27)

2.2. FERROCHROME SMELTER PLANT CHAPTER 2. LITERATURE STUDY Figure 2.4: Process flo w of ferrochrome smelter plants in South Africa [13 ]

(28)

2.2. FERROCHROME SMELTER PLANT CHAPTER 2. LITERATURE STUDY Figure 2.5: Outokumpu ferrochrome process [14 ]

(29)

2.2. FERROCHROME SMELTER PLANT CHAPTER 2. LITERATURE STUDY

2.2.1 Ferrochrome furnace off-gas

In the process of producing ferrochrome in a closed submerged arc furnace, CO-rich off-gas is produced as a by-product. Figure 2.6 shows a typical material balance of a FeCr-producing submerged arc furnace. The figure shows that around 34 kmol of off-gas is produced per tonne of ferrochrome. This means that in a well-sealed closed submerged arc furnace off-gas is produced at a rate of 650 Nm3/t to 750 Nm3/t, or 220 Nm3/h/MW to 250 Nm3/h/MW, with reaction energy of between 2100 kWh and 2300 kWh (7550 MJ to 8300 MJ) [15] [16].

Figure 2.6: Material balance for the production of one tonne of ferrochrome [15]

The typical off-gas composition of the gas produced during the ferrochrome smelting process is given in table 5.8. As seen in the table, a very high percentage of CO gas is produced in this process. The solid content or particles in the unclean furnace off-gas can typically be between 35 g/Nm3and 45 g/Nm3[16].

Table 2.2: Typical composition of ferrochrome furnace off-gas [15] Gas Composition (Volume - %)

CO 85 − 90

CO2 2 − 5

H2 5 − 7

N2 2 − 5

(30)

2.3. GAS CLEANING CHAPTER 2. LITERATURE STUDY

2.3 Gas cleaning

Furnace off-gas needs to be cleaned, before being emitted to the atmosphere, to stay within the regulations formulated by the government. The main aim of these regulations is to reduce atmospheric pollution caused by the industrial and mining sector.

Gas-cleaning plants (GCPs) are installed at the furnaces of smelters and differ depending on the off-gas composition and particulate matter content. At certain plants the off-gas goes through an additional cleaning process to reduce particulate matter content, when used for heating or power generation purposes. This is done to make sure that the off-gas supplied to the plants is within their specifications.

The particulate matter mentioned includes small particles and mist droplets, which need to be removed by a gas-cleaning system. Particles and droplets may be classified as grit, dust, fume, smoke, mist, aerosol or smog [17] (see table 2.3). These particles in the off-gas can cause damage to and corrosion of equipment and pipelines used for transporting the gas.

Table 2.3: Categories of particles [17]

Grit Coarse particles, larger than 76 um (size of opening in 200 mesh sieve). Dust Particles between 1 um and 76 um (able to pass through 200 mesh

sieve).

Fume Particles smaller than 1 um (solid). Mist Particles smaller than 10 um (liquid). Fog Mist that is dense enough to obscure vision.

Smoke Generally waste products from combustion and may be fly ash or products of incomplete combustion, or both. Particles may be liquid or solid.

Smog A combination of smoke and fog. There are two kinds: Los Angeles -Photo-chemical and comes from motor-car exhausts; London - Comes from incomplete combustion of coal.

Soot Aggregated particles of unburned carbon produced by incomplete combustion.

Aerosols Generally applied to all air-borne suspensions.

2.3.1 Techniques

Process gases flow along a duct and/or pipelines to the handling equipment or gas-cleaning plant. There are various methods for gas-cleaning off-gas and these methods depend on the nature of the material that needs to be removed. If the material that needs to be removed is a gas, it can either be brought into contact with a material that absorbs it or changed chemically.

The unwanted gas can be absorbed into either a liquid or a solid. A liquid is usually used for gases that form a great part of the gas stream, such as hydrochloric acid vapour, ammonia, sulphur dioxide and carbon dioxide. A solid is more commonly used for absorption when a small quantity of the gas stream contains gases such as water vapour

(31)

2.4. CALORIFIC VALUE CHAPTER 2. LITERATURE STUDY

on silica gel, carbon dioxide on lime, or organic vapours on activated carbon [17].

Chemical reaction, absorption and adsorption form key parts in removing gases. In all gases, one or more of these are used, one at a time. When it comes to the removal of particles and droplets, however, it could be more complex, as several mechanisms must often be used in combination. Basic mechanisms that can be used include:

• Gravity separation. • Centrifugal separation. • Inertial impaction - filtration (micron sizes). • Direct interception - micron. • Brownian diffusion - filtration (sub-micron sizes). • Eddy diffusion.

• Thermal precipitation. • Electrostatic precipitation. • Magnetic precipitation. • Brownian agglomeration. • Sonic agglomeration. • Turbulent deposition.

In most cleaning systems, more than one of the above-mentioned mechanisms are used to clean the off-gas. Before considering the method used, it is important first to gather information on the gas and the impurities. This information includes gas-flow rate, gas temperature, gas composition, materials to be removed and the degree of removal required. As mentioned, the main reason for a GCP is to remove certain unwanted gases or particles in the gas mixture. Particles can be removed using wet scrubbers such as the venturi scrubber, jet scrubber and a disintegrator scrubber [16].

2.4 Calorific value

In order to determine the energy available in a fuel, such as furnace off-gas, the heating value (lower, higher and gross) or calorific value of the fuel needs to be known. In this dissertation the lower heating values of the various gases that are present in the off-gas produced are used to calculate the CV. Table 2.4 shows the heating values and the molecular mass of various fuels.

Table 2.4: Fuel properties [18]

Fuel Molecular weight HHV (kJ/kg) LHV (kJ/kg) Gasoline C8H15 111 47300 43000 Light diesel C12.3H22.2 170 44800 42500 Heavy diesel C14.6H24.8 200 43800 41400 Methanol CH3OH 32 22540 20050 Ethanol C2H5OH 46 29710 26950 Methane CH4 16 55260 49770 Propane C3H8 44 50180 46190 Carbon monoxide CO 28 10100 10100 Coal (carbon) C 12 33800 33800 Hydrogen H2 2 141800 120000

The CV of FeCr furnace off-gas can be calculated using the off-gas composition and the heating values of the various gases present, mainly CO and H2.

(32)

2.5. ENGINES CHAPTER 2. LITERATURE STUDY

2.5 Engines

In the previous sections, the basic concepts of how and where furnace off-gas is produced were discussed. The gas composition, fuel energy content and reasons for various gas-cleaning techniques were also considered. In the following sections, some of the available systems and technologies for utilizing the furnace waste energy (off-gas) are considered. The various types of a particular technology, as well as the systems in which they are used, are considered.

In this section the most common prime mover, the combustion engine, is discussed. There are many different types of engines, each classified by its different applications (truck, locomotive, power generation, etc.), design, fuel used and method of cooling (air, water or uncooled). In this section, a few of these engines are briefly discussed in order to get a basic understanding of how they operate.

2.5.1 Internal combustion engine

Internal combustion engines have been developed for years. Nicolaus Otto developed the spark-ignition engine in 1876 and Rudolf Diesel the compression ignition engine in 1892 [19]. An internal combustion engine converts the chemical energy of a fuel source into mechanical energy. This is done by converting the chemical energy from the fuel into thermal energy by means of combustion or oxidation of the air and the fuel source used. As the thermal energy (temperature) increases, the pressure of the gases inside the engine increases, which causes the gases to expand and force the mechanical parts of the engine to move, creating rotating mechanical energy.

Internal combustion engines are classified by their different applications (locomotive, truck, power generation, etc.), design (reciprocating, rotary, etc.), fuel used (diesel, petrol, natural gas (NG), etc.), working cycle (four-stroke or two-stroke), method of ignition (compression or spark), position and number of cylinders (in-line, V, W, opposed piston, radial, etc.), method of cooling (air, water or uncooled), valve location (in head or block), air intake process (naturally aspirated, supercharged, crankcase compressed and turbo-charged) and the method of fuel input (carbureted, multipoint port or throttle body fuel injection) [20]. Spark ignition and compressed ignition engines are the most common engine types.

Most internal combustion engines, spark ignition and compression ignition, operate on either a two-stroke or four-stroke cycle. For most engines, these cycles are fairly similar, with slight variations for specific designs. Most automobile engines operate on the four-stroke cycle even though the two-four-stroke cycle has greater power output per unit weight. This is because most two-stroke engines struggle to pass modern emission standards. Spark ignition

In a spark ignition engine, a premixed stoichiometric (one-to-one ratio) fuel-air mixture is compressed at high pressure and then ignited using a spark plug. This ignition occurs at an optimal time in the stroke cycle of the engine. The fuel-air mixture is distributed and prepared through carburettors and fuel injectors. The power output of this engine is

(33)

controlled by regulating the air intake and fuel injection into the engine, usually referred to as throttling.

Most spark ignition engines use petrol as a fuel source and operate on two- and four-stroke cycles. In figure 2.8a, a basic four-stroke spark ignition design is shown. An engine design that also makes use of spark ignition is the Wankel engine, which is a rotary engine seen in figure 2.7.

Figure 2.7: Wankel combustion engine - operating cycle [21]

The operating cycle of a Wankel spark ignition engine follows the four-stroke cycle of a piston engine [21]. Three working chambers are formed in this engine and provide one power stroke each time the crankshaft rotates 360 degrees. The auxiliary equipment, such as the fuel system and ignition system, used in this engine is the same found in a basic piston engine. A Wankel engine has some advantages, including the small size, low weight and simpler design of the engine. It also has no valves, low friction and no reciprocating unbalance [21], but tends to produce high hydrocarbon emissions due to the nature of the sealing grids of the engine.

(a) Spark ignition (b) Compression ignition

Figure 2.8: Spark and compression ignition internal combustion engines (four-stroke) [22]

(34)

Compression ignition

Compression ignition engines differ from spark ignition engines in many ways, but mainly in the way the fuel-air mixture is ignited. In a compression ignition engine, the mixture is ignited because of the high temperatures that occur in the combustion chamber, due to the high compression. The chemical reaction and thus the combustion occur when the fuel and air are compressed to a critical pressure and temperature [23]. The fuel is injected in the combustion chamber using fuel injectors. In a compression ignition engine, the injection of fuel can be either direct or indirect. Direct injection engines inject fuel directly into the compressed, air-filled cylinder and an indirect injection engine makes use of a pre-combustion chamber where the air and fuel are premixed before injection. The main reason for the use of a pre-combustion chamber is to speed up the combustion process, increasing the engine output by increasing the engine speed [24]. In figure 2.8b, the pre-combustion chamber can be seen in the design for the cylinder of a compressed ignition four-stroke diesel engine.

The most commonly used internal combustion engine is the diesel engine. In diesel engines electrical components aid in the speed of the combustion process by using an electrical source (battery) to help with heating the mixture [24]. This pre-heating of the mixture helps with cold starting of diesel engines. A diesel engine has high thermal efficiency that leads to low fuel consumption, but has a low power output, relative to the weight of the engine, when compared to spark ignition engines.

2.5.2 External combustion engine

In an external combustion engine the energy used inside the engine can be generated outside the engine and is not generated inside the cylinder of the engine, as in an internal combustion engine. In this section, two external combustion engines are considered: the steam engine and the Stirling engine. Both engines use thermal energy and pressure to move the piston(s) inside the displacer cylinder of the engine.

Steam engines

The reciprocating steam engine played a major part in the early industrial era, when it was mainly used to power mills and steam locomotives [25]. A steam engine uses the expansion power of steam to generate power. Water is heated by burning a fuel source, such as coal, until the water is vaporized and steam is generated. The steam is supplied to a piston inside a cylinder via a valve, which drives the piston. This linear movement of the piston is then used to turn a fly-wheel, which generates the rotational power of the engine.

Figure 2.9 shows the construction and operation of a basic cylinder design of a steam engine. In this figure, the double-action steam engine is shown. In a double-action steam engine the high-pressure steam first enters on the one side of the piston, pushing it to one side, after which the steam is released through the exhaust port and then steam enters from the other side of the piston, pushing it to the opposite side. The pushing and pulling of the piston by the entry of steam is known as compression and expansion. The piston is connected to the piston rod, which is connected to a fly-wheel (or crank pin) through a chain of links. This connection converts the linear mechanical energy into rotational

(35)

energy that is transmitted to the crankshaft.

Figure 2.9: Basic steam engine cylinder operation [26]

Various steam engines also have condensers, which cools the exhaust steam until the vapour returns to the liquid form. This water, which is still warm, then moves to the boiler to be heated again. This closed system is more efficient, because the water that is vaporized is already at a high temperature when it reaches the boiler and less fuel needs to be burned to vaporize the liquid again.

Over the years, steam engines were replaced by internal combustion engines, but steam turbines were developed and are widely used in the industrial sector. These turbines are discussed in a following section.

Stirling engines

The Stirling engine was invented in 1816 by Robert Stirling as a competitor for the steam engine, but had a very slow start and never really succeeded for various reasons [27]. In recent years, this engine gained more favour because of its high efficiency and the ability of the engine to work with heat from various sources.

The Stirling engine is a closed system engine that works with a constant volume of compressible fluids, such as air, hydrogen and helium, which are heated and cooled. Its operation is based on the Stirling cycle where the working gas is cooled and heated in certain parts of the system in order to generate the compression and expansion of the gas. The compression and expansion process takes place in a cylinder where mechanical movement of a power piston is generated. A displacer piston, which has a smaller

(36)

diameter than the cylinder, is used to move the working gas back and forth through the whole Stirling cycle.

The main thermal components of this system, seen in figure 2.10, are the heater (increases thermal energy of the working fluid - expansion), cooler (decreases thermal energy of working fluid - compression) and the regenerator, which stores the heat of the working fluid before it is cooled. The regenerator causes higher thermal efficiency of the system, as the waste heat inside the engine is stored and re-used.

Figure 2.10: Basic configurations of the Stirling engine [28]

Three mechanical configurations of the Stirling engine are shown in figure 2.10, namely the alpha, beta and gamma configurations. Each configuration differs mechanically, but operates on the same thermodynamic cycle. In the alpha-configuration, no displacer is found and the cooled and heated versions of the gas drive two separate pistons. The beta-configuration makes use of a displacer and a power piston in a single cylinder, whereas in the gamma-configuration the displacer and power piston are used in separate cylinders. Of all the configurations, the gamma-configuration has the highest theoretical mechanical efficiency [28].

A Stirling engine is theoretically an engine with very high efficiency, when converting thermal energy into mechanical energy. It can have efficiencies ranging between 30% and 40% resulting from temperatures in the range of 650oC to 800oC and normal operating speeds of between 2000 rpm and 4000 rpm [28]. Because of its simple design the Stirling engine can be used as an inexpensive power source for electrical energy generation, using a variety of fuels, such as biomass [29].

2.5.3 Gas engine

Gas engines use gaseous fuel, such as coal gas, producer gas, biogas, landfill gas and NG. Most gas engines are spark-ignition internal combustion reciprocating engines using gaseous fuel as primary fuel source [30]. In this section a producer gas fuelled gas engine is considered and briefly discussed.

(37)

Over the past few years, gas engines have evolved dramatically and are able to use a wide range of fuels. Major global manufacturers of these engines (GE-Jenbacher, Cummins and Wärtsilä) have been working hard to improve their machines’ efficiency and to reduce emissions [31]. There are companies such as GE that have manufactured machines like the GE Jenbacher engine, which operates on furnace off-gas from steel, ferrochrome, ferromanganese and calcium carbide smelter processes to generate electrical energy [30]. The Jenbacher engine is discussed in more detail in the following subsection.

Many spark ignition engines (petrol) and diesel engines run on producer gas. Spark ignition engines can be run on producer gas alone and compression ignition engines can be modified to run on producer gas by lowering the compression ratio of the engine and installing a spark ignition system [32]. Another option is to use an unconverted diesel engine to operate in dual-fuel mode, where diesel is used for ignition of the gas-air mixture and the remaining power output (up to 90%) can be drawn from the producer gas. The latter option can then operate on diesel alone or in dual-fuel mode, which makes it a flexible option. However, not all types of diesel engines can be converted for dual-fuel operation because of the high compression ratios of certain designs (ante-chamber and turbulence chamber), but direct injection diesel engines can generally be converted because of lower compression ratios [32].

Jenbacher engine

The GE Jenbacher engine is a gas engine that can use the off-gas from a furnace for electrical energy generation. These engines can also run on landfill and natural gas and have already been installed in many parts of South Africa. In most cases these engines are installed for electrical energy generation.

(38)

In figure 2.11 the basic components of a stationary gas engine and generator system are displayed. The engine can be fuelled by different gases and burns the gas in the cylinders of the engine, which forces the crankshaft to turn. A generator is turned by the crankshaft, which results in the generation of electrical energy. The thermal energy, generated during the combustion process, is released from the cylinders, where it can either be recovered using combined heat and power (CHP) configurations or dissipated via dump radiators. An advanced control system is installed to control and regulate the performance of the generator. A LEANOX lean burn combustion control system in the GE Jenbacher controls the air/fuel mixture under all operating conditions and contributes to the stable operation of the engine [33].

GE Jenbacher gas engines are available in an electric output range of between 0.3 MW to 9.5 MW and have efficiencies of up to 47.8%. The pie charts in figure 2.12 show the energy efficiency of the GE Jenbacher engine. From the gas energy supplied, 42% is converted to mechanical energy, of which 40% is converted to electrical energy. The remaining 58% is thermal energy, of which 50% is usable thermal energy [33].

(a) Energy conversion from 100% gas supply

(b) Mechanical energy used (c) Thermal energy used

Figure 2.12: Gas engine energy balance [33]

The efficiency of the Jenbacher engine can be increased by incorporating combined heat and power and combined heat, power and cooling (CHPC) systems. These systems use the thermal energy generated in the gas engine and are considered later in this chapter.

(39)

2.5.4 Dual-fuel engine

A dual-fuel engine is similar to the basic diesel engine (compression ignition), but with some dual-fuel specific components (see figure 2.13). As suggested in the name of this engine, both gas and diesel are used as fuel for this system. These types of engines can either run in dual-fuel mode, which includes the use of diesel and a gaseous fuel mixture, or can operate on 100% diesel fuel. In this engine the compression ratio, when gaseous fuel is used, is too low to effect ignition at the desired time [34]. For that reason, this engine cannot operate on gaseous fuel alone and thus the diesel fuel is injected as in a normal diesel engine.

When the engine operates in dual-fuel mode the gaseous fuel is introduced in the intake of the engine. This air-to-gas (fuel) mixture is then drawn into the cylinder, as in an internal combustion engine, but with a leaner air-to-fuel ratio. As in a diesel engine, the diesel is injected near the end of the compression stroke of the cylinder. The diesel then ignites, causing the gaseous fuel to burn.

The pilot liquid fuel (diesel) only contributes to a small fraction of the power output of the engine. The quantity of pilot diesel that is used is usually fixed for a given engine and at full load the volume used represents less than 10% of the total energy supplied to the engine [35]. In most dual-fuel engines the engine output is only dependent on the volume of gaseous fuel added to the air during the induction stroke [34]. The Cummins dual-fuel engine has achieved maximum substitution rates. The fraction of the total fuel energy provided by the gaseous fuel is around 70%, with high load factors [36].

Figure 2.13: Dual-fuel operation [36]

Dual-fuel engines could be a possible substitute for the basic diesel engine, if gaseous fuel is available. The reason for this is the low specific fuel consumption, high thermal efficiency, possibly lower fuel costs, no pre-ignition, low CO and moderate hydrocarbon emissions at low and moderate loads without exhaust treatment, suitability for two-stroke operation and durability [21]. Lower compression ignition engines, such as dual-fuel engines, have a simpler construction, are lightweight and have lower initial costs, lower operating expenses and higher mechanical efficiency than higher-compression engines.

Utilizing waste energy from a submerged arc furnace at a ferrochrome smelter plant