ITP Aluminum: Energy Requirements for the U.S. Aluminum Industry

U.S. Energy Requirements for Aluminum Production

Historical Perspective, Theoretical Limits

and Current Practices

Prepared for

Industrial Technologies Program

Energy Efficiency and Renewable Energy

U.S. Department of Energy

February 2007

Preface

This report provides reliable and comprehensive statistical data over the period 1960 to 2003 for the

evaluation of energy trends and issues in the U.S. aluminum industry. It should be noted, however,

that these trends need careful interpretation as they incorporate unusual circumstances of a single

year, i.e., 2001. During the summer of 2001, the extensive heat wave in the western United States

produced an increased demand for electricity. Simultaneously, the ability to generate hydroelectric

power was reduced due to historically low snow packs in the Columbia River basin and new

regulations mandating the spill of water to aid migrating salmon. The combination of high electricity

demand and limited water supply contributed to a significant increase in the market price of

electricity during this time. This price increase in the Pacific Northwest made it more economical for

aluminum smelters to stop metal production and sell back power from their low-cost, fixed-price

electric contracts to aid in minimizing the shortfall in energy supply. As a result, the majority of

aluminum smelting capacity in the Pacific Northwest, representing approximately 43 percent of all

U.S. primary aluminum capacity, shut down.

The dramatic and relatively quick shutdown of a substantial amount of the U.S. primary aluminum

capacity can make certain trend numbers appear misleading. Throughout this report, percentages are

given for 10-year trends in various sectors throughout the aluminum industry. For example, U.S.

primary aluminum production has a 10-year annual growth rate of -2.5%, decreasing from 3,695

thousand metric tons in 1993 to 2,704 thousand metric tons in 2003. These numbers would seem to

imply a steady decline throughout the specified time period, but this is not the case. In fact, the total

drop in primary production over this time period (991 thousand metric tons) is less than the 1,031

thousand metric ton drop from 2000 (3,668 thousand metric tons) to 2001 (2,637 metric tons).

This

one year drop coincides with the majority of the Pacific Northwest shutdown. Other primary

production numbers in this report have similar trends, and this drastic shutdown should be taken into

account when considering these numbers.

It is currently too early to accurately assess the long-term impact of these sudden shutdowns and

changing conditions on the aluminum industry. It remains to be seen whether the shutdowns will lead

to a permanent decline of primary metal production in the Pacific Northwest, or whether the industry

will emerge robustly with additional self-generated power capacity and energy efficiency

improvements. It was announced (October 8, 2004) that one of the shutdown smelters will be

restarted at half capacity (+110,000 metric tons/yr), but it is difficult to judge whether or not this will

be the common trend. Whatever the industry's future, it is clear that the local and global pressures to

increase overall energy efficiency will determine its vitality. The energy efficiency opportunities

discussed in this report are pertinent to the future of the aluminum industry.

Another recent trend to make note of is China's rapid growth in both primary and secondary

aluminum production. While the United States has been slipping in its rank, China has taken

worldwide lead in primary aluminum production, producing 5,450 thousand metric tons of primary

aluminum in 2003. Additionally, China has been buying up much of the excess aluminum scrap

supply. In 2003, the United States exported 568,721 metric tons of scrap and dross, with 43 percent

(244,374 metric tons) going to China.

These trends are manifested in the slight decline of secondary

aluminum production within the United States.

Authors’ Note

A complete accounting of the energy consumed in the production of any product should include the

energy required to produce the fuels and electricity that are used plus the “feedstock energy”

associated with any fuels that are used as materials (e.g., carbon used in anode production). This

document, in many places, reports energy as a set of two numbers, e.g., 1.0 (2.0

) kWh. The first

value represents the energy consumed within a facility (on-site energy consumption), while the value

with "tf" superscript (tacit-feedstock energy) is a measure of energy that includes the energy used to

produce and transmit the energy consumed within a facility and raw material feedstock energy. This

adjustment is very significant in processes that consume electricity.

The U.S. average grid connection requires about 3.01 kWh (10,270 Btu) of fuel energy to deliver 1

kWh (3,412 Btu) of electrical energy (Table D.1, Appendix D). The average U.S. grid is supplied with

approximately 7 percent hydroelectric generation. The aluminum industry is located near low-cost

power sources and many of these are hydroelectric. Even today with the Pacific Northwest shut

down, the average primary aluminum plant connection is 39.4 percent hydroelectric generation. The

average primary aluminum plant grid connection requires 2.24 kWh (7,624 Btu) of fuel energy to

deliver 1 kWh (3,412 Btu) of electrical energy. It should be noted that values reported in this

document use the U.S. average grid connection values. The use of U.S. average grid values results in

a higher reported tacit energy consumption value than the actual tacit value required for primary

aluminum production. The advantage of reporting primary production energy consumption based on

the U.S. average grid is that it allows easier and same basis comparisons to other U.S. manufacturing

industries.

Contents

Executive Summary ................................................................................................... i

1. Introduction................................................................................................................1

1.1 Purpose of Report .....................................................................................................1

1.2 Energy and Environmental Overview.......................................................................2

2. Methodology, Metrics and Benchmarks ....................................................................4

2.1 Theoretical, Practical Minimum and Current Practice Benchmarks ........................ 4

2.2 Tacit, Process, Feedstock and “Secondary” Energies............................................... 5

2.3 Life Cycle Assessment.............................................................................................. 7

2.4 Energy Value Chain Analysis ................................................................................... 8

2.5 Transportation Energy............................................................................................... 8

2.6 Emissions .................................................................................................................. 9

3. Aluminum Production..............................................................................................10

3.1 U.S. Aluminum Supply.......................................................................................... 12

4. Primary Aluminum Raw Materials ..........................................................................14

4.1 Bauxite .................................................................................................................... 14

4.2 Alumina (Al

) ..................................................................................................... 16

4.3 Carbon Anode ......................................................................................................... 20

5. Primary Aluminum Production................................................................................24

5.1 Production, Capacity, and Growth ..........................................................................24

5.2 Historical Hall-Héroult Energy Utilization............................................................. 25

5.3 Theoretical Minimum Energy Requirement for Reduction .................................... 27

5.4 Hall-Héroult Reduction Process ............................................................................. 30

5.5 Environmental Considerations................................................................................38

5.6 Technological Change in the Next Decade ............................................................. 40

6. Advanced Hall-Héroult Cells...................................................................................43

6.1 Wetted Drained Cathode ........................................................................................ 43

6.2 Hall-Héroult Inert Anode........................................................................................47

6.3 Multipolar Cells ...................................................................................................... 52

7. Alternative Primary Aluminum Processes ...............................................................54

7.1 Carbothermic Technology.......................................................................................54

7.2 Kaolinite Reduction Technology ............................................................................58

8. Secondary Aluminum (Recycling)...........................................................................64

8.1 Secondary Aluminum Production...........................................................................65

8.2 Production, Capacity and Growth...........................................................................66

8.3 Recycling Processes................................................................................................67

9. Aluminum Processing ..............................................................................................71

9.1 Melting, Alloying, and Melt Treatment ..................................................................71

9.2 Ingot Casting........................................................................................................... 75

9.3 Rolling..................................................................................................................... 78

9.4 Extrusion................................................................................................................. 81

9.5 Shape Casting.......................................................................................................... 83

9.6 Thermal Treatments ................................................................................................86

Endnotes...................................................................................................................89

Glossary....................................................................................................................94

Appendix A: Summary of Production and Energy Data for the U.S. Aluminum Industry................98

Appendix B: Energy Intensity of Materials Produced in the United States.....................................100

Appendix C: Energy Values for Energy Sources and Materials .....................................................102

Appendix D: Hydroelectric Distribution and Electrical Energy Values .........................................104

Appendix E: Emission Data and Calculations ................................................................................107

Appendix F: U.S. Energy Use by Aluminum Processing Area ......................................................111

Appendix G: U.S. Primary, Secondary and Imported Aluminum Quantities, 1960-2003 ..............117

Appendix H: U.S. Bauxite and Alumina Quantities, 1960-2003 ....................................................119

Appendix I: Energy Requirements for Carbon Anodes .................................................................121

Appendix J: Theoretical Energy Data and Calculations ................................................................122

Appendix K: Alumnum Heat Capacity and Heat of Fusion Data ...................................................128

Appendix L: Impact of Secondary Metal Production on

Energy Requirements for U.S. Aluminum Production.................................................130

Appendix M:Impact of Using Different Technologies on

Energy Requirements for Producing Aluminum.........................................................131

List of Tables

Table A: U.S. Energy Requirements and Potential Savings ................................................................. iii

Table 4.1: Energy Associated with Carbon Anode Manufacturing........................................................ 22

Table 5.1: Typical Parameters of Aluminum Reduction Cells 1948 vs. 1999

................................... 26

Table 5.2: Ore-to-Metal Comparison of Near and Mid-Term Technology Improvements ................... 42

Table 6.1: Energy Consumption Associated with Various Wetted Cathode Arrangements .................. 46

Table 6.2: Energy Impact of Inert Anode Technology........................................................................... 50

Table 7.1: Comparison of Hall-Héroult and Carbothermic Reduction .................................................. 57

Table 7.2: Carbon Dioxide Equivalent Comparison of Hall-Héroult and Carbothermic Reduction ..... 58

Table 7.3: Comparison of Hall-Héroult and Kaolinite Reduction ........................................................ 62

Table 9.1: Primary Ingot Casting Distribution of Energy Consumption ............................................... 77

Table 9.2: Secondary Ingot Casting Distribution of Energy Consumption ........................................... 77

Table 9.3: Cold Rolling Distribution of Energy Consumption ............................................................. 79

Table 9.4: Hot Rolling Distribution of Energy Consumption ............................................................... 79

Table 9.5: Extrusion Distribution of Energy Consumption ................................................................... 82

Table 9.6: Shape Casting Distribution of Energy Consumption ........................................................... 85

Table A.1: Theoretical, Process, and Gross Energy Requirements......................................................... 98

Table A.2: United States Total Energy Requirements and Potential Savings......................................... 99

Table B.1: Energy Requirements to Produce Materials in the United States (2002)............................ 100

Table B.2: Gross Energy Requirements to Produce Materials in the United States (2002) ................. 101

Table C.1: Energy Values for Energy Sources and Materials ............................................................... 102

Table C.2: kWh/yr of Fuels Consumed Worldwide for U.S. Aluminum Processing

Excluding Electricity and Coke Feedstock Energy ............................................................ 103

Table C.3: Impact of Electric Tacit Conversion Factors on kWh/yr

Consumed Worldwide for U.S. Aluminum Production...................................................... 103

Table D.1: Electric Tacit Energy and Emission Data for Fuels Used for Aluminum Producion ......... 104

Table D.2: Energy Sources of Electrical Power in 2002 (electrical power used in gigawatt hours) .... 105

Table D.3: Sources of Supply of Electrical Power in 2002 (electrical power used in gigawatt hours) 105

Table D.4: Average U.S. Grid Connection Tacit Energy ...................................................................... 106

Table E.1: Carbon Dioxide Equivalent Emission Coefficients for Fuels

Associated with Aluminum Production.............................................................................. 107

Table E.2: Carbon Dioxide Equivalent (CDE) Emissions Associated

with Primary Aluminum Production.................................................................................... 108

Table E.3: Carbon Dioxide Equivalent (CDE) Emissions Associated with Aluminum Production .... 109

Table E.4: Carbon Dioxide Equivalent Emissions Associated

with New Aluminum Production Technologies................................................................... 110

Table F.1: Materials and Energy Associated with Primary Metal Electrolysis.....................................111

Table F.2: Materials and Energy Associated with Aluminum Manufacturing Operations ...................112

Table F.3: Theoretical Minimum Energy Requirements to Produce

Raw Materials and Aluminum Products..............................................................................113

Table F.4: U.S. Energy Use in the Production of Domestic Aluminum ...............................................113

Table F.5: Total U.S. Aluminum Industry Energy Consumption and Potential Savings ......................114

Table F.6: Energy Impact of Recycling.................................................................................................114

Table F.7: U.S. Electric On-site Energy Consumption in the Aluminum Industry...............................115

Table F.8: Smelting and Heating Fractions of Total U.S. Aluminum Industry Energy Consumed ......115

Table F.9: Total Worldwide Production and Energy Consumption

Associated with Producing Aluminum in the United States.................................................116

Table F.10: Percent Break-Up of Different Types of Energy Used

in Aluminum Production Operations....................................................................................116

Table G.1: U.S. Supply of Aluminum from 1960 to 2003 ....................................................................117

Table H.1: U.S. Supply of Bauxite and Alumina from 1960 to 2003 ...................................................119

Table I.1: Energy Associated with Aluminum Industry Carbon Anode Manufacturing..................... 121

Table I.2: Onsite and Tacit Energy Associated with Carbon Anode Production ................................ 121

Table J.1: Theoretical Minimum Energy for Hall Héroult Carbon Anode System ............................. 123

Table J.2: Theoretical Minimum Energy for Hall Héroult Inert Anode System ................................. 123

Table J.3: Theoretical Minimum Energy for Carbothermic Reduction............................................... 124

Table J.4: Theoretical Minimum Energy for Reduction of Aluminum Chloride ................................ 124

Table J.5: Theoretical Minimum Energy for Kaolinite / Aluminum Chloride Reduction................... 125

Table J.6: Thermochemistry Data for Elements and Compounds

Associated witb Aluminum Production................................................................................126

Table J.7: Changes in Heat of Formation Values as a Function of Temperature................................. 126

Table J.8: Theoretical Minimum Energy for Gibbsite Dehydration.................................................... 127

Table J.9: Heat of Capacity Equations for Gases Associated with Aluminum Production................. 127

Table K.1: Aluminum Energy Requirements for Heating and Melting................................................ 129

Table L.1: Impact of Secondary Metal Production on the Nominal

Energy Requirements to Produce Aluminum......................................................................130

Table M.1: Impact of Wetted Cathode Technology on Primary Metal Electrolysis............................. 131

Table M.2: Impact of Inert Anode and Wetted Cathode Technology on Primary Metal Electrolysis. 132

Table M.3: Estimate of Energy Requirement for Manufacturing an Inert Anode................................. 133

Table M.4: Estimate of Energy Requirement to Manufacture

Aluminum Using Different Technologies.............................................................................134

List of Figures

Figure A: Aluminum Industry Flow Diagram ........................................................................... i

Figure 2.1: Boundaries for Life Cycle and Value Chain Assessments ........................................ 8

Figure 3.1: Energy

Consumption of U.S. Aluminum Operations ............................................ 11

Figure 3.2: U.S. Aluminum Supply from 1960 to 2003 ............................................................. 12

Figure 4.1: U.S. Bauxite Supply 1960 to 2003 ........................................................................... 15

Figure 4.2: U.S. Alumina Supply 1960 to 2003 ......................................................................... 17

Figure 5.1: U.S. Production of Primary Aluminum from 1960 to 2003 ..................................... 24

Figure 5.2: Primary Aluminum Electric Energy Consumption 1900 to 2000 ............................ 25

Figure 5.3: Primary Aluminum Current Efficiency 1900 to 2000 ............................................. 26

Figure 5.4: Alumina to Aluminum Theoretical Minimum Energy ............................................ 27

Figure 5.5: Alumina and Carbon to Aluminum Theoretical Minimum Energy ......................... 29

Figure 5.6: A Typical Hall-Héroult Cell ..................................................................................... 30

Figure 5.7: Voltage Distribution in a Hall-Héroult Cell ............................................................. 32

Figure 5.8: Cryolite-Alumina Phase Diagram ............................................................................ 37

Figure 6.1: Conceptual Wetted Cathode Cells ............................................................................ 44

Figure 6.2: Theoretical Energy Schematic for Inert Anode ....................................................... 48

Figure 6.3: Inert Anode .............................................................................................................. 51

Figure 6.4: Multipolar Cells ....................................................................................................... 52

Figure 7.1: Carbothermic Reactor .............................................................................................. 55

Figure 7.2: Carbothermic Theoretical Minimum Energy ........................................................... 56

Figure 7.3: Clay to Aluminum Process Schematic ..................................................................... 59

Figure 8.1: U.S. Aluminum Market and Growth ........................................................................ 65

Figure 8.2: U.S. Production of Secondary Aluminum 1960 to 2003 ......................................... 67

Figure 9.1: Typical Rolling Mill Processing Operations ............................................................ 78

Figure 9.2: Typical Aluminum Extrusion Processing Operations .............................................. 81

Figure 9.3: Typical Aluminum Product Shape Casting Operations ........................................... 84

Executive Summary

Figure A: Aluminum Industry Flow Diagram

Executive Summary

The United States aluminum industry is the world’s largest, processing 9.6 million metric tons of

metal and producing about $40 billion in products and exports in 2003. It operates more than 400

plants in 41 states and employs more than 145,000 people. Aluminum impacts every community and

person in the country, through either its use and recycling or the economic benefits of manufacturing

facilities.

Energy reduction in the U.S. aluminum industry is the result of technical progress and the growth of

recycling. These two factors have contributed 22 percent and 39 percent respectively to the total 61

percent energy reduction over the past forty years. By many measures, aluminum remains one of the

most energy-intensive materials to produce. Only paper, gasoline, steel, and ethylene manufacturing

consume more total energy in the United States than aluminum. Aluminum production is the largest

consumer of energy on a per-weight basis and is the largest electric energy consumer of all

manufactured products. The U.S. aluminum industry directly consumes 45.7 x 10

kilowatt hours

(0.16 quad) of electricity annually or 1.2 percent of all the electricity consumed by the residential,

commercial, and industrial sectors of the U.S. economy. This is equivalent to the electricity consumed

by 5,222,000 U.S. households annually.

The aluminum industry has large opportunities to further reduce its energy intensity. The annual sum

of all the energy required in the production of aluminum metal and products in the United States is

equivalent to 183 x 10

kilowatt hours (0.62 quad). The difference between the gross annual energy

required and the theoretical minimum requirement amounts to over 149 x 10

kilowatt hours (0.51

quad). This difference is a measure of the theoretical potential opportunity for reducing energy

consumption in the industry, although achievable cost-effective savings are smaller.

U.S. Energy Requirements for Aluminum Production, Historical Perspective, Theoretical Limits and

New Opportunities provides energy performance benchmarks for evaluating new process

developments, tracking progress toward performance targets, and facilitating comparisons of energy

use. The report provides a basic description of the processes and equipment involved, their

interrelationship, and their effects on the energy consumed and environmental impact of

manufacturing aluminum and aluminum products. This knowledge can help identify and understand

process areas where significant energy reductions and environmental impact improvements can be

made.

This report examines and carefully distinguishes between the actual “onsite” energy consumption

values and gross or “tacit” energy values. The “tacit” or gross energy value accounts for the

generation and transmission energy losses associated with electricity production, the “feedstock”

energy of fuels used as materials, and the “process energy” used to produce fuels. Onsite energy

improvements provide concomitant gross energy savings.

Executive Summary

Primary aluminum is produced globally by mining bauxite ore, refining the ore to alumina, and

combining the alumina and carbon in an electrolytic cell to produce aluminum metal. Secondary

aluminum is produced globally from recycled aluminum scrap. Primary and secondary aluminum

metal are cast into large ingots, billets, T-bar, slab or strip and then rolled, extruded, shape-cast, or

otherwise formed into the components and useful products we use daily. Figure A (page i) shows the

major processing operations required to produce aluminum and aluminum products. This report

examines these processes and the energy they require.

Identifying Energy Reduction Opportunities

Energy performance benchmarks, current practice, and theoretical minimums provide the basis for

evaluating energy reduction opportunities. These benchmarks and gross energy consumed during

aluminum production in the United States are summarized in Table A.

Table A: U.S. Energy Requirements and Potential Savings

U.S.

Theoretical U.S. Potential

Total U.S. Potential

Annual

Minimum Process Process U.S.

Gross Gross U.S.

Production

Energy Energy Energy

Energy

2003

Requirement Required Savings

Required Savings

metric tons

kWh (10

) / yr

(quad)

kWh (10

) / yr

(quad)

kWh (10

) / yr

(quad)

kWh (10

) / yr

(quad)

kWh (10

) / yr

(quad)

Bauxite Mining

Alumina Refining 2,661,500 0.37 (0.001) 10.02 (0.034) 9.65 (0.033) 10.89 (0.037) 10.52 (0.036)

Anode Production 1,230,000 12.12 (0.041) 15.75 (0.054) 3.63 (0.012) 16.45 (0.056) 4.33 (0.015)

Al Smelting 2,758,000 16.52 (0.056) 42.97 (0.147) 26.46 (0.090) 128.36 (0.438) 111.84 (0.382)

Primary Casting 2,704,000 0.90 (0.003) 2.73 (0.009) 1.83 (0.006) 3.94 (0.013) 3.04 (0.010)

Secondary Casting 2,820,000 0.94 (0.003) 7.05 (0.024) 6.11 (0.021) 7.93 (0.027) 6.99 (0.024)

Rolling 4,842,600 1.55 (0.005) 3.04 (0.010) 1.49 (0.005) 6.08 (0.021) 4.53 (0.015)

Extrusion 1,826,000 0.80 (0.003) 2.37 (0.008) 1.57 (0.005) 2.77 (0.009) 1.97 (0.007)

Shape Casting 2,413,000 0.80 (0.003) 6.17 (0.021) 5.36 (0.018) 6.37 (0.022) 5.56 (0.019)

Total 34.00 (0.116) 90.10 (0.307) 56.10 (0.191) 182.77 (0.624) 148.78 (0.508)

Industrial processes that consume energy at significantly higher rates than their theoretical

requirements are, on the surface, obvious targets for potential improvement. However, energy

performance is only one factor in identifying the best opportunities for improving energy efficiency.

Other factors, particularly market dynamics, process economics and forecasting of future demand are

very significant in identifying real opportunities. This report examines the energy performance of the

operations involved in manufacturing aluminum products.

iii

Executive Summary

The amounts of energy used onsite in the major processing operations of the U.S. aluminum industry

are shown in Figure B (page v). The bottom band on each bar shows the theoretical energy

requirement, while the top band of each bar shows the energy used above the theoretical minimum.

The size of the top band is an indication of how large the opportunity is for energy reduction in that

process step.

Smelting requires 46 percent of the total energy consumed in U.S. manufacturing of aluminum. This

process is the largest consumer of energy and the most technically complex operation. Smelting

requires more than twice its theoretical minimum energy and has the potential for the greatest energy

reduction of all operations. Electricity is required for smelting and accounts for over 98 percent of the

energy used in the process. Current research and development (R&D) efforts to advance existing

technology and to develop alternatives to the existing smelting process have the potential to lower

smelting energy consumption by more than 30 percent.

Process heating accounts for 27 percent of the total energy consumed in U.S. manufacturing of

aluminum. Process heating is required for holding, melting, purifying, alloying, and heat treating. It is

utilized in nearly all aluminum production operations. Heating is the second largest energy

consuming operation.

Recycled aluminum now accounts for over half of all U.S. produced aluminum. It requires less than 6

percent of the energy to produce aluminum from mined bauxite and provides significant

environmental benefits. R&D efforts that improve the ability to recycle aluminum offer some of the

greatest opportunities for energy reduction in the industry, since recycling displaces aluminum

produced by smelting.

The magnitude of the top bands of the energy bars in Figure B shows that large opportunities exist for

lowering energy consumption in the industry. The Aluminum Industry Vision, Sustainable Solutions

for a Dynamic World published by the Aluminum Association in 2001 recognizes these opportunities

and sets industry goals for achieving further energy reduction. In Hall-Héroult smelting technology,

the most energy-intensive process, the industry has set a target for reducing electrical energy usage

from 15.4 kWh/kg to 11 kWh/kg of aluminum produced by the year 2020, a 27 percent reduction

from 2000 practices. The 2003 U.S. average of 15.0 kWh/kg shows a slight improvement from 2000,

likely due to the ongoing upgrades of larger, older primary production plants.

kWh(10

)/yr

Bauxite

Mining

Al u m i n a

Refining

Anode

Production

Aluminum

Smelting

Casting Rolling Extrusion Shape

Casting

energy used above theoretical

minimum

theoretical minimum energy

required

Executive Summary

Figure B: Process Energy Used in U.S. Manufacturing of Aluminum Products

kWh(10

)/yr

energy used above theoretical

minimum

theoretical minimum energy

required

Bauxite Al u m i n a Anode Aluminum Casting Rolling Extrusion Shape

Mining Refining Production Smelting Casting

Evaluation of the many opportunities that exist for reducing energy consumption in the industry can

only be made by comparing processes using consistent system boundaries and measures. This report

provides data and information necessary for the reader to understand opportunities for energy savings

in the aluminum industry.

1. Introduction

Aluminum is an essential material for modern manufacturing. It is a lightweight, high-strength,

corrosion-resistant metal with high electrical and thermal conductivity, and it is easy to recycle. The

U.S. aluminum industry is the largest in the world in terms of consumption. The U.S. aluminum

industry utilized 9,592,000 metric tons of metal in 2003 to produce an enormous variety of products.

U.S. per capita consumption was 29.5 kilograms. The industry operated more than 400 plants in 41

states and employed over 145,000 people to make aluminum products.

These products are shipped to

thousands of businesses in the United States from which they are distributed or are incorporated into

other products. Aluminum, per unit mass, is the most energy-intensive material produced in large

quantities in the United States. Only paper, gasoline, steel and ethylene manufacturing consume more

total energy for manufacturing in the United States than aluminum (Appendix B).

Research and development (R&D) efforts to reduce energy consumption are important, since energy

consumption correlates to manufacturing economics, environmental impact and United States

dependence on imported energy sources. Identifying process areas where opportunities for energy use

reduction exist and applying resources to capture these opportunities will benefit the industry and the

nation.

Aluminum manufacturing is energy intensive and roughly one-third of the cost to produce aluminum

from ore is associated with the use of energy and environmental compliance. The aluminum industry,

in the past forty years, reduced its overall energy intensity by nearly 61 percent (Appendix L).

However, even with the large reduction in energy intensity, the industry consumes nearly three times

the theoretical minimum energy required. Significant opportunities for further energy improvements

still remain.

1.1 Purpose of Report

The energy consumption and environmental effects associated with product manufacturing and use

are important measures of the product’s impact on society. Energy consumption and environmental

impact measures are becoming key decision tools for consumers and corporations when choosing a

product. In the near future, manufactured products will compete not only on price and performance,

but also on their impact on society.

The purpose of this report, U.S. Energy Requirements for Aluminum Production, Historical

Perspective, Theoretical Limits and New Opportunities, is:

• to provide an understanding of the processes involved, the energy consumed and the

environmental impact of manufacturing aluminum and aluminum products;

• to provide a common set of terms, benchmarks and values for comparing processes and issues

related to the aluminum industry;

Introduction

• to identify process areas in which significant energy reductions and environmental impact

improvements could be made;

• to strengthen public and work force awareness, education and training (identified as an industry

goal in Aluminum Industry Vision

This report focuses on the most energy-intensive manufacturing operations for aluminum, electrolysis

(smelting) and process heating operations. These two operations account for over 69 percent of the

energy used by the industry (Appendix F, Table F.8). There is a large difference between the

theoretical minimum energy requirements and current practice energy values in electrolysis and

melting. The magnitude of the energy consumed and the difference between current practice and

theoretical energy levels means improvement in electrolysis and process heating will have the largest

impact on the performance of the industry. This report documents existing operations and explores

potential new technology opportunities.

The science and technologies associated with the production of aluminum and aluminum products are

complex. This report attempts to provide the reader with a basic understanding of the science,

technology and energy usage of the aluminum industry. More detailed books

3, 4

are available for the

reader who requires further in-depth study of the subject.

1.2 Energy and Environmental Overview

Technologies, practices and product use determine the energy consumption and environmental impact

of aluminum. Many of the current technologies and practices used to produce aluminum metal and

aluminum products are mature. New technologies and practices are being proposed and studied to

improve aluminum manufacturing from an energy and environmental standpoint. The history and

explanation of current state-of-the-art technologies and practices are presented so the reader can

appreciate the values and benefits that new technologies or practices might bring to the aluminum

industry. Current U.S. production levels, historical production levels and projected growth rates of

aluminum are presented. These production values are needed to measure the magnitude of the impact

of a change in technology or practice. The energy and environmental impacts from the use of

aluminum products are generally low and in some applications may be significantly better than the

impacts of alternative materials. As significant as these impacts on the use of aluminum products are,

they are beyond the scope of this report.

The greatest impact on the future energy intensity of aluminum has been the structural change in the

industry itself. More than 51 percent of the aluminum produced by U.S. industry in 2003 came from

recycled material. In 1960, recycled material was used to generate less than 18 percent of U.S.

produced aluminum (Appendix G). Recovering aluminum from wastes and scraps requires less than 6

percent of the energy of aluminum production from bauxite mining (Appendix F, Table F.6). This

report examines how “urban mining” (recycling) will continue to change the structure of the

Introduction

aluminum industry and continue to lower the overall energy associated with aluminum production.

Recycling is the largest contributor to the reduction of the energy intensity of U.S. produced

aluminum.

Aluminum is an “energy bank” in that nearly all of the original energy stored in the metal can be

recovered again and again every time the product is recycled. Small fractions of the recycled metal

are lost to oxidation (melt loss) and entrapment in purifying fluxes (dross) during the recycling

process. Aluminum can be recycled indefinitely, allowing this saved energy to be collected again and

again.

Greenhouse gas (GHG) emission reduction is a key environmental and sustainability issue for the

twenty-first century. Energy-intensive manufactured materials (such as aluminum) could be

significantly affected both in terms of price and use by GHG emission-reduction policies. However,

contrary to common belief, aluminum production could be positively affected by GHG emission

reduction policies. A combination of emission mitigation in production and significant GHG

emission reduction further down the product chain enhance the attractiveness of aluminum for end-

use applications.

Additional energy and environmental savings can be achieved in the aluminum product chain through

the introduction of new alloys and improved (light weight) product design. These options will not be

considered in this study, but their potential is at least of the same order of magnitude as changes to

production practices and processes.

2. Methodology, Metrics and Benchmarks

There are a variety of metrics, measurements, benchmarks, boundaries, systems and units that are

used differently by various analytical groups. These variations can cause confusion when comparing

values stated in one report to those in another. Two commonly confused values are the relationship

between onsite and tacit energy values, and between U.S. energy requirements and worldwide energy

requirements. Onsite energy values are based on physical measurements. Tacit energy values have

assumptions associated with them. These assumptions can create large differences in the reported

values. The onsite and tacit values used in this report are explained in Section 2.2. The United States

does not mine ore for aluminum production, but refines roughly half of the ore required domestically.

This report focuses on energy consumption within the United States. The total energy associated with

production of metal from ore is an important value and is reported as the “worldwide” energy

requirement in this report.

2.1 Theoretical, Practical Minimum and Current Practice Benchmarks

When examining industrial processes, two metric values for energy requirement are obtainable with

little debate: the process theoretical minimum value and the current practice value. The theoretical

minimum energy requirement for chemically transforming a material is based on the net chemical

reaction used to manufacture the product. In the case of aluminum made from alumina

(2Al

4Al + 3O

)

, the theoretical minimum energy is 9.03 kWh/kg of aluminum

produced (Appendix J). This minimum value is simplistic and represents the thermodynamically ideal

energy consumption. It requires any reaction to proceed infinitely slowly. The theoretical minimum

energy to transform a material from one shape to another shape is based on the mechanical properties

of the material. It is also an idealized value. Neither chemical nor mechanical theoretical minimums

can be realized in practice; however, these provide the benchmarks that no process can surpass

(Analogy: The theoretical minimum score for a full round of golf is 18).

The current practice value is the average of the actual measurements of existing processes and

practices (Analogy: The current practice value for golf is the average score of every player, which is

well above par). The boundaries drawn around the process or practice, the number of samples,

sampling techniques, etc., determine the precision and accuracy of this value. The difference between

the theoretical minimum and current practice metric is a valuable measure of the opportunities for

energy efficiency improvement in that process or practice.

Practical minimum energy is a term in common usage. However, its definition varies. In some

instances, it is used to describe the process energy value that represents the combination of integrated

unit operations using best available technology and best energy management practices. In other

instances, practical minimum energy is defined as the optimal design value projected with the

adoption of new, advanced technology. Practical minimum energy values are, in reality, a moving

Methodology, Metrics and Benchmarks

target since it is not possible to predict the new technologies, practices and materials that will impact

an industrial process. What is known about the practical minimum energy value is that it lies

somewhere between the current best available value and the theoretical minimum value. (Analogy:

The practical minimum score for golf is some value below par and over 18.)

The “Aluminum Industry Vision”

has selected a goal of 11 kWh/kg of aluminum as its smelting

current practice value for the year 2020. This represents a 27 percent reduction over 1995’s value of

15.4 kWh/kg of aluminum. The industry envisions this as an obtainable and practical minimum

smelting energy goal for 2020.

2.2 Tacit, Process, Feedstock and “Secondary” Energies

Current practice process measurements are actual measurements taken within a facility on existing

operations. These onsite process measurements are valuable because they are the benchmarks that

industry uses to compare performance between facilities and companies. More importantly, these

onsite process measurements are used to assess the value of new processes and practices. These are

the critical values used in the decision-making process to adopt new technologies and practices.

Onsite process measurements, however, do not account for the complete energy and environmental

impact of manufacturing a product. A full accounting of the impact of manufacturing must also

include the energy used to produce the electricity, the fuels and the raw materials used within a

manufacturing facility. These “secondary energy” requirements for electric power generation and

transmission, for the energy needed to produce fuels, and for the energy values of feedstock materials

are very important from a regional, national, and global energy perspective, but they are seldom

analyzed or accounted for within an individual plant site.

The process energy or “secondary energy” associated with the fuels used in aluminum processing is

presented in Appendix C, Table C.1. The process energy adds approximately 3 percent to the energy

values of the fuels used (Appendix C, Table C.2). Feedstock energy represents the energy inherent in

fuels that are taken into a manufacturing process, but used as materials rather than fuels. Aluminum

production uses coke as a raw material in the production of carbon anodes. Coke’s feedstock energy is

significant and is equivalent to a 30 percent increase in the onsite energy consumption of the Hall-

Héroult process (Appendix F, Table F.1). The energy contribution of feedstocks is expressed in terms

of calorific or fuel value plus the “secondary energy” used to produce the feedstock. (Note: fuel and

feedstock tacit energy values used in this report are the calorific fuel value plus the fuel processing

energy, Appendix C, Table C.1).

Tacit energy is a term frequently used to describe the combined total of onsite energy and the

“secondary energy” requirements. Tacit electrical energy and environmental impact measurements

account for the fact that substantial electrical generation inefficiencies and transmission losses occur

outside the facility. It can take as much as four units of hydrocarbon or coal calorific energy to

Methodology, Metrics and Benchmarks

produce one unit of electric energy. Saving 1 kilowatt-hour of onsite electricity is equivalent to saving

nearly 4 kilowatt-hour of the energy contained in the petroleum or coal-based fuels used to generate

electrical power.

Tacit electric conversion factors are variable since they are dependent on the sources of the energy

used to produce electricity. Each manufacturing facility has a different tacit conversion factor

depending on its location. Typical U.S. grid electricity requires about 10,270 Btu of energy to deliver

1 kWh of onsite electricity (3,412 Btu) for use. Electricity production from coal requires 10,388 Btu

to deliver 1 kWh of onsite electricity (3,412 Btu). Water has no fuel value and typically hydroelectric

facilities are assumed to have a tacit energy requirement of 3,412 Btu to deliver 1 kWh of onsite

electricity (3,412 Btu) and near zero greenhouse gas emissions (Appendix D). The onsite and tacit

electric energy requirements for a facility operating on hydroelectric power are equal.

Comparing energy values for the various steps used in the production of aluminum products is

simpler when a common unit is used for all processing steps. Since electricity is the single largest

source of energy consumed in the manufacture of aluminum, the common units of a kilowatt-hour

(kWh) are used in this report. Process energy values for production steps that consume fuels are

converted to kWh using the conversion factor of 3,412 Btu/kWh.

The large variations in tacit electric energy conversion values, 10,388 Btu per onsite kWh for coal

compared to 3,412 Btu per onsite kWh for hydroelectric, have a dramatic influence on the reported

tacit energy profile of an industry. Aluminum smelting energy is 98 percent electric energy. A modern

smelter operating from a hydroelectric utility requires onsite energy of 14.4 kWh/kg of aluminum

produced and tacit energy of 14.4 kWh/kg of aluminum, whereas an identical smelter operating from

a coal-fired utility requires onsite energy of 14.4 kWh/kg of aluminum and tacit energy of 36.0 kWh/

kg of aluminum (Appendix C, Table C.3). The U.S. primary aluminum industry has approximately 40

percent of its capacity connected to hydroelectric facilities.

All values reported in this document use the U.S. average grid connection values (i.e., 10,270 Btu/

kWh). The use of U.S. average grid values results in a higher energy consumption value than the

actual tacit value required for primary aluminum production. The advantage of reporting primary pro-

duction energy consumption based on the U.S. average grid is that it allows easier and same bases

comparisons to other U.S. manufacturing industries. This report, for clarity, distinguishes between the

onsite operating energy values and the secondary energy values that include tacit/feedstock

contributions with the use of a superscript, tf. Any value that includes tacit and/or feedstock

components is denoted with the superscript “tf”, e.g., 1.0

kWh.

Methodology, Metrics and Benchmarks

2.3 Life Cycle Assessment

Life Cycle Assessment (LCA) is recognized as the most complete analysis model of a product’s

impact on energy, environmental, economic and social values. LCA of an industrial product extends

from “cradle-to-grave”, i.e., from material acquisition and production, through manufacturing,

product use and maintenance, and finally, through the end of the product’s life in disposal or

recycling. LCA recognizes the importance of considering energy, economic and environmental

factors not only during the production of a product, but also over the product’s complete life cycle,

including use and disposal. The LCA is particularly useful in ensuring that the benefits derived in one

area do not shift the impact burden to other places within a product’s life cycle.

The LCA “use and maintenance” factor for aluminum varies by end-product and in many applications

is more significant in terms of energy and environmental impact than production. Aluminum, in some

cases, provides LCA “use and maintenance” energy savings that are significantly greater than the

energy used in its production. For example, production of an equal strength, but lighter aluminum

product in the transportation sector saves significant amounts of transportation fuel and provides

substantial reductions in greenhouse gases during the product’s “use” phase when compared to

traditional materials. In 2001, an estimated 2.2 billion gallons of gasoline use and 20 million metric

tons of CO

emissions were reduced due to the use of lightweight aluminum castings in automobiles.

Complete LCA for aluminum products must account for the significant portion of aluminum that, in

the acquisition phase, comes from “urban mining” (recycling). Aluminum’s ability to be easily

recycled is reflected in the fact that over half of the U.S. produced aluminum now originates from

recycled material. Recycling is the best option for disposal of nearly every product made from

aluminum. This makes aluminum a “cradle-to-cradle” LCA product.

Methodology, Metrics and Benchmarks

2.4 Energy Value Chain

Analysis

The energy values studied and presented

are based on an energy “value chain”

analysis. The value chain analysis or

“cradle-to-shipping dock” analysis

provided is an integral part of an LCA. It

provides valuable information and data

values for organizations performing LCA

on aluminum products. Value chain

analyses are similar to LCA; however, they

cover only a portion of a total LCA. Figure

2.1 shows the global boundaries of an LCA

study and the boundaries for this study’s

value chain.

Figure 2.1: Boundaries for Life Cycle and Valu

Chain Assessments

Value chain analysis allows for the capture of the direct energy and feedstock inputs of each

processing step (link) and builds a cumulative value of each product along the chain. This report

looks at a portion of the LCA, the energy value chain from “cradle-to-shipping dock.” It does not

account for the LCA “use and maintenance” phase energy or for “tertiary” energy inputs (i.e., the

energy used to make the equipment or buildings that house the process steps). The “cradle-to-

shipping dock” approach is valuable for providing decision-making analyses within the

manufacturing sphere.

2.5 Transportation Energy

The transportation energy associated with acquiring raw materials and distribution of intermediate

products is important for a full LCA. Transportation energy can account for a significant portion of

the total energy associated with manufacturing a final product. The energy required to transport

mined bauxite to refining operations, alumina to smelting operations, ingots to metal processors, and

scrap from collection to melting is not accounted for in the process energy requirements that are

developed in this report. This report focuses on the energy associated with the processing of raw

materials and the processes employed in aluminum production. The transportation energy associated

with these raw materials and processes is small in relation to the total energy consumed in production.

Methodology, Metrics and Benchmarks

Transportation energy calculations for raw materials that are mined globally are highly variable. They

are a function of the location and multiple modes of transportation, e.g., conveyors, trucks, trains,

ocean freight. Transportation energy requirements were evaluated in the Life Cycle Inventory Report

for the North American Aluminum Industry. Transportation of raw materials accounted for 2 percent

of the total energy associated with primary aluminum production in the United States.

Evaluation of the transportation energy requirements associated with secondary aluminum production

is complicated. Consumer scrap can require considerable transportation energy resulting from

individual consumer drop-off, curbside collection, transfer station collection and the actual

transportation to a secondary processor. Transportation energy, associated with industrial

manufacturing scrap and scrap originating at large automotive and white good scrap processing

centers, is more easily estimated since its boundaries are easier to define. The Life Cycle Inventory

Report for the North American Aluminum Industry estimates transportation energy from these sources

to account for 6 to 8 percent of the total energy associated with the production of secondary aluminum

products.

2.6 Emissions

Energy use and greenhouse gas emissions are closely related. This report provides overall carbon

emission data associated with fuels used for aluminum operations. Other fuel-related emissions (e.g.,

nitrous oxides, sulfur dioxide, volatile organic compounds) are not considered because their

quantities are typically small as compared to the carbon-based emissions. Emissions that are

aluminum process-related (e.g., perfluorocarbons, from cryolite) are reported. Energy and

Environmental Profile of the U.S. Aluminum Industry

7,8

provides detailed emission data for aluminum

operations.

Emission calculations for this report are shown in Appendix E. Greenhouse gases contribute to

climate change by increasing the ability of the atmosphere to trap heat. Gases differ in their ability to

trap heat. To express the greenhouse effect of different gases in a comparable way, atmospheric

scientists use a weighting factor, global warming potential (GWP). The heat-trapping ability of one

metric ton of CO

is the standard, and emissions are expressed in terms of a million metric tons of

equivalent or 10

TCDE. This report uses carbon dioxide equivalents (CDE). Emissions are also

commonly expressed in terms of a million metric tons of carbon equivalent (10

TCE). Carbon

comprises 12/44 of the mass of CO

; to convert from CO

equivalent to C equivalent multiply the

equivalent by 0.273.

3. Aluminum Production

Aluminum metal is classified as primary aluminum if it is produced from ore and as secondary

aluminum if it is produced predominantly from recycled scrap material. Primary aluminum metal

production consists of bauxite mining, refining bauxite to produce alumina, and finally, smelting

alumina to produce aluminum. Secondary aluminum is produced by sorting, melting and treating

scrap aluminum. Primary and secondary aluminum metal are further processed using traditional metal

working technologies-rolling, extrusion, forging, shaping and casting into thousands of products.

Aluminum is the most abundant metallic element in the Earth’s crust. However, it is never found in

natural deposits as a free metal, like copper and gold. Aluminum is typically found as one of several

aluminum oxides or silicates mixed with other minerals and must be processed to be recovered in its

pure form. All commercial primary aluminum is produced from one raw material, bauxite and by one

process, electrolytic reduction. For economic and strategic reasons, the aluminum industry continues

to perform research and development on alternative raw materials (e.g., kaolin clay) and processes

(e.g., chemical reduction). Although these alternatives hold promise for reducing costs, energy

consumption, and environmental impacts, none are near commercialization.

The markets for aluminum industry’s raw materials and products are global. Global primary

aluminum production has been growing at a rate of 3.6 percent annually over the last ten years.

The

U.S. aluminum total supply grew at an annual rate of 0.6 percent over the period of 1993 to 2003

(Appendix G). Aluminum is still in the growth phase of the product cycle. Demand for aluminum is

increasing, mainly due to aluminum substitution for other materials in the transportation sector and

other lightweight applications. Its light weight, corrosion resistance and processing possibilities

coupled with its ease and value for recycling strengthen its position as the material of choice in many

applications. Measured in either mass produced or economic value, aluminum’s use exceeds that of

any other metal except iron. It is important in virtually all segments of worldwide manufacturing.

The global estimate for economically recoverable bauxite reserves is 22,000,000,000 metric tons.

This quantity can address the demands for the next century. Two countries have nearly half of the

world’s identified bauxite resources (Guinea has 25 percent and Australia has 20 percent). Bauxite is

no longer mined in the United States as a commercial feedstock for aluminum production. Domestic

ore, which accounts for less than 1 percent of the U.S. requirement for bauxite, is used in the

production of non-metallurgical products such as abrasives, chemicals, flame retardants, and

refractories.

Alumina is produced by refining bauxite in a wet caustic chemical leaching process (Bayer Process).

Imported bauxite is refined in the United States, the largest importer of bauxite and the second largest

bauxite refiner after Australia. Alumina production is continuing to rise in Australia, Brazil, Jamaica,

Aluminum Production

Surinam, Venezuela and India, all countries with large indigenous bauxite reserves. The trend in

alumina production is towards placing refining capacity near the mineral resources, thereby reducing

transportation energy and costs, and adding more value to exports.

Primary aluminum (aluminum from ore) is globally produced by the Hall-Hèroult process, a method

that involves electrolysis or smelting of alumina. Companies choose their smelting locations where

production conditions are favorable, like the availability of skilled labor, proximity to a consumer

market, and provision for a highly-developed infrastructure and, especially, for low cost and reliable

energy. Hydroelectric power accounts for 50 percent of the energy used worldwide for electrolysis of

aluminum (Appendix D, Table D.2). The bulk of energy use in aluminum production is related to the

electricity required for primary electrolysis. Since energy costs are approximately one third of the

total cost of smelting primary aluminum, smelter production has been moving from sites close to

consumer markets to sites with low electricity costs. Most of the primary aluminum industry

restructuring began in the late 1970s and continues to this day. China and Russia have emerged as

major metal producers; other countries entering the world market include Canada, Australia, Brazil,

Norway and countries in the Persian Gulf area, all areas with low energy costs.

Secondary aluminum is produced from scrap or recycled aluminum. The world’s average share of

secondary aluminum production is roughly one quarter of total aluminum production. The United

States produces over half of its aluminum from recycled aluminum scrap (Appendix G). Aluminum

recycling is concentrated in the countries where the scrap is generated with the exception of Asia,

which imports significant amounts of aluminum scrap (driven by the demand for cast aluminum in

the Asian car industry).

Kwh(10

)/yr

140

120

100

tacit energy used above theoretical

minimum

theoretical minimum energy

Bauxite Alumina Anode Smelting Casting Rolling Extrusion Shape

Mining Refining Production Casting

Figure 3.1: Energy

Consumption of U.S. Aluminum Operations

Primary and secondary

aluminum are used to

manufacture numerous

products ranging from aircraft

components to household and

packaging foils. Each product

requires processing like

heating, melting, alloying and

mechanical working. Figure

3.1 shows the tacit energy

consumption of the major

processes in the aluminum

production chain. Production

of primary aluminum accounts

for 87.6

percent of the energy

Aluminum Production

consumed by the U.S. industry; production of secondary aluminum for 4.3

percent ; rolling for 3.3

percent; extrusion for 1.5

percent; and shape casting for the remaining 3.4

percent (Appendix F,

Table F.4).

Two operations, electrolysis and the heating/melting of aluminum, account for over 73(82

) percent

of energy consumed in aluminum processing (Appendix F, Table F.8). Heating and melting

technologies are used for holding, alloying, and treating metal as well as for recycling. Programs that

improve thermal efficiency of heating and melting while minimizing the formation of aluminum

oxide and/or dross, provide a much larger impact on decreasing industry energy usage than their

energy consumption indicates.

3.1 U.S. Aluminum Supply

The U.S. aluminum supply of 9,592,000 metric tons in 2003 originated from three basic sources:

primary aluminum (domestically produced), secondary aluminum (recycled domestic material), and

aluminum imports. This consisted of 2,704,000 metric tons of primary aluminum, 2,820,000 metric

tons of secondary aluminum and 4,068,000 metric tons of imported aluminum.

From 1993 to 2003,

the annual U.S. growths of these supplies were −2.5 percent, −0.4 percent and 5.1 percent

respectively. Since 1993, the total U.S. supply has risen at an annual rate of about 0.6 percent. Figure

3.2 shows the distribution of these supplies over the past 43 years (Appendix G).

Figure 3.2:U.S. Aluminum Supply from 1960 to 2003

Primary Secondary Imports

Source: Appendix G

The United States is the fourth leading producer of primary aluminum metal in the world. However,

its dominance in the global industry has declined. The U.S. share of world production in 1960

accounted for slightly more than 40 percent of the primary aluminum produced. By 2003, the U.S.

share of world production had decreased to 9.8 percent. U.S. primary production peaked in 1980, and

over the past twenty-three years has been gradually declining. Significant year-to-year variations

occur as a result of U.S. electrical costs and global market changes.

Million Metric Tons

1960 1965 1970 1975 1980 1985 1990 1995 2000

Aluminum Production

Secondary (recycled) aluminum is of growing importance to the U.S. supply. In 1960, only 401,000

metric tons of aluminum were recovered. In 2003, 2,820,000 metric tons of aluminum were

recovered. For the years 1993 through 2003, the secondary production of aluminum has dropped at an

annual rate of −0.4 percent (Appendix G). Recently, the U.S. secondary aluminum growth rate has

been slowing due to a combination of factors. Scrap collection programs are beginning to reach their

maturation stage, and the market growth of scrap sources has slowed. Additionally, the U.S. has now

become a net exporter of scrap and dross. In 1993, the U.S. had net imports of scrap and dross of

98,540 metric tons, while in 2003, the U.S. had net scrap and dross exports of 152,000 metric tons.

The main cause for this change is again China’s radical increase in demand for scrap. In 2003,

244,000 metric tons of the 567,700 metric tons of scrap and dross exported by the U.S. (43%) went to

China.

These trends of decreasing U.S. secondary aluminum production are, however, expected to

change. Use of aluminum in the automotive industry grew at over 5 percent annually between 1993

and 2003. This large and growing supply is now beginning to enter the scrap markets and will spur

new growth in secondary aluminum.

Imported aluminum is the fastest growing source of U.S. supply with an annual rate of 5.1 percent

over the 1993 to 2003 time frame (Appendix G). New primary aluminum facilities are being located

outside the United States, near new sources of low-cost electricity.

4. Primary Aluminum Raw Materials

The total energy associated with producing the raw materials required for aluminum

production from bauxite ore was approximately 8.25(14.23

) kWh/kg of aluminum in 2003.

This accounts for 28 percent of the total energy required to produce primary aluminum metal

and consists of:

• 0.32( 0.34

) kWh per kg aluminum for bauxite mining

• 7.27( 7.87

) kWh per kg aluminum for bauxite refining to alumina, and

• 0.66( 6.02

) kWh per kg aluminum for carbon anode production.

A complete account of the energy requirements and environmental impacts to produce any product

must include the energy requirements and environmental impact associated with the production of the

raw materials used. The raw material energy requirements and environmental impacts associated with

primary aluminum production can be divided into the major operations required to produce it. These

are bauxite mining, bauxite refining and carbon anode manufacturing. Roughly 5,900 kg of earth are

mined to produce 5,100 kg of bauxite, which is refined into 1,930 kg of alumina. The 1,930 kg of

alumina are electrolytically processed with 446 kg of carbon to produce one metric ton (1,000 kg) of

aluminum (Appendix F, Table F.1).

Cryolite and other fluoride salts are used as the electrolytic bath for aluminum production. These

materials are theoretically not consumed in the process or combined as part of the final product.

However, approximately 19 kg of bath material is lost for every metric ton of aluminum produced

(Appendix F, Table F.1). These losses are a result of process upsets and bath drag-out when molten

aluminum is removed from the smelting operation. Since these salts represent only a small portion of

the energy requirement for producing the raw materials required for aluminum production, they are

not addressed in this report.

4.1 Bauxite

Aluminum, never found as a free metal, occurs naturally in the form of hydrated aluminum oxides or

silicates. Since the silicates are mixed with other metals such as sodium, potassium, iron, calcium,

and magnesium and it is chemically difficult and expensive to extract aluminum from them, the

silicates are not a practical source of alumnium. The oxides are, therefore, used for producing

aluminum. The aluminum oxides commonly found as naturally occurring minerals include

• corundum (alumina, Al

)

• böehmite (α-Al

•H

O, a monohydrate containing 85 weight percent alumina)

• diaspore (β-Al

•H

, chemically same as böehmite but with a different crystal structure)

1960

1965

1970

1975

1980

1985

1990

1995

2000

Primary Aluminum Raw Materials

• gibbsite (γ-Al

•3H

O, a trihydrate containing 65.4 weight percent alumina)

Alumina, used for the production of aluminum, is obtained from bauxite deposits. Bauxite is not a

true mineral but a rock that contains mostly böehmite and gibbsite along with diaspore, corundum and

numerous impurities (mostly compounds of iron, silicon and titanium). Bauxite commonly appears as

a collection of small, reddish-brown nodules in a light-brown, earthy matrix. The alumina available in

commercial bauxite ranges from 30 to 60 weight percent. Bauxite is typically classified according to

its intended commercial application: abrasive, cement, chemical, metallurgical, refractory, and other

end uses. The bulk of world bauxite production (approximately 85 percent) is metallurgical and used

as feedstock for the manufacture of aluminum.

The United States mines less than 1 percent of the bauxite it uses annually, virtually all of which is

used in the production of non-metallurgical products, such as abrasives, chemicals, and refractories.

Nearly all bauxite consumed in the United States is imported. Figure 4.1 tracks the the domestic and

imported components of U.S. bauxite supply from 1960 to 2003 (Appendix H).

Figure 4.1: U.S. Bauxite Supply 1960 to 2003

Million Mteric Tons

U.S. Mined Bauxite Imports of Bauxite

Source: Appendix H

In 2003, the United States imported a total of 8,300,000 metric tons of bauxite. About 95 percent of

the imported bauxite is refined to produce alumina and approximately 90 percent of the refined

alumina is used to produce primary aluminum. Approximately 7,097,000 metric tons of the imported

bauxite were refined for the primary production of aluminum.

Primary Aluminum Raw Materials

4.1.1 Bauxite Energy Requirements (Onsite and Theoretical)

Approximately 0.32(0.34

) kWh of process energy were required in 2003 to produce the 5.1

kilograms of bauxite needed to produce 1.0 kilogram of aluminum. Approximately 16.7

kilograms of carbon dioxide equivalent (CDE) were released for each metric ton of bauxite

mined.

The energy demand associated with the extraction of bauxite is typical of most mining operations.

Bauxite ore is generally strip-mined by removing the overburden (the soil on top of the deposit) and

excavating it with mechanical equipment. The overburden is saved for reclamation operations which

are extensively practiced to ecologically restore mined areas. The soft earthy nature of many bauxite

deposits generally does not require drilling or blasting operations. After mining, the bauxite is

crushed, sometimes washed and dried, and transported to refining plants via ship, barge, rail, truck or

conveyor belt.

Approximately 5.1 kilograms of bauxite are required to produce a kilogram of aluminum. The energy

requirement per kilogram of mined bauxite is 0.06 kWh for typical extraction.

Since electricity

accounts for less than 1 percent of the energy used in bauxite production, the tacit addition is

negligible (Appendix F, Table F.1).

Calculation of a theoretical minimum energy requirement for mining bauxite is dependent on the

system boundaries applied and the processes used. The laws of thermodynamics state that separating

the constituents of a mixture, such as bauxite from bauxite-rich soil, requires a certain minimum

expenditure of energy. Bauxite is the major constituent of bauxite-rich soils, and there is no change in

the chemical nature of bauxite in the mining process. So the theoretical minimum energy for

preparing bauxite is negligible. In addition, since it is theoretically possible to find bauxite on the

surface, the theoretical minimum energy requirement to produce bauxite is very close to zero. In the

interest of simplicity, this report uses a zero theoretical minimum energy requirement for mining

bauxite.

Emissions from fuels used in the extraction of bauxite are listed in Appendix E, Table E.2. These

emissions are typically from surface mining operations and result from a variety of fuels used in the

production of bauxite. Nearly 0.0167 kg CDE are emitted for each kilogram of bauxite mined.

4.2 Alumina (Al

)

Theoretically, from the stoichiometric equation (

2Al

+ 3C 4 Al + 3CO

), 1.89 kilograms

of alumina is required to produce 1 kilogram of aluminum. In practice, a very small portion of the

alumina supply is lost and the industry requires approximately 1.93 kilograms of alumina for

1960

1965

1970

1975

1980

1985

1990

1995

2000

Primary Aluminum Raw Materials

production of each kilogram of aluminum. In 2003, the United States produced 3,488,400 metric tons

of alumina from bauxite and imported an additional 2,300,000 metric tons of alumina to make

aluminum.

Figure 4.2:U.S. Alumina Supply 1960 to 2003

Million Metric Tons

Alumina Imports

From Imported Bauxite

From Domestic Bauxite

Source: Appendix H

Figure 4.2 shows the alumina supply sources from 1960 to 2003. The United States had three Bayer

refineries in operation in 2003, and one temporarily idled at midyear. These refiners processed about

7,752,000 metric tons of bauxite into 3,488,400 metric tons of alumina. About 10 percent of the

alumina produced is used to manufacture abrasive, refractory and other products. Approximately

3,139,560 metric tons of U.S. refined alumina were transferred to the primary aluminum industry.

This quantity of alumina was not sufficient to supply the U.S. demand for alumina; therefore, an

additional 1,300,000 (net) metric tons of alumina were imported (Appendix H)

All commercial alumina is refined from bauxite using the Bayer refining process. The process,

developed by Karl Bayer in 1888, consists of four major steps. Bauxite composition varies and

refining plant designs are slightly different to account for the site-specific quality of the bauxite.

1) Digestion – Crushed, ground, and sized bauxite is dissolved under pressure with a hot (180°C to

250°C) sodium hydroxide and sodium carbonate solution in a series of steam-heated digesters. The

concentrations, temperatures, and pressures employed vary depending on the properties of the

bauxite. Gibbsite is soluble in caustic soda above 100°C, while böehmite and diaspore are soluble in

caustic soda above 200°C. Since the treatments of böehmite and diaspore require higher temperatures

and longer digestion times, they are more expensive than treatment of gibbsite. The aluminum oxides

in the bauxite react to form soluble sodium aluminate or “green liquor.”

•

O + 2NaOH 2NaAlO

+ 2H

and

•

O + 2NaOH 2NaAlO

+ 4H

Primary Aluminum Raw Materials

Silicas in the bauxite are detrimental to the digestion efficiency. They react to form sodium aluminum

silicate, which precipitates. This precipitate chemically binds the aluminum from the bauxite and the

sodium from the sodium hydroxide into a solid from which the alumina cannot be economically

recovered. This decreases the yield of alumina and increases the costs associated with sodium

hydroxide. Chemical additions and the adjustment of refining practices can effectively provide

desilication and decalcification of specific alumina streams.

2) Clarification – The green liquor produced by digestion is clarified to remove sand, undissolved

iron oxides, titanium oxides, silica and other impurities. The insoluble materials, called “bauxite

residue” or “red mud,” are thickened, washed, and dewatered to recover sodium hydroxide. Bauxite

residue is a large-quantity waste product that is generally stored adjacent to refinery sites in landfills

or lagoons. After weathering, the landfills can sustain vegetation.

3) Precipitation – The clarified liquid that results from clarification is cooled and “seeded” with

crystals of gibbsite to aid precipitation of alumina trihydrate (Al

•3H

O). This is the reverse of the

digesting step (2NaAlO

+ 4H

O Al

•3H

O + 2NaOH). However, by carefully controlling

the seeding, temperature and cooling rate, specific physical properties can be given to the

precipitating alumina trihydrate.

4) Calcination – Alumina trihydrate is typically calcined in a fluid bed or rotary kiln at about 980°C

to 1,300°C to remove the water of crystallization and produce the dry white powder, alumina

(Al

•3H

O Al

+ 3H

O). Calcining rates and temperatures are carefully controlled and

vary depending on the final physical properties specified for the alumina.

Alumina used for electrolysis not only has a chemical purity specification, but also a physical

specification on particle size, surface area, bulk density and attrition behavior. These properties affect

alumina’s free flowing properties (how it flows in feeders), the rate at which it dissolves in cryolite,

dust levels, the strength of the alumina crust, its insulating properties and other properties important

in the aluminum electrolysis cell operation. Alumina’s bulk density is 880 to 1,100 kg/m

while its

specific gravity is 3.9 grams/cm

Bauxite residue (red mud) is a byproduct of the Bayer process and contains the insoluble impurities of

bauxite. The amount of residue generated per kilogram of alumina produced varies greatly depending

on the type of bauxite used, from 0.3 kilograms for high-grade bauxite to 2.5 kilograms for low-grade

bauxite. Its chemical and physical properties depend primarily on the bauxite used and, to a lesser

extent, the manner in which it is processed.

Although a great deal of effort has been expended over several decades to find and develop uses for

bauxite residue, a cost-effective, large-scale bulk application has yet to be found.

Numerous

attempts have been made to recover additional metals from the residue, such as iron, titanium, and

gallium. Other possible uses for the residue have included production of ceramic bricks or tiles, use as

Primary Aluminum Raw Materials

roadbed material or as filler material for plastics, or production of cement. Accordingly, the current

industry efforts focus on minimizing the amount of residue generated and improving its storage

conditions.

Probably the most promising and recent application for the residue has occurred in Western Australia,

where it is being evaluated as a soil amendment or conditioner. The soils in Western Australia are

sandy and drain freely, allowing fertilizers to leach into waterways where they boost nutrient levels

and can lead to problems such as algal blooms. The application of bauxite residue to these sandy soils

aids the retention of phosphates and moisture, and reduces the need to apply lime for soil pH

adjustments. As part of the Bauxite Roadmap project on bauxite residue, a model farm has been

planned to run for 5-7 years with the application of red mud, and all its aspects and results will be

studied in great detail.

4.2.1 Alumina Energy Requirements

Approximately 7.27(7.87

) kWh of energy were required and 1.62 kg CDE were released to

refine the 1.93 kg of alumina from bauxite needed to produce one kilogram of aluminum in

2003.

The energy required to produce alumina from bauxite in 1985 was estimated to range from 2 kWh/ kg

to 9 kWh/ kg of alumina.

This broad range of energy intensity reflects both bauxite quality (alumina

content) and refinery design. It was estimated that in 1991, U.S. refiners averaged 3.66 kWh/kg of

alumina produced.

The most recent available refining data lists 3.76 kWh/kg of alumina for 1995.

Calcination is the most energy-intensive operation of the Bayer process. On average, 1.93 kg of

alumina are consumed to make one kilogram of aluminum.

The alumina energy requirement for one

kilogram of aluminum can be estimated as 3.76 kWh/kg of alumina times 1.93 kg alumina, or 7.27

kWh/kg of aluminum, a tacit value of 7.87

kWh/kg of aluminum.

The Alumina Technology Roadmap provides insight into the high-priority research and development

needs of the global alumina industry.

The Roadmap recognizes the need and the opportunity for a 25

percent energy reduction by 2020 and improved, more sustainable handling of bauxite residues.

Better chemical process knowledge, waste heat utilization, and cogeneration are opportunities for

energy reduction in the refining process.

Emissions from fuels used in the refining process are listed in Appendix E, Table E.2. These

emissions are predominately related to natural gas and coal consumption for digestion and

calcination. Nearly 1.62 kg CDE are emitted for each kilogram of refined alumina.

Primary Aluminum Raw Materials

4.2.2 Alumina Theoretical Minimum Energy Requirements

The theoretical minimum energy required to produce alumina is 0.13 kWh/kg of aluminum

produced.

The theoretical minimum energy requirements to produce metallurgical grade alumina from bauxite

can be calculated from the reactions required in the process. The minimum energy requirements for

the digestion, clarification, and precipitation steps are related to the two chemical reactions that take

place during these processing steps.

•3H

O + 2NaOH 2NaAlO

+ 4H

O Al

•3H

O + 2NaOH

However, since there is no net chemical or temperature change for the combined reactions the

theoretical minimum energy requirement for these steps is near zero. The final reaction, calcination,

requires energy to remove the hydrated water from alumina (Al

•3H

O Al

+3H

O). The

theoretical minimum energy requirement to calcine (dehydrate) the alumina is 0.13 kWh/kg of

aluminum produced (Appendix J, Table J.10). This value assumes that the precipitated alumina

trihydrate is completely bone dry before entering the calcining process. At 8 percent moisture, this

would add about 0.01 kWh/kg of aluminum to the minimum requirement.

4.3 Carbon Anode

The United States consumed 1,651,000 metric tons of carbon anode in 2000. Approximately 0.45

kilograms of carbon anode were needed to produce one kilogram of aluminum. (Appendix F, Table

F.1) All commercial production of aluminum uses carbon as the anode material for the electrical

reduction of alumina to aluminum. The carbon anode net reduction reaction

(

2Al

+3C 4Al+3CO

) requires three carbon atoms for the reaction to free four aluminum

atoms. The theoretical minimum anode consumption is 0.33 kg of carbon per kilogram aluminum [(3

x 12.01 carbon molecular weight )/( 4 x 26.98 aluminum molecular weight)]. Anode material quality

is important since all the impurities dissolve into the bath and ultimately contaminate the molten

aluminum. The anode’s physical quality also affects both the energy efficiency and productivity of

smelting cells.

Anode consumption rates in practice, typically about 0.45 kg carbon per kg of aluminum, are

35 percent higher than the theoretical requirement. Excess carbon usage results from the need to

protect the iron electrical connection within the carbon anode, and from air burning and dusting. The

surface of the carbon is hot enough at cell operating temperatures to oxidize at a slow rate. This is

minimized by coating the anode surface and covering the anode with alumina, which insulates it from

air exposure.

Primary Aluminum Raw Materials

Dusting, breaking off small particles of carbon into the air or bath, can account for half of the excess

carbon used in the smelting process. Dusting is a direct function of the anode material uniformity. It is

caused by selective electrolytic oxidation and air burning of the binder pitch, which releases

aggregate carbon particles into the bath. Anode carbon dust is unavailable for aluminum production.

Two different anode technologies are utilized by the U.S. industry. “Prebaked” carbon anodes account

for more than 86 percent of the U.S. capacity

Older, “in-situ-baked” Söderberg anodes account for

the remainder of the capacity. New prebaked anode reduction cells have surpassed Söderberg anodes

in terms of current efficiency and emissions control. Only three operational smelters in the United

States are currently using Söderberg anodes. No new Söderberg cells are being built, and those that

exist are progressively being replaced, converted, or shut down. This report will focus mainly on

“prebaked” anodes for smelting technology.

Carbon “prebaked” anodes are made by mixing ground used carbon anodes, calcined petroleum coke

and coal tar or petroleum pitch. Pitch acts as a binder to hold the anode mass in a “green” formed

shape. Compacting the anode by using vacuum and vibrating the mixture when forming produces a

denser, more conductive and lower dusting anode. Baking carbonizes the pitch and creates a solid

bond between the particles of calcined coke and used anode material. Cast iron is poured into

preformed sockets in the baked anode to form an electrical connection. “Prebaked” anodes can weigh

as much as 1,250 kg and have a working face size of about 0.70 m x 1.25 m and a 0.5 m height.

“Prebaked” anodes are removed before they are completely consumed. Used anodes are recycled into

the anode production system to recover the carbon and the iron rods used for electrical connections.

Used anodes can account for 15 to 30 percent of the mass used in green anode makeup.

Calcined petroleum coke is a byproduct of the crude oil refining industry. Green or raw coke contains

8 to 10 percent moisture and 5 to 15 percent volatile organic materials. Raw coke must be calcined at

about 1,200°C to 1,350°C in gas-fired kilns or rotary hearths to remove the moisture, drive off

volatile matter, and to increase the density, strength and conductivity of the product.

Calcined coke

has a bulk density of about 800 kg/m

. Worldwide, about 25 percent of all raw coke is calcined and

about 70 percent of all calcined coke goes to aluminum production. Modern calcining hearth and kiln

designs capture and use the volatile organic matter in raw coke as their major fuel source.

4.3.1 Carbon Anode Energy Requirements

Approximately 0.66(6.02

) kWh of energy were required and 0.12 kg CDE were released in the

manufacturing of the 0.45 kg of carbon anode needed to produce one kilogram of aluminum in

2003.

Anode blocks are typically baked in a natural gas-fired furnace for several weeks. Quality anodes

Primary Aluminum Raw Materials

depend upon careful baking controls to gradually raise the temperature to about 1,250°C. Volatile

hydrocarbons from the pitch are gradually released during the baking process. Theoretically, these

volatile compounds could provide sufficient heat for anode baking and no additional energy would be

required. However, in practice, volatile organic compounds account for only 46 percent of the energy

input to the prebake ovens. The remaining 54 percent of the energy needed comes from fuel. Only

about 30 percent of the input energy goes into making the anode, 24 percent is lost from oven

surfaces, 29 percent goes up the stack and 17 percent is lost in other ways. New prebake furnace

ovens with computer controls are more efficient with both regenerative and recuperative elements.

Emissions associated with prebaked anodes result mainly from the combustion of natural gas and the

volatile organic compounds contained in the pitch. These amount to 0.27 kg CDE per kilogram of

anode or 0.12 kg CDE per kilogram of aluminum (Appendix E, Table E.2).

The process energy used to produce a

Table 4.1: Energy Associated with Carbon Anode Manufacturing

carbon anode is 1.36(1.66

) kWh/kg of

anode from the most recent available

U.S. data for 1995.

The total energy

(see Appendix I) to produce a carbon

anode is shown in Table 4.1. The

energy per kg of aluminum produced is

obtained by multiplying 0.45 times the

total energy required to produce a

kilogram of carbon anode.

kWh per kg of Anode kWh per kg of Aluminum

Pitch

Coke

anode

0.003 (2.58

)

0.13 ( 8.99

)

1.36 (1.93

)

0.45 X a

0.001 (1.15

)

0.06 (4.01

)

0.61 (0.86

)

Total

1.49 (13.50

) 0.66 (6.02

)

Söderberg anodes use green coke and are baked in-situ. These anodes can be baked only to the

maximum cell-operating temperature, which results in an anode with 30 percent higher electrical

resistivity and a greater dusting propensity than a prebaked anode.

The cell emission control system

is also more complex than for a prebake cell, since emission systems must be designed to handle the

5 percent to 15 percent volatile organic material content of the green coke.

Some plants employ both

wet and dry gas scrubbing systems to meet the environmental regulations.

4.3.2 Carbon Anode Theoretical Energy Values

The minimum theoretical energy requirement to manufacture a carbon anode is the energy necessary

to convert the coal tar pitch by destructive distillation to a coke-based binder. Approximately one-

third of the pitch binder mass is lost in the baking process.

A portion of this loss consists of volatile

components, while the remainder is the carbonization of the pitch. Pitch contains approximately

85 percent carbon. Approximately 79 percent of the pitch is actually carbonized, while 21 percent is

volatilized. The fuel energy value of the pitch that is carbonized, 0.75 kWh/kg of anode, is a measure

Primary Aluminum Raw Materials

of the theoretical energy required for anode manufacturing. In addition, the tacit or inherent energy of

11.55

kWh/kg of anode must be accounted for as part of the total theoretical requirement. Therefore,

the total theoretical minimum energy associated with the production of pre-baked carbon anodes is

12.30

kWh/kg of carbon anode. Further, since 0.33 kg of carbon anode are required to produce a

kilogram of aluminum in theory, the minimum energy requirement to produce anodes is 4.1 kWh/kg

of aluminum.

1960

1965

1970

1975

1980

1985

1990

1995

000

5. Primary Aluminum Production

The total energy associated with primary aluminum production from bauxite ore was

approximately 23.78(45.21

) kWh/kg of aluminum in 2003. This consisted of:

• 8.20(14.11

) kWh/kg aluminum for raw materials, and

• 15.58(31.10

) kWh/kg aluminum for electrolytic reduction.

Alumina is insoluble in all ordinary chemical reagents at room temperature and has a high melting

point (above 2,000°C). These properties make conventional chemical processes used for reducing

oxides difficult and impractical for conversion of alumina into aluminum.

Commercial primary aluminum is produced by the electrochemical reduction of alumina. Charles

Martin Hall in the United States and Paul Lewis Toussaint Héroult in France independently developed

and patented a commercially successful process for alumina reduction in 1886. This process,

commonly referred to as the Hall-Héroult process, is still the primary method in use for aluminum

production. Though the engineering has improved vastly, the process fundamentals are basically

unchanged today. The Hall-Héroult process takes place in an electrolytic cell or pot. The cell consists

of two electrodes (an anode and a cathode) and contains a molten bath of fluoride compounds

(cryolite), which serves as an electrolyte and solvent for alumina. An electric current is passed

through the bath, which reduces the alumina to form liquid aluminum and oxygen gas. The oxygen

gas reacts with the carbon anode to form carbon dioxide. Molten aluminum collects at the cathode in

the bottom of the cell and is removed by siphon.

5.1 Production, Capacity, and Growth

In 2003, 10 companies operated 15 primary aluminum production facilities in the United States.

These facilities having a production capacity of approximately 4,149,000 metric tons, operated at

approximately 65 percent capacity to produce 2,704,500 metric tons of aluminum in 2003.

Figure 5.1:U.S. Production of Primary Aluminum from 1960 to 2003 (Source: Appendix G)

Million Metric Tons

Primary Aluminum Production

Figure 5.1 shows variations in the production of primary aluminum in the United States from 1960 to

2003. These production variations are more representative of the costs to produce aluminum than of

the domestic demand. Primary aluminum is traded on a global market, and global supply has been

growing steadily at a rate of 3.6 percent annually for the past ten years. The United States accounted

for 9.8 percent of the world’s primary aluminum production in 2003.

5.2 Historical Hall-Héroult Energy Utilization

“The first commercial aluminum cells at Neuhausen, Switzerland (Héroult) and Pittsburgh,

Pennsylvania (Hall) required more than 40 kWh/kg of aluminum produced and had current

efficiencies ranging from 75 percent to 78 percent.”

The Hall-Héroult process is still electric energy

intensive. Since electricity costs are an important component (about one third) of the total production

costs, energy efficiency continues to be a major area of focus for the aluminum industry.

The electrical energy consumed in a primary

aluminum cell is measured by the number of

watts consumed over a period of time.

Wattage is determined by multiplying the

cell voltage by cell amperage. Figure 5.2

shows the significant electrical energy

improvements made between 1900 and

2000. Total electricity usage (excluding tacit

generation and transmission losses) varies

from less than 13 kWh/kg of aluminum for

the state-of-the-art plants up to more than 20

kWh/kg for older Söderberg facilities (U.S.

plants in 1995 averaged around 15.4 kWh/kg

of aluminum). The theoretical minimum

energy requirement for carbon anode

aluminum electrolysis is approximately 5.99

kWh/kg of aluminum (see Section 5.3.1).

Compared to theoretical values, U.S. facilities are operating at roughly 38 percent energy efficiency.

Significant engineering changes in cell design and operation have occurred over the past fifty years.

Table 5.2 shows the changes in operating parameters of a typical cell from 1948 to 1999.

Each new

or updated primary facility tries to increase productivity and incorporate energy-reducing

technologies to lower production costs. This results in gradual changes in the industry. The most

significant change is the new equipment and techniques allowing for smaller and more frequent

alumina additions (point feeders). This, combined with higher amperage, lower current density and

larger cells, has dramatically improved current efficiencies and productivity.

Figure 5.2: Primary Aluminum Electric Energy

Consumption 1900 to 2000

Primary Aluminum Production

Table 5.1: Typical Parameters of Aluminum Reduction Cells 1948 vs. 1999

Parameter 1948 1999

Current Rating (kA)

Aluminum Production (kg Al /pot day)

Energy Consumption (DC kWh/kg Al)

Anodic Current Density (A/cm

)

Area of Cavity (m

)

Nominal Anode Area (m

)

Ratio of Area (Anode/Cavity)

Average Velocity of Flow in Cathode (cm/s)

Cathode Life (days)

Potroom Workers (hours/metric ton Al)

Interval for Alumina Additions (minutes)

Emissions (kg /metric ton Al)

Anode Effects per Pot Day

Net Anode Carbon Consumption (kg C/kg Al)

Number of Pots per Potline

World Primary Production (10

) metric tons

50 - 60

385

18.5 - 19

1.2 - 1.3

4 - 5

~ 0.55

10 - 15

600 - 800

5 - 8

80 - 240

~ 30

~ 1.5

3 - 4

0.55

~ 40

~ 1

300 - 325

2,475

12.9 - 13.5

0.8 - 0.85

40 - 45

~ 0.9

4 - 6

2,500 - 3,000

1.7

0.7 - 1.5

< 0.5

0.05

0.43

~ 288

~ 20

There is a minimum cell amperage

(electrical current) required to produce

aluminum (see Section 5.3). Production

in the the United States now operates at

about 95 percent current efficiency, a

significant improvement over the past

several years, as shown in Figure 5.3.

The high current efficiency of existing

technologies leaves little opportunity for

process or technology improvements to

further reduce amperage and save

additional energy. Since current

efficiency is high, lowering the voltage

requirements of cells presents the largest

challenge and the best opportunity for

improving Hall-Héroult efficiencies. The

voltage requirements of a cell are

described in section 5.4.2.

Figure 5.3: Primary Aluminum Current Efficiency

1900 to 2000

Source: Production of Aluminum and Alumina

Primary Aluminum Production

5.3 Theoretical Minimum Energy Requirement for Reduction

Figure 5.4: Alumina to Aluminum Theoretical

Minimum Energy

All current primary production facilities and most alternative processes for aluminum production use

alumina as their raw material. Any process that starts with alumina to produce aluminum has the same

theoretical energy requirement. Different processes do not offer any theoretical energy advantage.

However, they do offer significant tradeoffs between efficiencies, emissions, footprints, and sources

of input energy (electricity, carbon, and fuels). The theoretical limits required to manufacture

aluminum provide a valuable insight into Hall-Héroult cell operation and potential future reduction

processes.

The product of primary reduction process is molten

aluminum. This report calculates the theoretical

minimum energy by assuming the reactants enter

and the byproducts leave the system at room

temperature and that molten aluminum leaves the

system at 960°C. The molten metal temperature,

960°C (1233°K), is an approximation of an average

commercial operating cell. Figure 5.4 illustrates the

theoretical boundaries for a system that reduces

alumina to form aluminum and oxygen. Changes in

the operating temperature of a cell have a minor

effect on the theoretical energy requirements.

Operating changes of 100°C in a Hall-Héroult cell, operating in the range of 700°C to 1,100°C, result

in less than a 1 percent change in the theoretical minimum energy requirements (Appendix J, Table

J.7).

Some studies assume that the gases evolved during reduction leave the system at the molten metal

temperature. In these studies, the theoretical minimum is 2.5 percent to 3 percent higher than the

values presented in this report (Appendix J). Theoretically, it is possible to capture all the energy

associated with these gaseous emissions. In practice, however, the gas stream is collected from

hundreds of hooded pots and treated before release to the atmosphere. Only a very small portion of

the heat is actually absorbed and returned to the system.

Three energy factors must be examined in the production of aluminum: the energy required to drive

the reduction reaction forward, the energy required to maintain the system at constant pressure and

temperature, and the energy required to change the temperature of the reactant and/or product. The

thermodynamics and chemical equilibrium of the reactions are described by the following equation:

∆G = ∆H – T∆S

The energy required to drive the reaction forward is the energy for the electrolytic reduction of

alumina (

2Al

4Al + 3O

) and is given by the change in the Gibbs free energy value (∆G).

The energy required to maintain system equilibrium is the difference between the heat of reaction

----------------------------------------- ---------------------------- ---------------------- ---------------------------------- ----------------------

Primary Aluminum Production

(∆H) and the Gibbs free energy value (∆G), which equals the entropy term (T∆S). Since the Gibbs

free energy requirement is less than the heat of reaction for alumina reduction, additional energy must

be added to the system to maintain the system temperature. Otherwise the system will cool as the

reaction proceeds. Hence, for the alumina reduction reaction, the ∆H term provides the minimum

theoretical energy requirement (reaction and equilibrium). Reduction cells operate at atmospheric

conditions and no pressure change results from the reduction. Numeric values for these

thermodynamic measures for the elements and compounds common to aluminum processing and the

calculations used to determine the theoretical minimum energy requirement are given in Appendix J.

The energy required to change the temperature of reactants and products is calculated from their heat

capacities (C

), which are provided in Appendix J.

Faraday’s law provides the minimum amperage requirement for electrolytic reduction. This law states

that 96,485 coulombs of electricity are passed through a cell to produce a one gram equivalent of an

element or compound. Aluminum has an atomic weight of 26.98, a charge of 3

and therefore, has an

8.99 gram equivalent weight. Faraday’s law is converted to more common measurements:

96,485 coulombs

gm equivalent

Ampere sec

coulomb

hour

3,600 sec

gm equivalent 1,000 gm

= 2,980 Ah/kg

8.99 gm kg

The value 2,980 Ah/kg of aluminum is the theoretical minimum amperage (current) required for

production. This value assumes perfect conditions, where there are no reverse or parasitic reactions

that consume amperage and no limitation to the ionic species availability to react at the electrodes (no

concentration gradients or gas bubbles). The Gibbs free energy (∆G) divided by the Faraday

amperage provides the minimum voltage required to drive the reaction forward. Cell voltage and

current efficiency are variables that are controllable by design and they determine the electrical power

required for reducing alumina. In practice, electrolytic cells have significant inefficiencies and

operate above the minimum voltage requirement. This excess voltage provides the thermal energy

required to maintain system equilibrium (∆H – ∆G) and to produce molten material (C

In the case of aluminum made directly from alumina (2Al

4Al+3O

), shown in

Figure 5.4, the energy required to drive the reaction forward (∆G) is 8.16 kWh/kg, the thermal energy

(∆H – ∆G) required to maintain thermal equilibrium is 0.48 kWh/kg, and the thermal energy (C

)

associated with producing the molten aluminum is 0.39 kWh/kg of aluminum. The theoretical

minimum energy requirement is 9.03 kWh/kg of aluminum. (Note: if the gas emission at 960°C is

included, the total theoretical minimum energy requirement is 9.30 kWh/kg of aluminum) (Appendix

J, Table J.2).

Primary Aluminum Production

5.3.1 Theoretical Energy for Hall-Héroult Carbon Anode Reduction

The theoretical minimum energy requirement for producing molten aluminum at 960°C in a

Hall-Héroult cell with a carbon anode is 5.99 kWh/kg.

All commercial aluminum production uses a carbon anode in a Hall-Héroult cell. The carbon is

consumed during the electrolytic process and supplies part of the energy necessary for the reduction

of alumina. This gives the Hall-Héroult carbon anode process a lower energy requirement than the

direct reduction of alumina to aluminum. The theoretical energy required for reduction is the same for

prebaked or Söderberg carbon anodes.

The net reaction for the carbon anode Hall-Héroult process is 2Al

+3C 4Al+3CO

Figure 5.5 shows an idealized Hall-Héroult cell

for the production of aluminum. In this cell, it is

assumed that the reactants (alumina and carbon)

enter the cell at 25°C, the carbon dioxide

byproduct leaves the cell at 25°C, and the

aluminum product leaves as molten metal at the

cell operating temperature of 960°C. The

reaction is assumed to occur under perfect

conditions, where there are no reverse reactions,

no parasitic reactions consuming additional

anode carbon, no limitations to the ionic species

reacting at the electrodes, and no heat or energy

losses external to the system.

Appendix J, Table J.1 details the calculation of theoretical minimum energy for this reaction. The

results show that the energy required to drive the reaction forward (∆G) is 5.11 kWh/kg, the thermal

energy required to maintain equilibrium is 0.49 kWh/kg and the thermal energy associated with the

molten aluminum is 0.39 kWh/kg of aluminum. Therefore, the theoretical minimum energy

requirement for the reduction of alumina in a carbon anode cell adds up to 5.99 kWh/kg of aluminum.

(Note: if the CO

gas emission at 960°C is included, the total theoretical minimum energy

requirement is 6.16 kWh/kg of aluminum).

In actual carbon anode cell operations, current efficiencies of less than 100 percent result from

reverse oxidation reactions between part of the aluminum metal that is dissolved in the cryolite and

carbon dioxide gas produced

(2Al+3CO

+3CO)

and to a lesser extent between the

carbon dioxide gas and the carbon anode

(CO

+C 2CO)

. Current efficiency losses can also

result from direct shorting of the anode to the aluminum pad.

Figure 5.5: Alumina and Carbon to Aluminum

Theoretical Minimum Energy

Primary Aluminum Production

Today’s state-of-the-art reduction cells are achieving current efficiency levels of more than 96 percent

and energy consumption levels of less than 13.0 kWh/kg of aluminum.

The theoretical minimum

energy requirement at 100 percent current efficiency is 5.99 kWh/kg of aluminum. The energy

efficiency levels of present state-of-the-art carbon anode reduction cells are about 46 percent.

5.4 Hall-Héroult Reduction Process

The engineering, materials, and process knowledge of existing components and processes form the

foundation for developing new components, processes, and techniques for producing aluminum. The

Hall-Héroult cell works as a system where all the components perform together. Therefore,

improving one component may not necessarily result in an improved cell or a more energy-efficient

operation. To understand the impact of component and system changes on the cell performance, it is

helpful to know about the Hall-Héroult process in terms of its components, operations and

interrelationships. This section explains the process by describing a typical prebake anode operation.

5.4.1 Typical Hall-Héroult Cell Operation

A typical modern aluminum electrolysis

Hall-Héroult reduction cell (pot) is a

rectangular steel shell 9 m-12 m long, 3 m-

4 m wide and 1 m-1.5 m deep (Figure 5.6). It

has an inner lining of carbon, which is

surrounded by refractory thermal insulation,

that keeps it isolated thermally and

electrically from the steel shell. Commercial

cells range in capacity from 60,000 amperes

to more than 500,000 amperes and can

produce more than 450 to 4,000 kilograms of

aluminum per day, respectively.

A cell typically operates at 950°C to 980°C

and yields molten aluminum and carbon

dioxide. The molten aluminum has a higher

density than the electrolyte (cryolite bath)

and settles to the bottom of the cell on top of the carbon lining. Molten aluminum at about 99.7

percent purity is periodically “tapped” by a vacuum siphon from the cell bottom. The tapped metal is

transferred to holding furnaces where the metal is alloyed, and entrained gases and impurities are

removed prior to casting. The carbon dioxide and other gases generated in the cell during the

reduction process are collected and treated to meet environmental regulations.

Figure 5.6: A Typical Hall-Héroult Cell

Primary Aluminum Production

Electric current enters the cell through the carbon anode and flows through 3 cm–6 cm of electrolyte

(bath) to the aluminum pad and carbon lining cathode. The aluminum pad is in intimate contact with

the carbon lining and serves as the charged surface of the cathode. Steel collector bars are set near the

bottom of the carbon lining to conduct the current to the anode of the next cell.

The 950°C to 980°C molten cryolite and aluminum used in a typical reduction cell are corrosive.

Molten cryolite has low viscosity and interfacial tension that allows it to easily penetrate any porosity

in the cell lining. To protect the carbon lining, the thermal insulation is adjusted to provide sufficient

heat loss to freeze a protective coating of the electrolyte, known as “ledge,” on the inner walls. The

molten aluminum pad protects the carbon bottom of the cell. The cell is never tapped completely dry

of molten aluminum. It is essential that no alumina or frozen ledge form under the metal pad. The

carbon cathode must remain bare for good electrical contact with the aluminum pad and to minimize

wear of the carbon surface.

The reduction reaction is continuous and alumina must be supplied to the bath at a controlled rate to

maintain constant conditions. This is accomplished with automatic feeders that break the surface crust

and deposit alumina into the molten bath where it is dissolved for reaction. Alumina is also used to

cover the carbon anodes and the frozen bath surface. The alumina covering serves as thermal

insulation and as a protective cover to reduce air burning of the anode.

The electrolytic reaction in a Hall-Héroult cell consumes the carbon anode. Approximately 0.45

kilograms of the carbon anode is consumed for each kilogram of aluminum produced. The carbon

anodes provide a necessary part of the energy required to operate a cell. The distance between the

carbon anode and the metal pad is kept constant by adjusting the anode as it is consumed. The

consumable carbon anodes must be replaced periodically, typically about every four weeks in a

modern plant. The frequency of anode changing depends on the anode design and the cell operation.

Anode changing represents the most frequent cell and productivity disruption. The removed portion

of an anode (known as a “butt”) is recycled or sold as a fuel. The pot cover, which is part of the gas

collection system, must be removed, the used anode must be pulled from the frozen surface crust, and

the new anode must be inserted into the space of the consumed anode. This has to be accomplished

without significant pot crust breakage or alumina falling into the bath. Anode changing is the single

largest thermal, current, and magnetic disturbance in cell operation.

Cells are arranged in long rows called “potlines.” They are placed as close as possible to each other

while maintaining sufficient room for anode changing, alumina feeding, and reasonably low

electromagnetic interference. The cells are connected electrically in series. Rectifiers, which convert

alternating current to direct current, are chosen to minimize capital investment and typically provide

about 700 V(dc). Typically a reduction cell’s design requires about 4.6 V(dc) so that a potline of

roughly 150 to 180 cells would be used.

Primary Aluminum Production

Figure 5.7: Voltage Distribution in a Hall-

Héroult Cell

5.4.2 Voltage Requirements

The energy consumed in an electrolytic reaction is a

function of the voltage used and the current efficiency of the

operating cell (the minimum current is fixed by Faraday’s

Law, see Section 5.3 on page 27). Modern Hall-Héroult

cells operate at high (96+ percent) current efficiencies. The

approximate voltage components of a conventional cell are

shown in Figure 5.7.

The electric current flows through the cell and the cell

voltage components can be described as a set of resistors in

series.

E = cell reaction + overvoltage + bath + cathode + anode

+ connectors

The cell reaction voltage is a function of temperature and at

960°C, is fixed at 1.2 V(dc).

This is the theoretical

minimum voltage required for the reduction reaction to take

place and no cell can operate at 960°C below this voltage.

The total cell operating voltage includes the addition of voltages required to overcome the ohmic

resistance of the other cell components. These are described in the following section of the report.

5.4.3 Cell Subsystems and Variables

Busbars and Pot Connectors

Busbars electrically connect in series all the cells of a single potline, which typically contains more

than 150 pots or cells. They are fabricated from highly conductive aluminum alloy and are sized for

minimum overall system cost. Any voltage drop in the busbar and connector system results in energy

loss.

Electrolyte

The electrolyte or bath used in Hall-Héroult is cryolite (Na

AlF

). This is modified with the addition

of aluminum fluoride (AlF

), calcium fluoride (CaF

) and other additives to control the operating

temperature, solubilities, activities of ionic species, conductivity, viscosity, interfacial tension, bath

density, vapor pressure, hardness of the crust, and other factors. Bath chemistry, physical properties

Primary Aluminum Production

and thermodynamics are very complex. The bulk electrical conductivity of the bath is influenced not

only by its composition and temperature, but also by the presence of anode gas bubbles and carbon

dust.

Aluminum fluoride (AlF

) is the most common bath additive. It lowers the operating temperature, the

solubility of the reduced aluminum, surface tension, viscosity and density. However, it has the

undesirable effect of decreasing alumina solubility and electrical conductivity. The weight ratio of

NaF to AlF

is referred to as the bath ratio. Controlling this ratio is important for efficient cell

operation (Note: Outside the United States, many countries use the cryolite or molar ratio, which is

twice the bath ratio).

The fluid bath circulates within the cell. As gas molecules are formed at the anode, they accumulate

and coalesce into fine bubbles that aggregate into larger bubbles. These bubbles collect and move

across the anode surface to escape around the edges of the anode. The buoyancy of the gas creates

movement, which contributes to the motion of the bath and pad. The bath movement results from, in

decreasing order of magnitude, gas bubble drag and electrolyte density difference caused by the

bubbles generated at the anode, electromagnetic forces on the molten metal pad, and temperature

gradients. This motion influences the concentration gradients of dissolved alumina and affects current

efficiency. The motion also influences the heat transfer from the bath to the protective frozen ledge.

Anode-Cathode Distance

The anode-cathode distance (ACD) is the distance between electrode surfaces. In the Hall-Héroult

cell, it is the distance from the lower face of the carbon anode to the top surface of the aluminum pad.

This distance is typically about 4 cm–5 cm. The electrolytic bath occupies the space between the

carbon anode and the aluminum pad. The voltage required for current to pass through the bath is

related to the bath conductivity and the distance between the anode and the cathode. Decreasing the

ACD lowers the voltage and energy requirements of the cell. The operating ACD is a compromise

between keeping a low value of bath resistance, while at the same time enabling electrolyte rich in

alumina ionic species to reach the charged surfaces and allowing reactant gas bubbles to escape. The

ACD also must be large enough to ensure that the liquid metal does not contact the anode and short

circuit the cell.

The heat required to keep the bath molten is in part supplied by the electrical resistance of the bath as

current passes through it. The amount of heat developed depends on the current path or the ACD.

Changing the ACD is one method of controlling the desired bath operating temperature.

Primary Aluminum Production

Aluminum Pad

The molten aluminum pad that forms at the bottom of the cell is the cathodically-charged surface for

the reduction reaction. The large amperage flowing through the cell creates electromagnetic forces

and torques that cause the metallic pad to rotate. These magnetohydrodynamic (MHD) forces create

this motion which deforms the molten aluminum/cryolite interface. The MHD forces cause local

metal velocities of about 5 cm/sec to occur. Control of MHD effects is one of the key factors for

successful cell operation with high current efficiency and low energy consumption. Cells are designed

to minimize these forces and today, velocities are typically one third of what was common fifty years

ago. Movement of the aluminum pad is mainly caused by electromagnetic forces in the cell and, to a

smaller extent, by the interfacial drag of the bath fluid. Joint discontinuities in the carbon blocks

create additional flow disturbances in the moving pad. The combination of all these forces causes the

pad to undulate or roll which could result in waves forming on the surface of the pad. These waves

can approach the anode and result in an electrical short circuit. The current that flows during this

shorting produces no aluminum and results in a major loss of power and productivity. The motion of

the aluminum pad also produces erosion of the carbon lining and shortens cell life. Since the distance

between the anode and cathode is constantly changing as a result of the undulating pad, the ACD is

kept large enough to avoid contact between the anode and the pad. This requires that the anode be

backed away from its optimal position. Designing systems to minimize movements of the metal pad

is a key factor in the efficient operation of a cell. A stable pad surface will allow the ACD to be

decreased.

The newer concept of using a “drained cathode cell” is an approach to circumvent the difficulties

associated with keeping the metal pad stable. Essentially the bulk of the metal is drained to a sump

and the cathode is left wetted only by a thin metal sheet (Advanced Hall-Héroult Cells, Section 6.1 on

page 43).

Cathode

The bottom lining of the cell also serves as a cathode and carries current from the molten metal pad.

Since 10 percent of the total cell voltage drop is in the carbon cathode blocks, it is important that they

have the highest density and electrical conductivity possible.

In an attempt to reduce the resistance of the cathode, some have experimented with cathode blocks

containing a higher content of graphitic carbon. However, while less resistive, graphite is also less

wear resistant and this compromises the life of the cathode. The cathode life generally determines cell

life, since cathode replacement requires the complete dismantling of a cell. The advent of hard

titanium diboride (TiB

) coatings may offer an opportunity to increase the graphitic content of

cathode blocks, and lower cell resistance, without reducing the cell life. Under optimal conditions, the

life of a cathode or cell is in the range of 7 to 10 years.

Primary Aluminum Production

Current Density

Current density is a measure of the productivity of a cell. It is calculated by dividing the amperage

supplied to an anode by the geometric face area of the anode. It is generally expressed in amperes per

square centimeter (A/cm

). Most potlines operate in the range of 0.8 A/cm

to 1.0 A/cm

. The

quantity of aluminum produced per cell increases with increasing current density. The tradeoff is that

as current density and productivity increase, current efficiency decreases, which results in a higher

energy consumption per unit of metal produced. Lower current densities are more energy efficient,

but increase capital and labor costs per unit of output.

Cell Polarization/Overvoltage

The reactions occurring at the anode and the cathode create localized conditions that are different

from the bulk of the bath. The reactions deplete the supply of reactants and increase the quantity of

products. This creates concentration gradients, which in turn cause concentration polarization.

Additionally, the gas generated at the anode forms bubbles, which lower the effective bath

conductivity. Localized conditions at the anode and cathode are unavoidable and require a voltage

higher than the minimum reaction voltage to be applied to the cell.

No free aluminum (Al

) or oxygen (O

2–

) ions are present in the bath. Alumina dissolves and

dissociates into salt complexes in the bath. The dynamics at the anode are complicated by the release

of gas bubbles. Oxygen-containing ionic species are transported through the bath and discharged on

the carbon anode. These anode reactions are:

2–

–

C CO

+ 2AlF

+ 4e

–

and

2–

+ 4F

–

+ C CO

+ 2AlF

–

+ 4e

–

At least three phenomena have been identified as contributing to the anode overvoltage:

1. an increase in the local current density due to the presence of gas bubbles adjacent to the anode

surface, which displace the electrolyte bath,

2. an ohmic component from an increase in the resistance of the electrolyte due to the presence of

the bubbles, and

3. the concentration polarization overvoltage.

Polarization effects at the cathode contribute much less to overvoltage than at the anode. The cathode

–

reactions are (

AlF

3–

–

Al + 6F

–

) and (

AlF

–

Al + 4F

–

). The aluminum ion

–

complexes, AlF

3–

and AlF

have higher ionic mobility than their anodic counterparts, which lowers

the concentration polarization effect. In addition, no gas bubbles, which influence both resistance and

concentration polarization, are produced at the cathode.

Primary Aluminum Production

Anode Effects

Control of the quantity of alumina dissolved in the bath is important for proper operation of a cell.

Alumina saturation is reached at about 7 percent alumina dissolved in a typical bath. The normal

operating level is about 3 percent alumina. If the level goes above 4 percent, some of the added

alumina may not dissolve rapidly and can settle to form a sludge on the cell bottom, thereby reducing

the cell conductivity. If it falls below 1 percent alumina the cell is starved of the reactant and an

“anode effect” ensues. When this occurs, the production of metal is interrupted and fluorine (F

hydrogen flouride (HF), and two perfluorocarbon gases – tetrafluoromethane (CF

) and hexafluoro-

ethane (C

) – are discharged instead of carbon dioxide. These perfluorocarbon gases have 6,000 to

9,000 times the global warming potential of carbon dioxide and require gas scrubbing to meet

environmental regulations.

Significantly, alumina is a good absorbent of fluorine and hydrogen fluoride, but not perfluorocarbon

gases. This allows primary production facilities to use their alumina raw material as an absorbent in

dry gas scrubbing systems. The fluorides absorbed on the alumina in the scrubbers are recycled into

the pot feeding systems, so that both the alumina and fluorides can be reused in the process.

Alumina Feed

Alumina is ideally added to the cell at a rate that exactly replaces the alumina that has been reduced.

If alumina is fed too fast or in large increments, it may not dissolve and can form sludge. Sludge

affects fluid flows within the cell and contributes to erosion of the cathode block surface. Under-

feeding the cell results in an anode effect. There is no current technology for in-situ, real-time bath

analysis to provide precise control of alumina concentration in the bath and alumina feed rate.

Alumina is fed by automatic handling and conveyor systems. The specifications for alumina are

complex with numerous tradeoffs. It must dissolve rapidly; it should contain few impurities, and have

a high surface area, yet be a relatively large particle (this apparent inconsistency is overcome by the

particle having significant porosity). The particle must also be robust, create little dust, and resist

breakage during handling. The introduction of dry scrubbing systems for the cell offgases, wherein

the alumina is used as an adsorbent for fluoride emissions, further complicates the alumina

specifications.

Alumina is “side-worked” in older prebake cell designs and Söderberg cells. Side-worked cells

introduce alumina into the bath using automated crust-breaking and feeding machines that move

along the length of the cell. Side-working is time-consuming and can take one to four hours before the

machine feeds the same section of the cell again. Newer prebaked cell designs utilize the spaces

between the anodes to feed alumina into the cells. “Point feeders” pierce the crust and dispense small

quantities of alumina at numerous points, typically in the center of the cell. One-minute intervals

between point feeding is common. This frequent addition of small quantities of alumina takes

Primary Aluminum Production

Figure 5.8: Cryolite-Alumina Phase Diagram

advantage of the motion of the bath and provides significantly better control of local alumina

concentration. This provides for better current efficiency, fewer anode effects and less erosion caused

by solid alumina. Point feeders have proven to be significantly better and have replaced most feeding

systems in both prebaked and Söderberg cells.

Cell Operating Temperature

Bath chemistry controls the operating temperature of the

Hall-Héroult process. Most commercial cells operate near

960°C (1,233°K). Reducing the cell operating

temperature is an obvious approach to saving energy and

reducing capital costs by lowering insulation

requirements. However, controlling the operating

variables of a cell becomes more critical as temperature is

lowered. Dissolved alumina must be available for

reduction in the bath to have high cell productivity, and

minimal energy-consuming concentration polarization

and anode effects. Lowering the cell temperature lowers

the solubility range for alumina in a cryolite bath. The

solubility or phase diagram for a cryolite alumina system

is V-shaped with temperature on the y-axis and alumina

concentration on the x-axis, as shown in Figure 5.8. Cells

must operate within the V; operations to the left will

precipitate cryolite, and those to the right will precipitate

alumina. Decreasing the temperature narrows the cell

operating range with respect to alumina solubility. In practice, bath chemistry is modified with the

addition of aluminum fluoride, calcium fluoride and other salts. These modifications lower the liquid

phase temperature. However, the basic V-shape of the diagram is retained.

Heat Balance

Controlling the thermal balance of the cell is of prime importance for efficient operation and long cell

life. There is no commercially available material that can retain its insulating value and resist

penetration and chemical attack by cryolite at cell-operating temperatures. Accordingly, the method

used to protect the side walls is to allow some heat loss so that the temperature of the exposed surface

of the cell lining is below the freezing point of the bath. This creates a frozen layer or ledge of bath

that protects the linings. Sidewall heat losses can account for 35 percent to 45 percent of the total heat

loss in a cell.

The frozen ledge, due to phase relationships, differs in composition from the bath. If

the temperature rises above steady state, the ledge begins to dissolve and the bath ratio (sodium

fluoride to aluminum fluoride) changes. If the temperature is allowed to fall too much, ledge

formation is excessive, anodes are changed with great difficulty, alumina does not dissolve as readily,

Primary Aluminum Production

and the bath ratio is affected. The frozen electrolyte ledge also provides the electrical insulation of the

side walls. The thermal conductivity of frozen cryolite is an order of magnitude lower than that of the

molten cryolite.

Large cell designs require less energy to maintain operating temperatures because of the lower ratio

of cell surface area to volume. This is one factor in the trend to use larger cells in newer smelting

plants.

5.5 Environmental Considerations

The Energy and Environmental Profile of the U.S. Aluminum Industry,

compiled by the U.S.

Department of Energy, Office of Industrial Technologies Program, and Life Cycle Inventory Report

for the North American Aluminum Industry,

published by The Aluminum Association, provide

detailed environmental information on the overall aluminum production process. The byproducts

generated in the Hall-Héroult process that are of environmental concern can be grouped into three

areas: electrolysis, anode production, and cell waste products.

Electrolysis

Electrolysis green house gas (GHG) emissions in the Hall-Héroult process can be split into three

groups: reduction reaction emissions, carbon dioxide (CO

) and carbon monoxide (CO); process

upset perfluorocarbons emissions; and hydrogen fluoride (HF) formed from the inclusion of moisture

O) in the raw materials. Hydrogen fluoride gas is almost completely captured and returned to the

cells by the alumina dry scrubbing system used in modern facilities.

The carbon-based emissions associated with the reduction reaction come from three sources:

1. Reaction Products – The reaction produces oxygen that reacts with the carbon anode to produce

and small quantities of CO; this reaction produces 1.22 kg of carbon dioxide equivalents for

each kilogram of aluminum produced (Appendix E, Table E.4).

2. Air Burning – The carbon anode loses mass to oxidation with the atmosphere. This produces

0.30 kg of CO

for each kilogram of aluminum produced (Appendix E, Table E.4).

3. Electricity Generation and Transmission – The emissions related to the fuels used in

electricity generation for U.S. primary facilities are 5.38 kg of carbon dioxide for each kilogram

of aluminum produced (Appendix E, Table E.2).

Primary Aluminum Production

A total of 6.90 kilograms of CO

gas is generated from the reduction process for each kilogram of

aluminum produced in an average U.S. primary facility. It should be noted, that electricity-related

emissions for specific potlines vary widely. Potlines operating on electricity obtained from coal-fired

power plants produce 16.0 kg of CO

gas for each kilogram of aluminum produced, while potlines

using electricity from hydro-power plants produce close to zero CO

gas emissions.

The perfluorocarbon emissions are related to the “anode effect.” If the concentration of alumina in the

bath becomes too low, other reactions between the carbon anode and the bath occur and

tetrafluoromethane (CF

) and hexafluoroethane (C

) are generated. These gases have a high global

warming potential (GWP). The GWP of a GHG is a ratio developed to compare the ability of each

greenhouse gas to trap heat in the atmosphere relative to CO

gas. The GWP of CF

and C

is 6,500

and 9,200 respectively. In other words, 1 kg of CF

released to the atmosphere is equivalent in its

warming potential to 6,500 kg of CO

Aluminum smelting is the principal quantifiable source of perfluorocarbon in the United States. The

U.S. Environmental Protection Agency (EPA) estimates U.S. emissions from aluminum production at

500 metric tons of CF

gas and 50 metric tons of C

gas in 2003.

In 1995, the aluminum industry

entered into a “Voluntary Aluminum Industry Partnership” (VAIP) with the EPA to reduce

perfluorocarbon emissions by 46 percent during the next decade. Reductions in primary aluminum

production and efficiency improvements to reduce anode effects have reduced emissions of CF

and

gases since 1990 by 71 percent each. The U.S. aluminum industry and EPA are continuing the

VAIP to seek further GHG reductions beyond the original achievements. The total 2002 U.S.

aluminum perfluorocarbon emissions are 2.0 CO

equivalent metric tons per ton of aluminum.

The emissions of perfluorocarbon for older side-fed cells are one order of magnitude higher than the

emissions for cells with point feeders. The industry continues to improve point feed systems.

Ultimately, with the entire industry on point feeders and advanced cell control systems, it should be

possible to virtually eliminate anode effects and, hence, perfluorocarbon generation.

Anode Production

The CO

emissions for the carbon anode manufacturing amount to approximately 0.13 kg of CO

equivalents per kg of aluminum. In this report, the feedstock portion of the CO

emissions for carbon

anode manufacturing are included in the electrolysis. More specific information on this topic can be

found in the comprehensive life cycle information published by The Aluminum Association.

Primary Aluminum Production

Cell Waste Products

Aluminum electrolysis carbon slime (AECS) and spent potlining (SPL) are unavoidable byproducts

of the aluminum smelting process and are listed in the United States as hazardous wastes.

Development work is underway to mitigate problems associated with spent pot linings. Most

development efforts attempt to combust the carbon linings to destroy any remaining toxic chemicals,

to recover the valuable fluoride as AlF

, and to render the remaining material inert through

vitrification. In some areas of the world, SPL is destroyed by combustion in the production of cement.

5.6 Technological Change in the Next Decade

The Hall-Héroult electrolysis process, utilizing a carbon anode and cryolite bath, is a mature

technology. However, gradual improvements in both productivity and environmental performance are

still possible. The typical Hall-Héroult cell life ranges from seven to ten years. Adoption rates of new

technology and systems are governed to some degree by the cell life. There is a slow autonomous

efficiency improvement in the Hall-Héroult process because of the continued adoption of advanced

cell designs, feeding systems, bath composition, control systems, and other practices. This trend has

resulted in a gradual decline in energy consumption in the range of 0.2 percent to 0.5 percent per year.

Energy savings are actively pursued by aluminum producers since electricity costs constitute a high

percentage of total production costs. Since current efficiencies are already over 95 percent, the goal is

generally to reduce the overvoltage in the aluminum cells to increase the overall electric efficiency.

A number of technological and engineering improvement options exist and are being adopted by the

aluminum industry. These include:

• Point Feeders – Point feeders enable more precise, incremental alumina feeding for better cell

operation. Point feeders are generally located in the center of the cell and thereby cut down on the

diffusion required to move dissolved alumina to the anodic reaction sites. The controlled addition

of discrete amounts of alumina enhances the dissolution process, which aids in improving cell

stability and control, minimizing anode effects, and decreasing the formation of undissolved

sludge on the cathode. In the jargon of modern commerce, point feeders enable “just-in-time

alumina supply” to permit optimum cell operation. Point feeder improvements continue to be

made as more accurate cell controllers become available.

• Improved Process Controls – Advanced process controllers reduce the frequency of anode

effects and control operational variables, particularly bath chemistry and alumina saturation, so

that cells remain at their optimal conditions.

• Slotted Anodes – The use of transversally slotted anodes helps to facilitate the removal of anode

gases from underneath the anode. This consequently lowers the overall anode polarization,

allowing for incremental improvements in productivity and cell amperage.

Primary Aluminum Production

5.6.1 Changes to the Hall-Héroult Cell and Alternative Technologies

Two innovative technological changes to the Hall-Héroult process, the wetted drained cathode and

the inert anode, are on the near-term horizon for improving energy efficiency. These technologies can,

with cell modifications, be retrofit into existing potlines and supporting infrastructure. Wetted

cathodes are anticipated to lower energy consumption of a Hall-Héroult cell by 18

percent when

compared to a modern Hall-Héroult cell. This report defines a modern cell as one that operates at 4.6

V(dc) and 95 percent current efficiency with the voltage distribution shown in Figure 5.7 (page 32).

The combination of an inert anode with a wetted cathode could provide a 22

percent reduction in

energy consumption and the elimination of cell CO

emissions. These technologies are described in

Section 6.0, Advanced Hall-Héroult Cell (page 43), and the energy impacts are calculated in

Appendix M. Multipolar cells using Hall-Héroult chemistry require the use of inert anode and wetted

cathode technologies. The multipolar design allows for a more compact, more productive cell with

significant thermal energy savings. Section 6.0 also describes multipolar electrolytic cells.

Two alternative technologies to the Hall-Héroult process, Carbothermic Reduction and Kaolinite

Reduction, have been studied by several groups for many years. These alternative technologies could

displace existing Hall-Héroult cells in the future. They are described in Section 7, Alternative

Primary Aluminum Processes (page 54). Both these processes could potentially change where and

how the aluminum industry operates, while lowering energy consumption. These alternatives

consume more carbon and have higher onsite carbon emissions than the Hall-Héroult process.

However, their electrical demands are lower which results in lower overall (utility-to-metal) CO

emissions. The carbothermic process is anticipated to save 20

percent in energy and be economical

at a much smaller scale than the Hall-Héroult facilities. The kaolinite reduction process is anticipated

to save about 12

percent of the energy required for a modern Hall-Héroult system. This value is

impacted by the need to prepare additional ore mass and carbon for the process. However, the

kaolinite reduction process is commercially interesting because of its lower onsite energy demands,

domestically available ore, and lower-cost raw materials.

Table 5.2 summarizes the estimated energy performance of these near and midterm technologies

(Appendix M, Table M.4). The onsite and tacit energy values in the table allow the processes to be

compared on a reaction, raw material or a complete ore-to-metal basis. The table provides the energy

associated with anode production and feedstock energies. The energy performance of the near-term

technologies, wetted cathode and inert anode, are based on voltage changes in the electrolysis cell

(Appendix M, Tables M.1 and M.2). These voltage changes are supported by theory and reported

experiments and provide a good estimate of energy use. The energy values reported for midterm

technologies, carbothermic reduction and kaolinite reduction, are approximations based on the

theoretical energy requirements and assumed reactor inefficiencies. Both midterm technologies

involve multiple reaction and separation zones. To date, no fully integrated reactor systems have been

built. These midterm energy approximations assume that there is significant heat integration

(recovery) within a facility.

Primary Aluminum Production

Table 5.2: Ore-to-Metal Comparison of Near and Mid-Term Technology Improvements

Energy Input kWh/kg Al

Modern Prebaked

Hall-Héroult

Wetted

Cathode

ACD = 2.0

Inert Anode

Wetted

Cathode

ACD = 2.0

Carbothermic

Reduction

Kaolinite

(AlCl

)

Reduction

ONSITE ENERGY DEMANDS

Raw Materi

als

Bauxite-Alumina

Kaolinite

Anode Materials

Reaction Carbon

Total

7.59

0.61

8.20

7.59

0.61

8.20

7.59

0.77

8.36

7.59

8.14

0.77

8.91

Reaction Energy

Reaction Thermal

Furnace Losses

Reaction

Cell Ohmic

Total Reaction

3.76

10.67

14.43

3.76

7.62

11.38

6.90

6.20

13.11

7.71

1.93

9.63

-1.90

0.40

6.48

2.93

7.91

TOTAL Onsite

Percent Energy

Savings:

22.63 19.58 21.46 17.22 16.82

Reactions

Reactions & anode

Reactions, anodes & ore

21%

20%

13%

33%

36%

24%

45%

42%

26%

TACIT ENERGY DEMANDS

Raw Materi

als

Bauxite-Alumina

Kaolinite

Anode Materials

Reaction Carbon

Total

8.24

6.02

14.26

8.24

6.02

14.26

8.24

0.77

9.00

8.24

8.43

16.67

8.84

0.77

11.37

20.97

Reaction En

ergy

Reaction Thermal

Furnace Losses

Reaction

Cell Ohmic

Total

8.41

23.82

32.23

8.41

17.01

25.41

15.41

13.86

29.27

17.21

4.30

21.52

-1.90

0.60

14.48

6.54

19.72

TOTAL Tacit

Percent Energy

Savings:

46.48 39.67 38.27 38.19 40.69

Reactions

Reactions & anode

Reactions, anodes & ore

21%

18%

15%

21%

18%

33%

44%

18%

39%

46%

12%

Many technical hurdles remain to be solved in these new processes before they become commercially

viable. Wetted cathodes and inert anodes will be adopted as their performances are proven and

existing cells need rebuilding. Industry will require significant demonstration time before adopting

any alternative to Hall-Héroult technology.

6. Advanced Hall-Héroult Cells

The Hall-Héroult cell system efficiency can be improved with the adoption of new cell technologies

described below. Wetted drained cathode technology offers the most significant improvement in

energy efficiency, but its adoption will be gradual. Typical cell life is seven to ten years. The industry

will optimize the capital invested in existing cells before retrofitting with new technology. Inert

anodes offer significant environmental benefits, lower maintenance costs, and can be retrofit into the

existing cells. Inert anodes could be adopted more quickly by industry since the carbon anodes are

replaced approximately every four weeks. However, their superior performance must still be proven

in industrial trials.

6.1 Wetted Drained Cathode

Wetted drained cathodes allow the anode-cathode distance to be reduced and are expected to

result in a 20(18

) percent reduction in the electrolysis energy required to produce aluminum.

Molten aluminum does not wet the carbon lining of a Hall-Héroult cell. The aluminum pad rests on an

extremely thin sheet of cryolite bath. This creates an electrical junction similar to the air gap between

two metal bars and causes a small voltage drop. If the bars are clamped tighter, providing better

contact, the junction voltage drop decreases. The thicker (heavier) the aluminum pad is, the thinner

the cryolite sheet becomes and the lower the junction voltage drop. Modifying the cathode surface to

make it more wettable would allow the same electrical contact with a decreased thickness in the pad.

A thinner pad would be more hydrodynamically stable, have lower wave height, and allow a decrease

in the anode-cathode distance (ACD). A decrease in the ACD results in energy savings. A cell lining

that is completely wetted and inert to cryolite would be even more efficient. This combination of

properties would allow the aluminum pad to be drained out of the anode-cathode spacing. Removing

the unstable aluminum pad would allow the ACD to be considerably reduced and provide significant

energy savings.

Titanium diboride (TiB

) has been found to be a wettable cathodic material, and several approaches to

incorporate TiB

into a Hall-Héroult cell are being studied.

Cathodes made wettable with TiB

appear to increase cell life by making the cathode less susceptible to penetration by bath material.

This is a considerable benefit, as cell rebuilding costs are a major contribution to primary aluminum

operational costs. Longer life cells also generate less cell lining waste material (spent pot liner) per

ton of aluminum produced. These benefits have to be balanced with the higher costs associated with

the TiB

material. Recent evidence also suggests that the wetted cathodes reduce the formation of

sludge (undissolved alumina) on the cell bottom and improve cell operations.

Advanced Hall-Héroult Cells

Several concepts for wetted cathode and draining cells have been proposed. Figure 6.1 shows:

a. a conventional cell,

b. a low metal pad with a wetted cathode in a conventional cell,

c. a hybrid cell with a metal sump, and

d. a fully drained slanted cell.

Figure 6.1: Conceptual Wetted Cathode Cells

The decrease in ACD is shown for each configuration. These configurations offer attractive

alternatives since they can be incorporated as a retrofit to the existing facilities. Some of these

configurations are currently being evaluated in commercial cells and could soon be available.

However, as ACD is decreased, less bath is available for circulation and mixing. This requires retrofit

cell designs to account for the change in bath dynamics. Designs must ensure that dissolved alumina

is available across the anode surface to maintain high current efficiency and productivity and to avoid

anode effects. Designs also need to compensate for the heat energy lost due to the lower cell voltage

operation.

Conceptually, the fully-drained inclined cathode design offers the greatest potential for energy

savings. By inclining the anode and cathode a few degrees, the molten metal pad (with all its

complexities) is removed completely to a sump. Without the pad, the anode-cathode-distance is

dimensionally stable and can be narrowed, which significantly reduces the total electrical resistance

of the bath. The alumina feeding of the bath is not compromised because the buoyancy of the gas

bubbles generated during reduction causes bath circulation and fresh alumina is drawn into the ACD

gap and across the electrode face.

Advanced Hall-Héroult Cells

6.1.1 Energy Savings for Wetted Cathode Technologies

Decreasing the anode-cathode distance (ACD) results in a proportional decrease in the voltage drop

associated with the electrolytic bath. Energy is saved when the ACD reduction is matched with the

ability to maintain current efficiency and heat balance. The energy savings for the different wetted

designs shown in Figure 6.1 are estimated in Appendix M, Table M.1.

The modern Hall-Héroult cell, shown in Figure 6.1 (a) has an ACD of 4.5 cm and a voltage

distribution similar to that shown in Figure 5.7 (page 32). The traditional cell voltage distribution has

1.75 V(dc) associated with the ACD which accounts for approximately 38 percent of the total 4.60

V(dc) drop across the cell. As discussed in Section 6.1, a wetted cathode provides better electrical

contact between the metal pad and the cathode allowing the ACD to be decreased. If the ACD were

lowered from 4.5 cm to 3.5 cm, as shown in Figure 6.1 (b), the voltage associated with the bath would

be proportionally lowered to 1.36 V(dc). The total voltage would decrease to 4.21 V(dc), which

would provide an 8 percent reduction in the electrical energy usage.

Draining the metal pad into a sump would eliminate the unevenness of the pad and permit an even

smaller ACD with greater energy savings. If the ACD were lowered from 4.5 cm to 2.5 cm, as shown

in Figure 6.1 (c), the voltage associated with the bath would be proportionally lowered to 0.97 V(dc).

The total voltage would decrease to 3.82 V(dc), providing a 16 percent reduction in the electrical

energy usage.

Reducing the ACD is limited by the ability to transport the reactant (dissolved alumina) to the

electrode interface and to remove products (aluminum and carbon dioxide) from the electrode

interface. Sloping the electrode interface slightly as shown in Figure 6.1 (d), removes products and

supplies reactants more effectively by using the buoyancy of the gas to induce bath circulation. It is

estimated that under the best conditions, the ACD for a sloped configuration could be reduced to as

little as 2.0 cm,

37, 38

in which case, the voltage associated with the resistance of the bath would be

0.78 V(dc). The total voltage required would then decrease to 3.63 V(dc), providing nearly a 20

percent reduction in the electrical energy usage.

Advanced Hall-Héroult Cells

Table 6.1: Energy Consumption Associated with Various Wetted Cathode Arrangements (kWh/kgAl)

Energy Input kWh/kg Al

Modern Prebaked

Hall-Héroult

ACD = 4.5

Wetted

Cathode

ACD = 3.5

Wetted

Cathode

ACD = 2.5

Wetted

Cathode

ACD = 2.0

ONSITE ENERGY DEMANDS

Raw

Mate

rials

bauxite-alumina

anode materials

TOTAL

7.59

0.61

8.20

7.59

0.61

8.20

7.59

0.61

8.20

7.59

0.61

8.20

Reacion

Energy

Reaction Electrolysis

Cell Ohmic

TOTAL Reaction

3.76

10.67

14.43

3.76

9.45

13.21

3.76

8.23

11.99

3.76

7.62

11.38

TOTAL Onsite

Percent Energy

Savings:

22.63

Reactions

Reactions & anode

Reactions, anodes & ore

21.41

20.19

17%

16%

11%

19.58

21%

20%

13%

TACIT ENERGY DEMANDS

Raw

Materi

als

bauxite-alumina

anode materials

TOTAL

8.24

6.02

14.26

8.24

6.02

14.26

8.24

6.02

14.26

8.24

6.02

14.26

Reaction

ergy

Reaction Electrolysis

Cell Ohmic

TOTAL Reaction

8.41

23.82

32.23

8.41

21.09

29.50

8.41

18.37

26.78

8.41

17.01

25.41

TOTAL Tacit

Percent Energy

Savings:

46.48

Reactions

Reactions & anode

Reactions, anodes & ore

43.76

41.03

17%

14%

12%

39.67

21%

18%

15%

Table 6.1 lists the data and measures the impact of the various wetted cathode arrangements.

Depending upon the cell design, the reduction energy impact is expected to be as high as 18

percent.

6.1.2 Environmental Impacts for Wetted Cathode Technologies

The byproducts generated in the Hall-Héroult process that are of environmental concern are grouped

into three areas: electrolysis, anode production, and cell waste products. Of these, the wetted cathode

technology impacts electrolysis and cell waste products.

Electrolysis

Electrolysis green house gas (GHG) emissions can be split into three groupings: reduction reaction

emissions, process upset perfluorocarbons emissions and hydrogen fluoride emissions from the bath.

Wetted cathode technology does not change the emissions related to process upsets or bath emissions.

The reduction reaction emissions come from three sources: the reaction products, anode air burning

Advanced Hall-Héroult Cells

and electricity generation and transmission. Wetted cathodes do not change the reaction products or

anode air burning. Both wetted and traditional cathodes will produce 1.22 kg of carbon dioxide from

the reduction reaction (2Al

+ 3C 4Al + 3CO

), 0.30 kg of carbon dioxide for each kilogram

of aluminum produced from air burning of the carbon anode, and 0.12 kg of carbon dioxide for each

kilogram of aluminum for the fuels associated with manufacturing the anodes.

The environmental benefit of wetted cathode technology is related to the emissions associated with

the electricity production. A wetted cathode lowers the electrical energy requirement, which in turn

reduces the emissions related to the fuels used in electricity generation and transmission. The

electricity production (14.4 kWh/kg of aluminum) for a modern Hall-Héroult cell emits 5.04 kg of

carbon dioxide equivalents (CDE) for each kilogram of aluminum produced. A wetted-sloped

cathode cell with a 2.0 cm ACD will lower the CDE emission associated with electricity generation

and transmission by nearly 21 percent to 3.98 kg CDE/kg of aluminum produced. This lowers the

total CDE emissions associated with a wetted-sloped cathode cell from 10.62 kg CDE/kg to 7.90 kg

CDE/kg of aluminum produced (Appendix E, Table E.4).

Cell Waste Products

One of the benefits of using wetted cathodes is longer cell life. Rebuilding cells less frequently will

lower the quantity of spent pot liner waste per unit of aluminum produced.

6.2 Hall-Héroult Inert Anode

Carbon anodes are consumed in the Hall-Héroult process, making the continuous manufacture of new

anodes and constant changing of the consumed anodes necessary. Anode changing upsets the

stability, production, and energy efficiency of the cell. For more than 100 years, Charles Hall and

other primary metal producers have attempted to find an inert anode that would eliminate the

manufacturing and handling of consumable anodes.

The material demands of an inert anode require

that it be highly conductive, and be thermally and mechanically stable at 800°C to 1,000°C. Further, it

should not react or dissolve to any significant extent in the cryolite, or react or corrode in the 800°C to

1000°C oxygen containing atmosphere. Any undesirable reaction with the bath must occur at a very

slow rate since this will result in the anode material contaminating the aluminum product.

Few materials are truly inert under the extreme conditions of a cell. Research has focused on three

classes of inert anode materials: ceramics, cermets, and metals. The challenges to finding the most

efficient material are substantial. Ceramics typically have poor thermal-shock resistance, are not

mechanically robust, display poor electrical conductivity, and are difficult to connect electrically.

Metals have good thermal, mechanical, and electric properties, but are attacked by the hot oxidizing

atmosphere. Metal oxides are somewhat soluble in cryolite and can resist the hot oxygen atmosphere,

but they exhibit lower electrical conductivity than metals. Once in solution, they electrochemically

Advanced Hall-Héroult Cells

reduce and contaminate the aluminum metal. Metal oxide solubility can be reduced dramatically in

cryolite by operating with a bath saturated in aluminum oxide, alumina. However, this presents

significant bath feeding challenges. Cermets combine the advantages and disadvantages of ceramics

and metals. In addition to overcoming technical hurdles, the likely higher-cost of manufacturing inert

anodes of commercial size must be compensated with longer life, lower energy consumption, and

higher productivity.

An inert anode would enable greater control of the critical anode-cathode distance (ACD), which

represents the largest voltage drop in the cell (Figure 5.7 on page 32). When used in conjunction with

a drained cathode, it is estimated that an inert anode may save up to 21

percent of the energy required

for reducing aluminum. Inert anodes offer a major environmental advantage and the potential of

producing a valuable coproduct. Replacing the carbon anode with an inert material results in oxygen

being discharged rather than carbon dioxide. If a significant market for oxygen is near the reduction

facility, the oxygen produced could be collected and sold as a coproduct. Carbon credits are an

unknown but potentially large economic force that could hasten the development of inert anodes.

Inert anode technology could potentially be retrofit into the existing cells with limited changes and

use the existing alumina and aluminum handling infrastructures. Some electrical infrastructure

changes would be required since the inert anode will operate at a higher voltage than the carbon

anodes. Since frequent access to the cells is not required for changing anodes, cells can be sealed

more effectively to provide better gas collection and treatments.

Notable progress in the production and testing of potential inert anode materials has been made in

recent years.

Some companies are now conducting trials with relatively stable materials that offer

the promise of inert anode performance. Using an inert anode and wetted cathode could also lead to

the design of multipolar, vertical electrode cells, which would increase productivity and further

reduce energy usage.

6.2.1 Theoretical Energy for Hall-Héroult Inert Anode

The theoretical minimum energy requirement for an

ideal Hall-Héroult cell using an inert anode can be

calculated from the thermodynamics of the reaction,

2Al

4Al + 3O

(Appendix J, Table J.2).

The production of aluminum shown in Figure 6.2

assumes that the reactant (alumina) enters the cell at

25°C, the oxygen byproduct leaves the cell at 25°C,

and the aluminum product leaves the cell as molten

metal at 960°C. The theoretical minimum energy

requirement is calculated under perfect conditions, an

Figure 6.2: Theoretical Energy Schematic for

environment with no reverse reactions, parasitic

Inert Anode

Advanced Hall-Héroult Cells

reactions, heat/energy losses external to the system, or limitations to the ionic species reacting at the

electrodes. The results show that the energy required to drive the reaction forward (∆G) is 8.16 kWh/

kg, the thermal energy (∆H – ∆G) required to maintain equilibrium is 0.48 kWh/kg and the thermal

energy (C

) associated with the molten aluminum is 0.39 kWh/kg of aluminum. The total theoretical

minimum energy requirement for an inert anode adds up to 9.03 kWh/kg of aluminum. (Note: If the

gas emission at 960°C is included, the total theoretical minimum energy requirement is 9.30 kWh/

kg of aluminum).

In actual cell operations, current efficiencies of less than 100 percent can result from reverse

oxidation reactions between part of the metallic aluminum and the oxygen gas produced by the

reaction. Inert anode cell designs must ensure that oxygen bubbles are discharged from the system

without contacting the molten aluminum being produced.

6.2.2 Comparative Energy Requirements for Carbon and Inert Anode Hall-Héroult Cells

The theoretical minimum energy requirement for an inert anode reaction is 51 percent higher than the

carbon anode requirement. Both, the Hall-Héroult carbon and inert anode processes, use the same

starting raw material, alumina, and require the same total theoretical energy, 9.03 kWh/kg of

aluminum. The Hall-Héroult carbon anode is consumed as part of the reduction reaction and provides

part of the energy to the system. The energy provided by oxidizing the carbon (3.04 kWh/kg of

aluminum) lowers the carbon anode system’s theoretical energy requirement to 5.99 kWh/kg of

aluminum. The inert anode must supply the full minimum energy requirement for alumina reduction.

The inert anode system requires 2.2 V(dc) compared to the Hall-Héroult reaction’s 1.2 V(dc).

This

increase in reaction voltage supplies the additional 3.04 kWh/kg of aluminum minimum energy

required by the inert anode system to reduce alumina.

Although the inert anode reaction voltage is 1 V(dc) higher than a carbon anode Hall-Héroult cell,

three factors are expected to provide the inert anode with an overall improved operational energy

performance than the carbon anodes:

a. elimination of carbon anode manufacturing,

b. reduction of anode polarization overvoltage, and

c. reduction in ACD that results from a stable anode surface.

The impacts of these three factors are shown in Table 6.2 and calculations are presented in Appendix

M, Table M.2.

Elimination of Carbon Anode

The energy inherent in the carbon and required for carbon anode manufacturing exceeds the

additional theoretical energy need for the inert anode system. A complete account of the energy

associated with the carbon-anode production shows that 6.02

kWh/kg of aluminum produced is

Advanced Hall-Héroult Cells

associated with the anode’s manufacture and fuel value. This value is higher than the 3.04 kWh/kg of

aluminum that the carbon anode supplies in the reaction for theoretical reduction. Since the actual

values for manufacturing inert anodes are unknown at this time, a conservative estimate is made in

Appendix M, Table M.3. This estimate, approximately 0.77 kWh/kg of aluminum, provides a basis

for comparing the two processes. The total energy associated with an inert anode operating cell is

0.77

kWh/kg for anode manufacturing, 3.14 kWh/kg for the onsite additional voltage requirement,

and 3.86

kWh/kg for the generation and transmission losses associated with the additional anode

voltage. The inert anode requires 7.77

kWh/kg of aluminum produced, 29

percent more than the

tacit energy associated with the Hall-Héroult carbon-anode requirement.

Table 6.2: Energy Impact of Inert Anode Technology

Energy Input kWh/kg Al

Modern

Prebaked Hall-

Héroult

ACD = 4.5

Inert Anode

a) Elimination

of Carbon

Anode

ACD = 4.5

Inert Anode

a) Elimination of

Carbon Anode

b) Low Polariza-

tion

ACD = 4.5

Inert Anode

a) Elimination of

Carbon Anode

b) Low Polarization

c) Wetted Cathode

ACD = 2.0

ONSITE ENERGY DEMANDS

Raw

Mate

rials

bauxite-alumina

anode materials

TOTAL

7.59

0.61

8.20

7.59

0.77

8.20

7.59

0.77

8.20

7.59

0.77

8.20

Reacion

Energy

Reaction Electrolysis

Cell Ohmic

TOTAL Reaction

3.76

10.67

14.43

6.90

10.67

17.57

6.90

9.25

16.15

6.90

6.20

13.11

TOTAL Onsite

Percent Energy

Savings:

22.63

Reactions

Reactions & anode

Reactions, anodes & ore

25.76

-22%

-14%

24.35

-12%

-13%

-8%

21.30

TACIT ENERGY DEMANDS

Raw

Materi

als

bauxite-alumina

anode materials

TOTAL

8.24

6.02

14.26

8.24

0.77

9.00

8.24

0.77

9.00

8.24

0.77

9.00

Reaction

ergy

Reaction Electrolysis

Cell Ohmic

TOTAL Reaction

8.41

23.82

32.23

15.41

23.82

39.23

15.41

20.67

36.08

15.41

13.86

29.27

TOTAL Tacit

Percent Energy

Savings:

46.48

Reactions

Reactions & anode

Reactions, anodes & ore

48.23

-22%

-5%

-4%

45.08

-12%

38.27

21%

18%

Advanced Hall-Héroult Cells

As gas molecules evolve on the anode surface, they coalesce and form bubbles, which creates a layer

of bubbles adjacent to the horizontal electrode. These bubbles jostle and move along as a result of

their density difference with the bath and by the motion imparted from the aluminum pad.

Experimental evidence shows that the oxygen gas evolved in the bath has a different froth/foam

dynamic than carbon dioxide. The molar volumes of gases that result from the reduction reaction are

equal. However, because oxygen is 27 percent lighter than carbon dioxide, less mass moves through

the electrode area. The physical properties, in practice, contribute to a lower anode overvoltage.

The

inert anode can also be shaped to allow for better release of the oxygen generated (Figure 6.3). The

lower overvoltage of the inert anode, compared to a carbon anode, reduces the energy consumed. The

overvoltage reduction combined with the elimination of carbon anode manufacturing, is estimated to

provide a 5 percent energy improvement.

Figure 6.3:Inert Anode

Inert Anode-Wetted Cathode Combination

A more stable inert anode surface will allow the anode-cathode distance (ACD) to be reduced and

will be simpler to control. Carbon anodes must be lowered as they are consumed to maintain an

optimal ACD. Combined with a wetted cathode, the inert anode probably offers the greatest potential

for bath voltage reduction. Inert anodes will also eliminate cell disruptions that occur with carbon

anode changing.

Overall, inert anodes, when combined with a wetted cathode and compared to the traditional Hall-

Héroult cells, are expected to provide

• a 10 percent reduction in operating costs (elimination of carbon anode plant and labor costs

associated with replacing anodes),

• a 5 percent increase in cell productivity,

and

• a 41 percent reduction in greenhouse gas emissions (Appendix E, Table E.4)

Advanced Hall-Héroult Cells

These practical scenarios provide the inert-anode cell with an overall lower-energy requirement than

the state-of-the-art Hall-Héroult cell. However, the challenging engineering designs for the inert-

anode cell systems must incorporate effective approaches for minimizing thermal losses from the

reduction cells, the current-carrying bus systems, and the connectors external to the cell.

6.3 Multipolar Cells

Present Hall-Héroult industrial cells consist of a single cathode surface and essentially a single anode

surface immersed horizontally one over the other in a bath. This single-pole arrangement makes

aluminum reduction a capital-intensive process. Use of multipolar electrolytic cells would greatly

increase productivity-per-unit-reactor volume by placing multiple electrodes in a single reactor. This

arrangement would also provide better control of heat losses. For practical engineering reasons, a

multipolar cell requires that the ACD be stable, which in turn requires an inert/stable anode. To date,

the lack of suitable materials of construction have ruled out multipolar cells using molten fluoride

electrolytes.

Two variations of electrochemical multipolar cells are possible, as shown conceptually in Figure 6.4.

One system uses a cell with an anodic surface on one side, multiple bipolar electrodes in the center

portion, and a cathodic surface on the opposite side, as in Figure 6.4 (a). One surface of a bipolar-

electrode plate acts anodically, while the opposite surface acts cathodically. Bipolar electrodes must

be electrically isolated and sealed to the cell walls to avoid bypass current efficiency losses.The

second system uses independent pairs of electrodes immersed in the same cell, as shown in Figure 6.4

(b). Multipolar cell designs are complicated by the need to remove liquid metal and collect the

gaseous reaction products.

Figure 6.4:Multipolar Cells

(a) (b)

The bipolar-multipolar concept was successfully demonstrated for electrolytic aluminum chloride

reduction.

In 1976, Alcoa Inc. began operating a multipolar, prototype plant with a capacity of over

18,000 kg of aluminum per day. Alcoa obtained 90 percent current efficiency at 9.26 kWh/kg of

aluminum. The Alcoa multipolar cell was electrically 40 percent more efficient than a present day

Hall-Héroult cell. However, the prototype plant was eventually shut down because of the combination

Advanced Hall-Héroult Cells

of the costs to produce anhydrous aluminum chloride feed, the failure to reach full design capacity,

the need to remove and destroy trace amounts of chlorinated biphenal byproducts, the reactor capital

costs, and general maintenance costs.

The Alcoa cell had distinct technical advantages over the classic Hall-Héroult cell: it worked at

substantially lower temperatures, it had relatively high current densities, its carbon anodes were not

consumed, it had no fluoride emissions, and the plant had a smaller footprint. The improvement in

electrical energy efficiency was a result of the higher electrical conductivity of the electrolyte and the

small interpolar distance. The cell feed consisted of 3 percent to 10 percent of purified aluminum

chloride (AlCl

), obtained by carbochlorination. The electrolyte consisted mainly of sodium chloride

and potassium chloride, or lithium chloride. Electrolysis was performed in a sealed cell consisting of

12 to 30 bipolar carbon electrodes stacked vertically at an interpolar distance of 1 cm. The electrodes

remained immersed in the electrolyte at an operating temperature of 700°C ± 30°C. A current density

of 0.8 A/cm

to 2.3 A/cm

and a single-cell voltage of 2.7 V(dc) were typical operating conditions. An

operating life of nearly three years was claimed for the electrodes.

A useful feature of the Alcoa cell was that the buoyant chlorine created a flow in the narrow gap

between the bipolar electrodes. This aided in sweeping the aluminum from the cathodes and

enhancing the formation of aluminum droplets. The gas also helped maintain a continuous flow of

electrolyte across the cell. Chlorine was collected at the top of the cell and molten aluminum was

collected in the bottom. The chlorine gas generated by the reduction reaction was completely recycled

to produce new aluminum chloride feed for the cell.

A Hall-Héroult multipolar design, involving multiple inert anodes and wettable cathodes arrayed

vertically in a fluoride electrolyte cell, has been explored by Northwest Aluminum.

In this concept,

the operating temperature was 750°C, much lower than Hall-Héroult technology. Alumina saturation

at the lower temperature of the electrolyte bath was maintained by controlling a suspension of fine

alumina particles in the bath. This technology offered all the benefits of reduced energy consumption

(low ACD), elimination of carbon anodes and associated emissions, as well as a significant increase

in productivity per cell, as a result of the multipolar design. This work came to an unfortunate halt

with the closure of Northwest Aluminum. However, recent multipolar cell work done at Argonne

National Laboratory shows much promise, although it is still in the early stages.

7. Alternative Primary Aluminum Processes

Carbothermic reduction of alumina and chloride reduction of kaolinite clay have been under study for

more than 40 years. These technologies have the potential to be alternative processes to the Hall-

Héroult process. The non-electrolytic, carbothermic reduction of alumina is estimated to produce

aluminum with significantly less energy consumption, reduced emissions, and lower investment

costs. Multipolar-reduction technology to produce aluminum using aluminum chloride obtained from

low-grade domestically available clays is also estimated to reduce energy consumption, capital costs,

and emissions. The aluminum industry has yet to adopt these processes for producing aluminum,

mainly because of the uncertain capital and operating economics associated with large-scale plants.

7.1 Carbothermic Technology

Carbothermic reduction of alumina is the only non-electrochemical process that has shown potential

for aluminum production. This technology has been an industry objective and the subject of extensive

research for more than forty years.

Carbothermic technology is projected to produce aluminum from

ore at 38.2

kWh/kg or 18 percent below a modern Hall-Héroult carbon-anode technology (Table 7.1

on page 57).

Carbothermic reduction produces aluminum using a

chemical reaction that takes place within the volume

Volumetric Vs. Area Processing

of a reactor. This volumetric reaction requires much

less physical space than the area reaction of Hall-

Héroult. These properties give carbothermic

technology a smaller footprint and make it much less

The Hall-Héroult reduction electrolytic

reaction occurs on the surface of an

electrode in a reactor (cell). Capacity is

doubled simply by doubling the electrode

dependent on the economies of scale (large and long

surface area. Non-electrolytic reduction

potlines) required for an economically efficient Hall-

chemistry occurs within a volume of fluid.

Héroult facility. If successful, this technology could

A spherical volumetric reactor can be

significantly change the structure of the aluminum

industry, allowing industry the freedom to relocate

from regions of inexpensive power to centers of

doubled by increasing its radius 1.4 times.

This difference between volume and

surface area reactions provides significant

savings in capital cost for a reactor.

manufacturing. Aluminum production “mini-mills”

could be placed adjacent to or within aluminum

casting facilities. These “mini-mill” installations could provide molten aluminum and/or specifically

alloyed metal directly to the casting operations. This would provide additional energy, economic, and

environmental benefits to the industry.

Alternative Primary Aluminum Processes

Figure 7.1: Carbothermic Reactor

Carbothermic reduction of alumina to

aluminum is a multi-step chemical reaction

process. The thermodynamic optimization

for the reactions requires a multi-zone

furnace operating at very high temperatures

(Figure 7.1). In the first stage net reaction,

alumina and carbon form an alumina-

aluminum carbide slag at ~1,900°C

(2Al

+ 9C Al

+ 6CO). In

the next stage net reaction, aluminum

carbide is reduced by alumina to form

aluminum metal at ~2,000 °C,

(Al

+ Al

6Al + 3CO

). The

thermodynamics (temperatures and chemical equilibria) of these reactions are very complex. A

significant portion of aluminum evolves as gas phase components (Al and Al

O) at these operating

temperatures. Careful process control is necessary to minimize the generation of volatiles. Recovery

of these components in the form of Al

in a vapor recovery system is required for the process to be

economically viable. If the Al and Al

O back react with CO to form Al

then the productivity of

the process is decreased and the Al

required to satisfy the process stoichiometry is deficient.

The complex thermodynamic controls, sophisticated equipment, and construction materials required

to successfully develop an economical commercial system have eluded the industry so far. Current

R&D efforts are reevaluating carbothermic technology in hopes of capitalizing on new, advanced,

high-intensity, electric-arc furnace technology, advanced thermodynamic and system modeling

techniques, and an improved understanding of the process dynamics.

7.1.1 Theoretical Energy for Carbothermic Reduction of Alumina

The theoretical minimum energy requirement for producing aluminum by the carbothermic

reduction of alumina is 7.32 kWh/kg.

Carbon is a reactant in the carbothermic reduction reaction process and supplies part of the energy

necessary to drive the reaction forward. This gives the carbothermic reduction process a lower

theoretical energy requirement than the direct electrolytic reduction of alumina to aluminum.

Alternative Primary Aluminum Processes

The net reaction for the carbothermic reduction is

+ 3C 2Al + 3CO. It is assumed that

the reactants (alumina and carbon) enter the reactor

system boundaries at 25°C, the carbon monoxide

byproduct leaves at 25°C, and the aluminum

product leaves the reactor system as molten metal at

960°C. The assumed molten metal temperature is

lower than the actual reactor metal discharge

temperature. Theoretically and practically, the

Figure 7.2: Carbothermic Theoretical Minimum

Energy

energy in the higher temperature reactor discharge

can be recovered efficiently (e.g., by mixing high temperature reactor aluminum with solid metal

scrap to recover the heat energy and lower the temperature). To compare the theoretical limits of the

various aluminum production processes, the same molten product temperature of 960°C is used

throughout this report. The theoretical reaction occurs under perfect conditions when there are no

reverse reactions, parasitic reactions, or heat/energy losses external to the system.

The calculation of theoretical minimum energy requirement for the carbothermic reaction is detailed

in Table J.3, Appendix J. The results show that the energy required to drive the reaction forward (∆G)

is 6.03 kWh/kg and this energy is supplied as thermal energy vs. the electrical energy used in a Hall-

Héroult cell. The thermal energy (∆H – ∆G) required to maintain equilibrium is 0.90 kWh/kg and the

thermal energy (C

) associated with the molten aluminum is 0.39 kWh/ kg of aluminum. The

theoretical minimum energy requirement under these conditions adds up to 7.32 kWh/kg of

aluminum. (Note: If the CO gas emission at 960°C is included, the total theoretical minimum energy

requirement is 7.51 kWh/kg of aluminum).

The carbon monoxide byproduct has a fuel value and would likely be captured and used to supply

thermal energy to the carbothermic facility.

7.1.2 Comparative Benefits for Carbothermic Reactors and Hall-Héroult Cells

Electric furnace technology provides 90 percent thermal efficiency in many applications. Heat losses

are limited to conduction and radiation losses from the furnace shell. Flue gas heat losses are

eliminated, as are any undesired reactions between flue gases and molten aluminum metal. It is

reasonable to assume that the carbothermic furnace/reactor can be designed with more than 85

percent thermal efficiency. If the thermodynamics of the reaction and offgas recovery can be

controlled within 95 percent of the theoretical requirements, the electrical energy required for an

operating carbothermic reactor would be (7.32 kWh/kg ÷ (0.85 x 0.95)) or 9.07 kWh/kg of aluminum

produced. This represents a 37 percent reduction in energy use when compared to the 14.4 kWh/kg of

aluminum from a modern Hall-Héroult Cell. Table 7.1 (page 57) shows the onsite and tacit energy

savings potential of carbothermic technology.

Alternative Primary Aluminum Processes

In addition to electrical energy savings, carbothermic technology is expected to provide other

benefits, like

• a reduction in capital costs by 50 percent or more as a result of volumetric processing through

high-intensity smelting;

• a reduction in production costs by 25 percent through lower electrical demand and elimination of

carbon anode manufacturing and handling,

• the production of a high purity CO/CO

stream for coproduct sale or energy integration,

• the potential to use small blocks of electrical power due to high turn-down ratio, and

• the potential of widely locating “mini-mills” with integral, captive smelters delivering molten

metal.

Table 7.1: Comparison of Hall-Héroult and Carbothermic Reduction

Energy Input kWh/kg Al

Modern Prebaked

Hall-Héroult

Carbothermic

Reduction

ONSITE ENERGY DEMANDS

Raw

Materi

als

Bauxite-Alumina

Kaolinite

Anode Materials

Reaction Carbon

Total

7.59

0.61

8.20

7.59

Reaction

Energy

Reaction Thermal

Furnace Losses

Reaction

Cell Ohmic

Total Reaction

3.76

10.67

14.43

7.71

1.93

9.63

TOTAL Onsite

Percent Energy

Savings:

22.63 17.22

Reactions

Reactions & anode

Reactions, anodes & ore

33%

36%

24%

TACIT ENERGY DEMANDS

Raw

Materi

als

Bauxite-Alumina

Kaolinite

Anode Materials

Reaction Carbon

Total

8.24

6.02

14.26

8.24

8.43

16.67

Reaction

ergy

Reaction Thermal

Furnace Losses

Reaction

Cell Ohmic

Total

8.41

23.82

32.23

17.21

4.30

21.52

TOTAL Tacit

Percent Energy

Savings:

46.48 38.19

Reactions

Reactions & anode

Reactions, anodes & ore

33%

44%

18%

Alternative Primary Aluminum Processes

7.1.3 Environmental Impacts of Carbothermic Technology

The carbothermic process, when compared to Hall-Héroult technology, results in significantly

reduced electrical consumption, and the elimination of perfluorocarbon emissions that result from

carbon anode effects, hazardous spent pot liners, and hydrocarbon emissions associated with the

baking of consumable carbon anodes.

The total carbon dioxide emissions from carbothermic reduction, as with Hall-Héroult, depend on the

source of electricity. Hydroelectric power generation emits almost no carbon dioxide, whereas the

carbon dioxide emissions associated with the average U.S. grid electricity are 0.51 kg CDE/kWh

(Appendix E, Table E.1). Since 40 percent of the U.S. primary industry operates on hydroelectric

power, the aluminum industry’s average electrical generation emission rate is 0.35 kg CDE/kWh. The

carbothermic reaction results in the generation of carbon-based greenhouse gases (GHG), mainly

carbon monoxide (CO), at twice the rate of the Hall-Héroult reaction. However, the carbothermic

process only requires electricity for heating and not for the reduction reaction. Assuming

carbothermic technology is used in “mini-mills” operating off the average U.S. electric grid, the total

GHG emissions from “utility-to-metal” for the carbothermic process are reduced relative to the

average U.S. Hall-Héroult system, because of carbothermic’s lower electrical intensity. Table 7.2

(Appendix E, Table E.4) shows the lower electrical demand of carbothermic technology resulting in

lower total carbon dioxide equivalent (CDE) emissions than for a Hall-Héroult system.

Table 7.2: Carbon Dioxide Equivalent Comparison of Hall-Héroult and Carbothermic Reduction

Emission Sources Modern Hall-Héroult Carbothermic

kg CDE/kg Al kg CDE/kg Al

Carbon Anode 1.66 2.45

Reaction Energy

Requirements:

kWh/kg Al

kg CDE/kWh

14.43

0.349

5.04

9.63

.511

4.92

Process 2.20 0.0

Total 8.91 7.37

7.2 Kaolinite Reduction Technology

Production of pure aluminum by reduction of aluminum chloride was discovered before the Hall-

Héroult process, in 1825. Alumina conversion to aluminum chloride and reduction to aluminum using

bipolar technology was demonstrated in the late 1970s, but it was not commercialized because of

problems with the product purity and projected high capital and operating costs (Section 6.3 on

Alternative Primary Aluminum Processes

page 52). New construction materials, improved thermodynamic understanding, and the potential to

use low-cost alumina containing clays have maintained the interest in chloride reduction technology

for producing aluminum.

Compared to the current Bayer refining and Hall-Héroult fluoride-based smelting of alumina, the

chlorination of alumina-containing clays promises many potential advantages:

• Raw materials are widely available, inexpensive, and indigenous to the United States.

• Thermodynamics provide high-speed, high-conversion reactions with lower electrical demand.

• No “bauxite residue” is produced, though there is waste from the clay chlorination.

• Conventional materials-of-construction (i.e., mild steel) can be used.

Industry has spent more than 30 years developing process technologies for the chlorination of widely

available, low-grade kaolin clays. These clays contain kaolinite (hydrated alumina silicate, i.e.,

•

2SiO

•

), significant amounts of titanium dioxide (1 percent to 5 percent), and other

materials. Titanium tetrachloride and other metal chloride byproducts are also produced when

processing kaolin clays.

The basic steps of the clay to aluminum process are shown in Figure 7.3.

Figure 7.3:Clay to Aluminum Process Schematic

First, the kaolin clay and process coke are dried. All feed materials to the clay chlorination units must

be dried to minimize chlorine absorption in any water and to reduce the corrosive effects of moisture

in the chlorination offgas stream. The drying process involves controlled, catalyzed heating of the

finely ground clay at ~800°C in a fluidized reactor, using coke and air to provide the heating energy.

The hot reactor produces a dehydrated calcined clay, which is sent to carbo-chlorination.

The carbo-chlorination step is a high-speed, catalyzed exothermic reaction of calcined clay with

chlorine and coke:

Alternative Primary Aluminum Processes

3(Al

•2SiO

) + 14C + 21Cl

6AlCl

+ 6SiCl

+ 7CO + 7CO

Chlorine is injected into the bottom of the fluidized reactor with the clay oxides and coke. The

metallic oxides, principally those of aluminum and silicon, are converted to their chlorides. The

chlorides exist as vapors at the reaction temperatures. The next process step is designed to suppress

the formation of silicon tetrachloride. The offgases of the chlorination reactor are treated with a small

amount of catalyst vapor and reacted in a second fluid bed reactor with additional calcined clay. The

alumina portion of the clay is converted to aluminum chloride, and silicon tetrachloride is converted

to solid silica and discharged (

3SiCl

+ 2Al

4AlCl

+ 3SiO

). The net effect of this step

is the preferential chlorination of alumina relative to silica. The hot vapors from the second reactor

are cooled and crude aluminum chloride is recovered. Several byproducts are produced as an

extension of the chlorination step. Titanium tetrachloride, silicon tetrachloride, and boron trichloride

are not condensed in the cooler and can be recovered in subsequent processing. Silicon tetrachloride

can be reacted with oxygen to recover chlorine, which is recycled back to the chlorination step as

below:

4SiCl

+ 5O

2SiO

+ 3O

+ 8Cl

The crude aluminum chloride must be purified for ease of operation of the electrolytic reduction cells

and for the final aluminum quality. The main impurity is iron, which is present as ferric chloride at

levels as high as ~50,000 ppm. Impurities are removed by chemical treatment.

The electrolysis of aluminum chloride to aluminum takes place in an aluminum chloride smelting

cell, which comprises a stack of horizontal bipolar graphite electrodes, between which the aluminum

chloride is converted into high-grade aluminum and chlorine gas

(2AlCl

2Al + 3Cl

)

. The

electrodes are immersed in a chloride bath, which is contained in fully enclosed, thermally lined

vessels. The bipolar electrodes are supported and separated by inert spacers resting on the electrode

below it, and the entire stack, consisting of bottom cathode, bipolar electrodes, and top anode, is

supported by the walls of the cell.

7.2.1 Theoretical Energy for Kaolinite Reduction of Alumina

The theoretical minimum energy requirement for producing aluminum from kaolinite is 5.76

kWh/kg of aluminum produced. The theoretical minimum for the chloride reduction step of

aluminum chloride is 7.66 kWh/kg of aluminum.

The theoretical minimum energy requirement can be calculated from the net chemical reaction in the

kaolinite to aluminum process:

Alternative Primary Aluminum Processes

7(Al

•2SiO

) + 14C 14Al + 14SiO

+ 7CO + 7CO

It is assumed that the reactants (kaolinite and carbon) enter the system boundaries at 25°C, the carbon

monoxide and carbon dioxide byproducts leave at 25°C, and the aluminum product leaves the system

as molten metal at 960°C. The theoretical net reaction occurs under perfect conditions when there are

no reverse reactions, parasitic reactions, or heat/energy losses external to the system. These

assumptions yield a theoretical minimum energy requirement of aluminum production from kaolinite

as 5.76 kWh/kg of aluminum (Appendix J, Table J.5). This assumes a pure kaolinite feedstock.

The kaolinite process is accomplished in two steps, carbo-chlorination and aluminum chloride

reduction. The electrolytic aluminum chloride reduction process used in the kaolinite process requires

the exact same minimum amperage (2,980 Ah/kg of aluminum) as any electrolytic process for

reducing aluminum. The theoretical chloride reduction reaction occurs under perfect conditions

(where there are no reverse reactions, parasitic reactions, or heat/energy losses external to the system)

and requires 7.66 kWh/kg of aluminum produced. The theoretical minimum energy requirement used

in this report is calculated at 960°C to provide comparison with the other processes. The proposed

multipolar system operates at about 700°C, which would lower the theoretical energy demand by

about 0.09 kWh/kg of aluminum. The exothermic nature of the carbo-chlorination reaction

(–1.90 kWh/kg of aluminum) results in the overall kaolinite to aluminum theoretical energy

requirement to be lesser than the minumum energy requirement for the aluminum chloride reduction

Alternative Primary Aluminum Processes

7.2.2 Comparative Benefits for Kaolinite Reduction and Hall-Héroult Cells

The kaolinite reduction process offers potential advantages when compared to the Hall-Héroult

process. Table 7.3 shows the onsite and tacit energy comparison for Hall-Héroult and Kaolinite

processes.

Table 7.3: Comparison of Hall-Héroult and Kaolinite Reduction

Energy Input kWh/kg Al

Modern Prebaked

Hall-Héroult

Kaolinite (AlCl

)

Reduction

ONSITE ENERGY DEMANDS

Raw Materi

als

Bauxite-Alumina

Kaolinite

Anode Materials

Reaction Carbon

Total

7.59

0.61

8.20

8.14

0.77

8.91

Reaction Energy

Reaction Thermal

Furnace Losses

Reaction

Cell Ohmic

Total Reaction

3.76

10.67

14.43

-1.90

0.40

6.48

2.93

7.91

TOTAL Onsite

Percent Energy

Savings:

22.63 16.82

Reactions

Reactions & anode

Reactions, anodes & ore

45%

42%

26%

TACIT ENERGY DEMANDS

Raw Materi

als

Bauxite-Alumina

Kaolinite

Anode Materials

Reaction Carbon

Total

8.24

6.02

14.26

8.84

0.77

11.37

20.97

Reaction En

ergy

Reaction Thermal

Furnace Losses

Reaction

Cell Ohmic

Total

8.41

23.82

32.23

-1.90

0.60

14.48

6.54

19.72

TOTAL Tacit

Percent Energy

Savings:

46.48 40.69

Reactions

Reactions & anode

Reactions, anodes & ore

39%

46%

12%

The combined clay carbo-chlorination and bipolar aluminum chloride smelting process is estimated

to be 35

percent more tacit energy efficient than the Hall-Héroult, has a smaller plant footprint, can

be more flexible in regard to use of off-peak power and power fluctuations, and produces fewer

emissions and process wastes. Onsite energy use is about 19 percent lower than a modern Hall-

Alternative Primary Aluminum Processes

Héroult cell. The large tacit energy improvement results from a decrease in electrical use. Compared

to the current Hall-Héroult smelting technology, the bipolar aluminum chloride smelting process

consumes less electricity, yielding 60 percent more metal for the same electrical input. Bipolar cell

designs provide significantly lower reactor volume per unit of product output. This lower volume

allows the cell to idle and hold temperature much more efficiently than a single electrode cell. These

properties allow multipolar cells to take better advantage of off-peak electrical costs. Chloride cells

also operate at lower temperatures, providing additional savings.

Raw materials used in the kaolinite process require approximately 47 percent more energy compared

to those in a modern Hall-Héroult process (Table 7.3). Kaolinite requires more energy for extraction

on an aluminum basis and the carbo-chlorination step requires 2.5 times the carbon of a Hall-Héroult

carbon anode system. The raw material energy expenses may be compensated for by the lower-cost

domestic supply of kaolinite and the ability to use lower-cost carbon than Hall-Héroult systems.

It is important to emphasize that while significant elements of the kaolinite process have been studied

and developed, no integrated production of aluminum from kaolinite clays has yet been attempted.

Significant work in developing an integrated production process has been reported.

7.2.3 Environmental Impacts of Kaolinite Technology

Table E.4 in Appendix E tabulates the CDE emissions for the kaolinite to aluminum process.

Operating off the average U.S. electrical grid, this process would produce 14 percent fewer CDE

emissions than a typical Hall-Héroult facility. However, the same plant operating on the average U.S.

smelter grid (50.2 percent hydroelectric) would actually result in 8 percent lower CDE emissions than

Hall-Héroult facilities.

8. Secondary Aluminum (Recycling)

The production of primary aluminum ingots from bauxite ore requires approximately 23.8(62.2

)

kWh/kg of aluminum. Recovering aluminum from scrap to produce secondary aluminum ingot

consumes about 6 percent of the energy required to produce primary aluminum.

This significant

energy difference drives the emphasis placed on aluminum recycling in today’s society and in the

aluminum industry.

Recycling in the United States saved more than 167 x 10

kilowatt hours (0.57 quad) of energy in

2003, the equivalent of 19,100 Megawatts. Each kilogram of aluminum that is recovered by recycling

saves 59.4

kWh of the 62.2

kWh of energy consumed in producing a primary aluminum ingot from

bauxite ore (Appendix F, Table F.6). Any process that improves the recovery of scrap aluminum is

effectively making an order of magnitude change in the energy associated with aluminum production.

The growth of aluminum recycling represents the greatest change in the structure of the industry and

in the energy associated with aluminum manufacturing. A common practice since the early 1900s,

recycling was a low-profile activity until 1968 when aluminum beverage can recycling vaulted the

industry into public consciousness. In 1960, recycled aluminum accounted for 18 percent of the

nation’s total aluminum supply (401,000 metric tons). Over the next 43 years, production of recycled

aluminum rose by 703 percent to 2,820,000 metric tons. During those same 43 years, the total U.S.

aluminum metal supply increased 300 percent (Appendix G). In 2003, 75 plants in the United States

produced secondary ingots. Over half – 51 percent – of the aluminum metal produced in the United

States in 2003 was from recycled material.

The growth of the market for recycled aluminum is due in large measure to economics. It is cheaper,

faster, and more energy-efficient to recycle aluminum than to manufacture it from ore. Recovered

aluminum is easily melted at relatively low temperatures (aluminum alloys typically melt at

temperatures below 660 ºC). Producing a recycled aluminum ingot consumes only about 6 percent of

the energy required to produce a primary aluminum ingot from bauxite ore. In addition, to achieve a

given output of ingot, recycled aluminum requires only about 10 percent of the capital equipment

costs compared with those required for the production of primary aluminum.

Aluminum products are corrosion resistant which allows them to be easily and repeatedly recycled

into new products. The corrosion resistance is due to the metal’s properties. When the surface of

aluminum is exposed to air, it rapidly forms a tenacious, self-limiting, protective oxide layer. Other

surface treatments can be applied to further enhance aluminum’s corrosion resistance.

Secondary Aluminum (Recycling)

Figure 8.1 shows the annual growth rates of the three sources of metal supply between 1993 and 2003

period, together with the growth in the major aluminum product markets. Aluminum imports are

growing at at a rate of 5.1 percent per year while both U.S. primary and secondary productions are in

decline (Appendix G).

Figure 8.1:U.S. Aluminum Market and Growth

Aluminum scrap is categorized as “new” or “old.” “New” scrap is generated when aluminum

products are manufactured. It includes defective products; scalping chips; edge and end trim from

rolling processes; skeleton scrap from stamping and blanking operations; flash, gates, and risers;

extrusion butts and ends; and turnings and borings. “Old” scrap (post consumer or obsolete products)

comes from discarded, used, worn-out, or out-of-date products that include automotive parts, white

good parts, containers such as used beverage cans and closures, wires, cables, and building materials.

“Runaround scrap” is “new” scrap that is recycled by the same company that generated it. Since

runaround is usually not sold or marketed, it is not reported in the U.S. recycling statistics.

8.1 Secondary Aluminum Production

Secondary aluminum producers represent a separate and vital segment of the aluminum industry

whose principal activities are converting purchased scrap, and metal recovered from skim and dross

generated in molten metal operations into usable aluminum alloy products. Recycling in primary

aluminum operations is typically confined to in-house or runaround scrap and manufacturing scrap

returned directly from their customers. Secondary aluminum producers specialize in melting and

processing a wide range of new and old, segregated and mixed, high and low-quality scrap.

Secondary Aluminum (Recycling)

The most desirable form of recycling is closed-loop in which scrap from specific product applications

are returned for remanufacturing of the same products. Rigid container stock used in beverage cans is

an example of closed-loop recycling since used beverage cans are processed exclusively to

remanufacture can sheet. Secondary aluminum producers are involved in closed-loop recycling as

contractors to primary producers for this and other products.

Scrap segregated by alloy has greater value than mixed scrap because it can be used predictably and

most efficiently in the production of the highest value-added compatible compositions. Mixed scrap

presents the greatest challenge, generally requiring added steps in melting, composition

identification, and often casting into ingot form before consumption.

Casting alloys are the largest and the most important, but not the only market for secondary

aluminum. Scrap is used, with primary metal, in the production of extrusion billet and fabricating

ingot. Reclaimed smelter ingot (RSI) is produced from scrap and dross, often on a toll conversion

basis.

Alloys used by aluminum foundries in the production of shape or engineered castings include

compositions designed to facilitate production from scrap. These alloys, which have been historically

prominent, typically specify broader element ranges and higher impurity limits than the alloys

developed for more specialized purposes and whose compositions require production from primary

metal sources. The increased use of aluminum in structural applications in the ground transportation

sector typically requires primary compositions, but the expanded use of aluminum in engine blocks,

cylinder heads, and other power train parts relies on castings in secondary alloys. The emphasis on

light weighting for improved fuel efficiency and reduced environmental degradation is encouraging

the adoption of various wrought aluminum products such as auto body sheet, fuel tanks, seat backs,

extruded drive shafts and stringers, and forged connecting rods. As the proportion of aluminum

increases in ground transportation designs, segregating and blending scrap into saleable alloy

compositions will become common practice.

New scrap sorting technology is developing, including those related to the recovery of aluminum

from the large transportation and white goods markets.

New technologies with computer screening

are using color segregation and laser-induced-breakdown spectroscopy (LIBS) to sort wrought from

cast and, in some instances, one alloy from another. Aluminum alloy separations will allow scrap to

be segregated into more specific alloy groupings with higher economic value.

8.2 Production, Capacity and Growth

Aluminum scrap is widely recycled and supports a large secondary aluminum industry. In 2003, the

United States produced 2,820,000 metric tons of secondary aluminum, amounting to nearly one-third

of its total supply of aluminum. This market had an annual growth rate of –0.4 percent over the last

ten years (Figure 8.2).

1960

1965

1970

1975

1980

1985

1990

1995

2000

Secondary Aluminum (Recycling)

Figure 8.2:U.S. Production of Secondary Aluminum 1960 to 2003

Million Metric Ton

Source: Appendix G

Market demand for recycled aluminum will remain strong due to its inherent low-energy cost relative

to primary metal. Aluminum use in consumer products has become widespread in a trend described as

the “urban mine.” Challenges in scrap recovery, alloy sorting and impurity removal can be addressed

with the current technologies. The limitation to secondary metal market growth is the economic

supply of scrap.

The urban mine has been accumulating scrap products for the past hundred years. There is a lag time

from when aluminum leaves the shipping dock to when it becomes available for recycling. This lag

time, or “use” phase life, varies significantly among aluminum products. For packaging products such

as beverage cans, this lag is only about 65 days. Long-life products, such as building frames, can have

a lag time of 40 to 50 years. Automobile lag time is 12 to 15 years. Aluminum in automobiles

accounted for 5 to 10 percent of scrapped automobiles’ weight in 2000, but represents 35 to 50

percent of its scrap value. This source will grow rapidly as new automobile designs continue to utilize

a greater proportion of aluminum.

Considering the large amounts of aluminum that are stored in long-life products (accumulated within

the “urban mine”) and the continued growth in aluminum demand, recycling will continue to increase

and be a significant contributor to the U.S. metal supply. In fact, recycling has overtaken primary

production as the main source of domestically produced aluminum in the United States.

8.3 Recycling Processes

The objectives of the recycling process are maximum metal recovery, minimum contamination, and

lowest conversion cost. Safety is of prime importance since components mixed with scrap can present

the risk of explosive reactions and other concerns. Since moisture is a safety concern, remelt ingot

and reclaimed smelter ingot (RSI) are routinely preheated before charging to the furnace hearth.

Preheating methods vary from heating on the charging doorsill to using dedicated preheating ovens.

Preheating standards from times of exposure and temperature also vary. In either case, energy is

consumed for the promotion of operational safety.

Secondary Aluminum (Recycling)

Scrap for recycling is available in many forms. Light scrap is typically baled or briquetted to reduce

transportation costs. Bales and briquettes are typically split for inspection of integrity and

contamination and/or are crushed, shredded, or shripped (sheared and ripped) to controlled flowable

particle sizes. Conveyor systems segregate particle fines for separate processing, provide magnetic

separation, and allow for visual or automated inspection.

Large volumes of aluminum scrap contain paint, enamel, lacquer, or porcelain coatings. These

coatings, which contain oxidizing compounds, would significantly reduce metal recovery if not

removed before melting. Conveyor furnaces or rotary kilns operating at temperatures near the melting

point are required for their removal. Turnings, borings, and other fine scrap may contain oil from

cutting fluids which must be removed for satisfactory metal recovery by swarf (metal fines and

chips) drying in which the oil contributes to the energy required, or by elevated temperature treatment

in rotating kilns. Fine scrap is conveyed to external charging wells for submergence by mechanical

methods including the use of recirculating molten metal pumps. Light scrap is charged directly to the

furnace hearth and is covered by additional heavier charge components.

Melting is typically accomplished in gas-fired furnaces ranging in size from 75,000 to more than

250,000 pounds. Coreless induction furnaces are in common use for rapidly melting fine scrap.

Molten metal from these furnaces is transported to the main furnaces for further processing. After

iterative alloying steps, molten metal is held and processed before casting. Furnaces have thermal

efficiencies ranging from approximately to 20 to 45 percent. Rotary kilns and conveyor furnaces

operate in essentially the same efficiency range. Induction melting is more efficient at approximately

90 percent but furnace capacities are limited and additional steps are required to address oxide

concentrations created by electromagnetic stirring. In-furnace and in-line molten metal treatments are

employed to remove dissolved hydrogen and entrained oxides and other nonmetallics before casting.

In-line systems are usually internally heated using gas-air or electric resistance immersion elements.

Troughing and filter basins are preheated using air-gas torches or resistance elements. Open molds for

casting remelt ingot and sow are routinely preheated before use.

The aluminum, recoverable from skim and dross obtained from foundries, and primary and secondary

molten metal processing units, is typically extraced in secondary producers. While other dross

treatment processes have been developed, rotary furnaces in which dross and salts are mixed and

heated remain the most commonly used. The products of dross recovery treatment by this process are

aluminum and black dross, a mixture of unrecovered aluminum, metallic oxides, and salt. Tertiary

processes have been developed for separating the components of black dross into saleable salt fluxes,

metal, and value-added nonmetallic derivatives such as calcium aluminate and additives for low-

density cement. The economics of tertiary processing are strongly and adversely influenced by the

current low-cost alternative of landfill disposal.

In summary, the energy intense operations common in recycling are:

• swarf drying for removal of combustible contaminants,

Secondary Aluminum (Recycling)

• kiln or conveyor delacquering of coatings,

• preheating charge components,

• melting,

• holding and melt processing, and

• rotary furnace operation

8.3.1 Theoretical Energy for Secondary Aluminum

The theoretical minimum energy required to produce secondary aluminum at 960°C is 0.39

kWh/kg of aluminum. On a theoretical and a practical basis, the energy required to produce

secondary aluminum is less than 6.5 percent of the energy required to produce primary metal.

If the system boundaries are drawn around a secondary aluminum facility, the material entering and

leaving is aluminum metal. Since no chemical change has occurred, the theoretical minimum energy

requirement for scrap conversion is only the energy required to melt and raise the metal temperature

to that required for casting. Molten metal furnace temperatures vary depending on the furnace, alloys,

and other proprietary factors. A molten metal temperature of 960°C, the same production temperature

as for primary aluminum, is used in this report for secondary aluminum to compare aluminum process

technologies easily. In reality, normal pouring temperatures are much lower, in the range of 650°C to

750°C.

The theoretical minimum energy requirement to bring room-temperature (25ºC) aluminum to its

molten form at 960°C is 0.39 kWh/kg, which is less than 6.5 percent of the theoretical energy

requirement for the primary production of aluminum. The theoretical energy required to heat

aluminum from room temperature to its melting point, melt it, and raise the molten aluminum to a

higher temperature is calculated and explained in Appendix K. Pure aluminum melts at 660°C and

requires 0.30 kWh/kg to melt, 23 percent lower than the value used in this report.

8.3.2 Technological Change in the Next Decade

The energy efficiency of the entire aluminum industry can be further increased by capturing a greater

percentage of material for recycling and by improving technology for scrap handling and melting.

Non-technological and nonmarket factors are also important for the continued growth of recycling.

Two of these factors, consumer awareness and incentives, can contribute significantly to the recycling

volume. Consumer awareness requires continuing educating the public about the energy and

environmental benefits of recycling. Offering incentives will aid in the return of aluminum scrap to

the manufacturing base.

Recycling energy efficiency will be enhanced by developing technologies that minimize oxidation

and improve thermal inefficiencies in scrap processing and melting. Improved collection systems and

separation devices (e.g., eddy current, color sorting, laser sorting) can increase aluminum scrap

Secondary Aluminum (Recycling)

recovery by 20 to 30 percent. Improved technology can be applied to increase scrap recovery rates,

especially with regard to aluminum in municipal solid waste. Incremental improvements in existing

furnaces can further reduce the recycling energy requirements. They can be achieved by recuperating

stock gas energy for preheating combustion air and metal feedstock, by modifying burner and furnace

designs, and controlling furnace practice and operating conditions.

53, 54

New technologies must also

be developed to ensure more significant progress.

9. Aluminum Processing

The aluminum industry can be divided into metal- and product-producing sectors. In 2003, the metal-

producing sector manufactured approximately 2,704,000 metric tons of primary metal and 2,820,000

metric tons of secondary metal. The product-producing sector processes these domestically produced

and imported metals into approximately 5,497,000 metric tons of rolled products, 1,719,000 metric

tons of extrusions, and 2,513,000 metric tons of shape castings.

Wire rod and bar, forgings, impacts,

and powder products total more than 700,000 metric tons, comprising about 7 percent of the

aluminum product market; these have not been studied as a part of this report.

Recoveries (yields) in each product processing sector are less than 100 percent. Handled, melted, and

worked tonnage exceeds that implied by the product statistics. The manufacturing of circles and

blanks from aluminum sheet, for example, may recover as little as 35 percent of the original ingot

weight as shipped product. The gross to net weight ratio in gravity castings is often 2:1, in other

words, fifty percent of the original cast weight is automatically designated for remelting.

Furthermore, many products have a genealogy of repeated melting and casting steps. Ingot cast at a

primary smelter is remelted at a secondary producer for casting remelt ingot, which is again remelted

at an aluminum foundry for casting production. On a yield basis, it may be conservatively estimated

that more than 18,000,000 metric tons of aluminum alloys are melted each year to support aluminum

industry shipments. Some casting operations are located sufficiently close to primary smelters and

secondary aluminum producers to allow the molten metal to be shipped in insulated and protected

crucibles, saving the energy of remelting.

Minimizing planned and unplanned scrap represents a large opportunity for energy and cost savings.

Each kilogram of metal that does not go into a final product must be remelted, recast, and reworked.

Remelting also leads to the loss of a percentage of metal to oxidation, which must be replaced with

energy-intensive primary metal. Melting and melt processing operations are the most energy intense

of all post-smelting processes.

9.1 Melting, Alloying, and Melt Treatment

In an aluminum primary smelting facility, molten metal is transferred from the smelting cells to

furnaces for alloying and melt treatment prior to casting. In secondary and other casting plants, ingot,

metallurgical metals, and master alloys must be melted and alloyed. The melting arrangement in most

larger plants provides high heat-input, high melt-rate furnaces for melting, and separate holding

furnaces to which molten metal is transferred for final alloying and preparation for casting. Some

operations have combination melting/holding furnaces.

Aluminum Processing

There are a wide variety of furnace types and designs for melting aluminum. Furnace choice depends

on the required melt volume, melt rate, availability, cost of fuel and electrical energy, and emission

standards. The most common in primary and secondary operations are natural gas-fired reverberatory

furnaces reaching capacities of more than 120,000 kg. Crucible furnaces with capacities ranging from

160 to 4,500 kg are more common in small- and medium-sized foundries. Other furnace types include

coreless and channel induction, electrical resistance and radiant tube furnaces.

Reverberatory furnaces are box-shaped and consist of an insulated steel shell with a refractory lining.

Fuel-fired reverberatory furnaces are used when the melt rate and/or capacity are large. The fuel-fired

reverberatory furnace fires natural gas, propane, or oil directly into the furnace from either the roof or

more typically, the sidewall. The heat is transferred to the surface of the molten aluminum

predominantly by refractory radiation and some convection. There are a large number of

reverberatory furnace design variations: charging and access doors, refractory specifications, side-

wells for charging and/or recirculation, hearth or sidewall induction stirring, split hearths, dry

hearths, divided zones for melting and holding, and various burner capacities and types. Recuperation

concepts include charge preheating, preheating combustion air, and cogeneration.

The growth in recycling has resulted in a number of specialized processes, furnaces, and systems to

improve metal recovery from scrap. Molten metal pumps have been incorporated into these designs to

provide rapid ingestion of fine scrap and more rapid melting of larger scrap forms. Salt flux additions

maintain system cleanliness and aid in the separation of oxides. Pump-induced flow external to the

furnace may include provisions for melt treatment and the separation of oxides as well as for melting.

Natural gas or oil-fired reverberatory furnaces use about 0.87 to 1.96 kWh/kg of aluminum.

addition, a gas furnace increases metal losses due to oxidation. Gas furnaces have 5 to 8 percent metal

loss compared to 0.5 to 3 percent loss in electric furnaces. Recent design innovations in fossil-fuel

reverberatory furnaces help capture the waste heat in the stack gas to preheat incoming materials.

This increases energy efficiency and reduces the time required to melt the metal. Recuperated waste

heat can also be used to preheat combustion air. These technologies can reduce fuel usage to less than

0.57 kWh/kg of aluminum.

Crucible furnaces are more versatile with regard to alloy changes and melt quantities. Combustion

occurs between an insulated steel shell and a crucible of silicon carbide, graphite, clay graphite or

other refractory material resistant to molten aluminum attack. Electrical resistance elements can be

substituted for gas burners in crucible furnace designs.

In recent years, much research has been done on using immersion heaters as a way to remelt

aluminum. Immersion heaters have very low rates of heat loss (~97 percent thermal efficiency), and

have energy usage levels of less than 0.50 kWh/kg of aluminum.

Aluminum Processing

Skimming

Oxide naturally forms on the surface of molten aluminum, resulting initially in a thin protective film.

With increase in time and temperature, the thickness of the oxide layer increases. Turbulence and

agitation accelerate oxide formation and result in the intermixing of metal and oxides. Oxidation

rates are influenced by alloy content and increase with temperature, especially when magnesium is

present in the alloy. The oxide layer also effectively insulates the bath from radiation heat transfer and

must be periodically removed to maintain thermal-efficiency in reverberatory furnaces.

If fluxes are employed to treat the skim before it is removed from the furnace, the oxides are typically

dewet and a large portion of the molten aluminum entrained in the skim layer separates to the melt. In

either case, untreated (skim) or flux-treated (dross) contains entrained free metal as a result of the

skimming action. Efforts are usually made to recover entrained free aluminum after skimming. While

still hot, metal can be drained from the skim gravimetrically and with vibration. Alternatively, skim

may be rapidly cooled by inert gas quenching or in rotating water-cooled steel drums after which free

aluminum may be physically separated. The residue comprising unrecovered aluminum and oxides is

normally further processed for its metal content.

Gross melt losses typically range from 1 to 8 percent. The magnitude of the loss is dependant on the

type of furnace and burners, the surface-to-volume ratio, the practices that are used, and the material

being melted.

Melt loss has significant economic impact since oxidized metal must be replaced in

the supply chain with new primary aluminum metal.

Alloying

Specific elements or combinations of elements are added to molten aluminum to produce aluminum

alloys. Alloying provides the basis for a remarkable range of physical and mechanical property

capabilities not displayed by unalloyed aluminum or by any other metal system. Among the elements

and combinations of elements added to molten aluminum are modifiers and refiners, which provide

finer grain structures, and controlled microstructural features, including metallurgical phases that

influence recrystallization behavior.

Molten Metal Treatment

Molten aluminum contains dissolved hydrogen, entrained oxides, and other nonmetallic inclusions

which, if not removed, would adversely affect metal acceptability and performance. Treatment with

salt fluxes or active fluxing gases changes the interfacial relationship of included particles with the

melt so that gravitational separation is facilitated. Fluxing with argon, nitrogen, and/or other gases

results in flotation of entrained matter while dissolved hydrogen is reduced by partial pressure

Aluminum Processing

diffusion. Metal treatment takes place in the melting furnace, holding furnace, or in-line between the

furnace and the casting unit. Rotary degassers have been developed to provide the finest dispersion

and intermixing of metal and fluxes for these purposes.

Molten metal filtration for the removal of particulate contamination was introduced in the 1950s and

has grown in importance and application since that time. The first and still among the most effective

filtration processes are deep bed filters using tabular alumina as the filtration medium. Crushed

carbon beds are capable of fine inclusion removal and a reduction in sodium content that is important

for many products. Porous foamed ceramics are widely used for commercial grade and higher quality

requirements. Fused ceramic and refractory filtration elements are also available.

9.1.1 Energy Requirements for Melting Aluminum

Melting is an energy-intensive process; it requires nearly the same amount of energy to raise one

kilogram of aluminum to a molten 700°C state as it does to raise one kilogram of iron to 1,500°C.

However, nearly three times the volume of aluminum is produced compared to iron because of

density differences.

The energy requirements for melting aluminum are presented in Section 8.3.1, Theoretical Energy for

Secondary Aluminum. When the system boundaries are drawn around an aluminum melting facility,

the material entering and leaving is aluminum metal. Since no chemical change has occurred, the

theoretical minimum energy requirement is only the energy required to melt the metal. Molten metal

furnace temperatures vary depending on the furnace, alloys, and other proprietary factors. The

theoretical minimum energy requirement to bring room-temperature (25°C) aluminum to a molten

960°C metal is 0.39 kWh/kg. The theoretical energy requirement for melting pure aluminum to

molten metal at various temperatures is presented in Appendix K.

Basic natural gas or oil-fired reverberatory furnaces range in efficiencies from approximately 20 to 45

percent. The more efficient furnaces employ recuperation of stack gas heat for reduced melting

energy requirements through charge preheating or for more efficient burner operation through

preheating combustion air. Furnace condition and operating practices have large effects on energy

performance. Because heat transfer in reverberatory furnaces takes place principally through

radiation, melt surface temperatures are considerably hotter, leading to more rapid oxidation and

higher melt losses.

Electric furnaces, typically used in small processing operations, do not require a flue and their heating

chambers can be made nearly airtight. A side-well is provided for charging metal and alloying

materials. The side-well removes the need to open the furnace door and prevents a major convective

heat loss. Energy losses (excluding electrical generation and transmission) in electrical furnace

heating are principally due to conduction and radiation losses from the exposed furnace shell. Losses

Aluminum Processing

are typically 0.49 to 0.81 kWh/kg of aluminum. Induction furnaces are typically more than 90

percent energy efficient, while gas-fired crucibles are 15 to 28 percent, and electrically heated

crucibles 83 percent energy efficient.

9.1.2 Technological Change in the Next Decade

The industry is constantly evaluating, adopting, and improving furnace technologies and practices.

This provides not only energy and environmental benefits, but also cost savings. New burner

technologies and oxygen-enhanced combustion systems are being developed and evaluated to further

improve efficiency and reduce the costs of melting without increasing emissions. All industry

segments have transitioned to greater reliance on scrap and recycling for their metal needs.

Technologies for sorting, handling, and remelting scrap in all forms with optimum metal recoveries

and lowest costs are continuously being developed and refined. New technologies for melting thin

gauge material to minimize oxidation losses are being developed and implemented. Also, industry is

continually seeking better methods to recover the aluminum that is trapped in dross. Additional

efforts are directed at the closed-loop recycling of dross-related wastes including saltcake.

New laser technologies will speed the in-situ chemical analysis of molten metal and minimize the

processing time required for alloy compliance. The development of better, longer-lasting, ceramic

materials for furnace linings is ongoing and will reduce the time required for furnace maintenance.

The aluminum industry has recently published a technology roadmap in conjunction with the

advanced ceramics industry, to encourage the development of superior furnace construction

materials.

Finally, new melting technologies now under development offer the prospects for revolutionary

improvements in melting efficiencies that may be applicable to the scale and operational demands of

much of the industry. One exciting development is that of immersion heating with high watt-density

elements for melting as well as temperature maintenance. The development of commercial immersion

heaters for aluminum remelting is very likely to occur within the next few years.

9.2 Ingot Casting

Ingot casting is the solidification of molten alloys into shapes that are suitable for subsequent

thermomechanical processing or for remelting. Ingots are made by controlled solidification in molds

designed to produce the desired geometrical configuration and metallurgical characteristics.

Ingot casting is by itself not energy intensive; however, casting is typically a batch process, and large

quantities of molten metal are held in furnaces in which alloying, fluxing, and degassing are

performed. Accordingly, conductive and radiant heat losses occur from these furnaces operations.

Aluminum Processing

Ingot for wrought product applications is almost universally cast by the semi-continuous direct-chill

(DC) casting process. The process includes different means of introducing and controlling the flow of

molten metal into the mold, lubrication methods, the use of insulation in mold construction, and the

injection of air or imposition of an electromagnetic field for reducing or eliminating contact between

molten metal and the mold. The process produces rectangular cross-section ingots for rolling, round

log-like billets for extrusion, squares for wire, rod, and bar products, and various shapes as fabricating

ingot in forging.

The DC casting process begins when aluminum flows from the furnace through troughs to the casting

station. At the casting station, the aluminum flows into one or multiple water-cooled stationary molds

that rest on the casting station table. The DC ingot molds are only a few inches deep and form the

cross-section of the ingot or the billet. The ingot is initially formed in the water-cooled mold. Once

perimeter solidification has begun, the casting table is gradually lowered into the casting pit, while

additional molten aluminum is supplied to the top of the mold. The water-cooled mold remains at the

top of the pit and continues to shape the casting. Water sprays impinging on the solid shell continue

the solidification process of the molten ingot core. The casting table is lowered into a casting pit until

the desired length is achieved. After casting, ingot intended for wrought fabrication may be stress-

relieved, scalped, cut to length, and homogenized. Cutting, shearing, forming, and other mechanical

operations, as well as melting, heating, casting, heat treating and other thermal operations are utilized

by the product-producing sector.

Casting operations attempt to control the crystal/grain structure and composition gradient of cast

products. Grain size and boundaries are important factors affecting the material’s physical and

mechanical properties in cast and wrought form. However, because there is a significant temperature

profile across the ingot cross-section during solidification, grain structure and composition can vary

from surface to center. For subsequent fabricating operations, it is usually necessary to remove the

skin layer by scalping so that the final product has consistent physical properties. The amount of

surface to be removed is dependent on shape, surface quality, and the depth of undesirable grain

structure and segregation. Scalpings are remelted and reprocessed, which results in additional energy

usage and metal losses due to oxidation.

A percentage of DC ingots are rejected for quality reasons. Cracking may occur during or after

solidification. Surface defects may form, which affect the acceptability of the ingot for wrought

processing. Other specialized standards concern grain structure, segregation, and microstructure. At

times, ingots are found to exceed alloy specification limits. The processing energy used to produce

the ingot is then lost and additional energy is required for remelting and reprocessing.

Aluminum Processing

9.2.1 Energy Requirements

More than 2,704,000 metric tons of primary aluminum were cast into ingots in 2003 (Appendix G).

The tacit energy consumed in aluminum processing operations can be divided into three categories:

1. fossil-thermal energy use, which includes furnaces, heating and heat treatment operations;

2. electrical energy required for heating, sawing and scalping, and for motor, pump and compressor

operation; and

3. other fuel-consuming operations, including transportation.

Primary ingot casting has typical metal yields from 88 to 98 percent and requires about 1.01(1.46

)

kWh/kg of cast ingot product (Appendix F, Table F.2).

Table 9.1: Primary Ingot Casting Distribution of Energy Consumption (Appendix F, Table F.10)

Energy Category Fossil-Thermal Electrical Other Fuels

Percent

79 (63

) % 21 (36

) % 0 (0

) %

The theoretical minimum energy requirements for primary and secondary castings are the same at

0.33 kWh/kg of aluminum (Appendix F, Table F.3). The difference in their actual energy usage results

from their respective initial materials. In primary casting, the initial material is molten aluminum in a

holding furnace, while the initial material in secondary casting is metal scrap. The scrap must first be

melted before it enters a holding furnace, giving secondary casting a higher actual energy use than

primary casting. Secondary aluminum was cast into 2,820,000 metric tons of ingots in 2003.

Secondary casting has typical yields of 96 percent and requires about 2.50 (2.81

) kWh/kg of product.

Table 9.2: Secondary Ingot Casting Distribution of Energy Consumption (Appendix F, Table F.10)

Energy Category Fossil-Thermal Electrical Other Fuels

Percent

77 (73

) % 5 (9

) % 18 (17

) %

9.2.2 Technological Change in the Next Decade

Ingot casters have strived to improve process yields and to refine practices to provide more consistent

internal and surface quality in wrought ingot manufacture. These efforts have resulted in significant

progress in surface, sub-surface, and metallurgical improvements. Grain refining by heterogeneous

nucleation agents has benefited from decades of constant research and development by primary

producers in cooperation with master alloy suppliers.

Molten metal fluxing and filtration processes continue to undergo changes leading to greater

efficiencies, higher product quality, reduced environmental impact, and reduced costs.

Numerous research programs are directed at the modeling and prediction of the solidification process

for reduced cracking incidence and improved structural uniformity. The evolution of mold designs

capable of improved surfaces, reduced scalping, and higher production rates continues in all wrought

ingot production.

Aluminum Processing

All of these developments reflect advances in sensors and instrumentation that permit more accurate

monitoring and comprehensive control of the casting process.

9.3 Rolling

Rolling is the process of reducing ingot thickness by passing it between counter-rotating steel rolls.

Aluminum-rolled products include plate (typically > 0.6 cm thick), sheet (typically 0.02 cm to 0.6 cm

thick), and foil (typically < 0.02 cm thick). Figure 9.1 shows the unit operations of a typical rolling

mill.

Figure 9.1: Typical Rolling Mill Processing Operations

Hot and cold rolling operations are used industrially to shape material into the broad category of flat

rolled products. Hot rolling is generally performed at various temperatures exceeding the

recrystallization temperature for differing rolling alloys. Cold rolling takes place at room temperature,

but the heat generated can result in metal temperatures as high as 150°C.

Large high strength alloy ingot may require thermal stress relief after casting. Most ingots are

homogenized to reduce intergranualar segregation and to modify intermetallic particle form.

Homogenization may be integral to preheating for hot rolling, or the preheating may take place

separately. Although some ingot may be rolled with the cast surface intact, rolling surface faces are

normally removed by scalping, and the head and butt of the ingot are “cropped” to assure finished

product uniformity.

Hot rolling begins with repetitive reversing mill reductions on stock preheated to temperatures in the

range of 400°C to 500°C. The slab is passed repeatedly back and forth through the rolls until the

desired reduction in thickness is achieved. While heat is generated during deformation, the number of

passes required at the reversing mills may – because of reducing section thicknesses, time, and

coolant application – cause losses in temperature that require reheating before continuing the hot

Aluminum Processing

rolling process. Slabs are lifted from the hot line and charged in reheating furnaces until working

temperatures are restored. End crops are taken during breakdown rolling as required to maintain

squareness.

Edge trimming knives remove stock from the edges of the sheet before coiling. The amount of edge

trim required is determined by the depth of edge cracks and the ragged conditions associated with the

unrestrained deformation that occurs during hot rolling. The amount of edge trim required with

cropping losses represents a significant percentage of planned scrap that must be returned to the cast

shop for remelting. Edge trim losses can be minimized by improved ingot quality and by edge rolling.

The hot rolling process completely changes the microstructure formed during casting and elongates

the grain structure in the direction of rolling. Even though deformation temperatures are typically

greater than the recrystallization temperature, there is inevitably some degree of equivalent cold work

so that annealing at reroll gauge will result in recovery or recrystallization before cold rolling.

9.3.1 Energy Requirements for Rolling Aluminum

About half of U.S. rolled aluminum products are cold rolled. 2,421,300 metric tons of cold rolled

products were produced in 2003 (Appendix F, Table F.4). Cold rolling has typical yields of about 84

percent and requires about 0.64(1.35

) kWh/kg of rolled product.

Table 9.3: Cold Rolling Distribution of Energy Consumption (Appendix F, Table F.10)

Energy Category Fossil-Thermal Electrical Other Fuels

Percent

42 (25

) % 55 (72

) % 3 (3

) %

In 2003, nearly 2,421,300 metric tons of hot-rolled products were produced (Appendix F, Table F.4).

Hot rolling has typical yields of about 82 percent and requires about 0.62(1.16

) kWh/kg of rolled

product.

Table 9.4: Hot Rolling Distribution of Energy Consumption (Appendix F, Table F.10)

Energy Category Fossil-Thermal Electrical Other Fuels

Percent

57 (38

) % 43 (62

) % 0 (0

) %

Theoretically, it is possible to roll products without the need for heat treatments and with no loss of

material due to trimming or slitting. In this case, the minimum theoretical energy to roll a product is

composed of only two components:

• the energy required to heat starting stock to the rolling temperature, and

• the energy required to deform the shape.

Aluminum Processing

The hotter the material, the lower the deformation energy required. The heat capacity equations

required to calculate the energy requirement for heating pure aluminum are listed in Appendix I. The

energy required for deforming is given by the equation E = εσ

c, where ε is the strain or deformation

defined as ε = ln (t

), where t

represents the initial and t

the final dimension, σ denotes the yield

stress, and c denotes a constant describing the shape of the stress strain curve. The yield stress value

for aluminum can vary by as much as a factor of ten over the hundreds of alloys that are used by the

industry.

The very large variations in alloy properties, particularly the shape and the magnitude of the stress

strain curves, make it possible to calculate theoretical minimum energy requirements for rolling

aluminum only for a specific rolling process with a specific alloy. There are a large number of heavy

equipment requirements in rolling mill operations that are not confined to rolling. Roller and tension

levelers, slitters, stretchers, roll formers, and paint or coating lines are examples of energy consuming

operations that rely on pumps, motors, and compressors as well as on mechanical design for

efficiency. A rough approximation of the rolling sector’s minimum energy can be made by assuming

overall process heating efficiencies and electric/hydraulic system efficiencies and by looking at the

entire rolling sector yield and energy consumption. If the overall sector heating efficiency is 50

percent and the electric/hydraulic system efficiency is 75 percent, the estimate of the minimum

energy requirement is 0.31 kWh/kg of product for hot rolling and 0.33 kWh/kg of product for cold

rolling. These assumptions imply that cold rolling is operating at about 52 percent overall energy

efficiency and hot rolling at about 50 percent overall efficiency (Appendix F, Table F.3).

9.3.2 Advanced Rolling Technology

Hot rolling typically requires numerous passes through the rolling mills and is energy intensive. One

approach to improving productivity and reducing heating energy is to continuously cast molten metal

into slab or strip.

Going directly to thin strip, continuous strip and slabcasting saves the energy

required for homogenization, scalping, preheating, end and side trim, and multiple passes through

rolling mills.

Continuous strip casting is in wide current use for some sheet and foil specifications and has

demonstrated energy savings of more than 25 percent relative to conventional ingot rolling. These

casters convert molten metal directly into reroll gauge sheet at 1 mm to 12 mm thickness. Continuous

strip casters employ twin counter-rotating water-cooled rolls or belts to accomplish solidification.

Strip casting is now restricted to certain low alloy content compositions. Technology is being

developed for more complex or more highly alloyed compositions. Very high solidification rates and

the extrusion component of casting between cylindrical rolls result in high degrees of segregation of

solute to the centerline and in cracking tendencies.

Aluminum Processing

While continuous strip casting is recognized as alloy-constrained, an alternative continuous process

for coiled reroll, slab casting is highly alloy tolerant and results in metallurgical structures that closely

correspond to those of the ingot and wrought products produced by the Direct Chill process. In slab

casting, solidification takes place between water-cooled belts or blocks. Slab thickness varies but

typically corresponds to continuous hot mill entry gauges of 75 mm to 150 mm. An individual slab

casting line has a production capacity ten times that of the largest strip caster. The cast slab is directly

fed into one or more in-line, low-speed, high-torque, hot reduction mills that reduces thickness to

coilable gauge. Slab casting processes have successfully produced high strength aluminum alloy

sheet as well as challenging products such as beverage can sheet, but subtle differences in product

performance—especially in formability and anisotropy—continue to favor the use of conventionally

hot-rolled sheet in these applications.

A further advancement in rolling technology is spray rolling. In this process, molten metal droplets

are sprayed directly into the nip of twin rolls and the material is solidified and consolidated directly

into sheet material in one step. In concept, this offers the most energy-efficient process, and recent

projects have sought to advance this technology.

9.4 Extrusion

Extrusion is the process of forcing an aluminum ingot or billet through a steel die to form an

elongated shape of consistent cross-section. Extruded products include rods, bars, tubes, and

specialized products interchangeably called shapes, sections, or profiles. Figure 9.2 shows the unit

operations of a typical extrusion plant. After rolling, extrusion is the second most common processing

technique for aluminum. Aluminum extrusion is remarkable because the process combines high

productivity with an essentially infinite variety of extremely complex shapes, cross-sections, or

profiles that cannot be economically duplicated in any other process. Furthermore, aluminum can be

readily extruded; this process is either extremely difficult or impractical for many other metals.

Figure 9.2: Typical Aluminum Extrusion Processing Operations

It is possible to produce almost any cross-sectional shape, within wide limits. Through the use of

hollow stock and floating or fixed mandrels, hollow shapes or cross-sections with complex enclosed

configurations can be produced. The extrusion process is capable of producing a cross-section with a

weight of a few grams to more than 300 kg/m, a thickness of less than 1 mm to over 250 mm,

circumscribing diameters of 5 mm to 1,000 mm and lengths in excess of 30 m. The appropriate choice

Aluminum Processing

of alloy and extrusion conditions can result in an optimum combination of properties for a particular

application. Such properties may include tensile strength, toughness, formability, corrosion

resistance, and machinability. Countless extrusion products are made in the United States. These

include frame and supporting shapes for windows and doors, carpet strips, household bath enclosures,

screens, bridge structures, automotive parts, aerospace components, and many other consumer

products.

Operations at individual plants vary widely depending on the cross sections and alloys produced.

Extrusion presses range in capacity up to 15,000 tons but the most common are perhaps 2,500 tons.

Presses may extrude vertically or horizontally but virtually all modern extrusion presses are

horizontal. Typically, billets in diameters up to 275 mm are preheated to temperatures ranging from

450°C to 550°C depending on the alloy, product design, and the desired mechanical characteristics.

The preheated billet is charged to the extrusion press container and forced by hydraulic pressure

through the extrusion die.

There are essentially two processes for extrusion production. In the direct extrusion process, the billet

is hydraulically pressed through the die, while in the indirect extrusion process, the die is forced over

the billet. In direct extrusion, the billet surface is retained in the extrusion container and contributes to

butt loss, which may total 8 percent of the starting billet weight. Because the billet surface is not

extruded to become part of the product, scalping is not required. For indirect extrusion, the billet

surface becomes an integral part of the product and so, scalping before extrusion is essential.

The single largest area for energy improvement in extrusion technology is the reduction of process

scrap, including butt losses. Extrusion product specifications include significant surface-quality and

chemical finishing criteria. Defects include torn surface, die pick-up which is often related to the

billet’s microstructure, and non-fill. Variable response to chemical finishing results in appearance and

color mismatches which affects product acceptability.

9.4.1 Energy Requirements for Extruding Aluminum

The United States produced over 1,826,000 metric tons of extruded aluminum products in 2003

(Appendix F, Table F.4). Extrusion processes have typical yields of 69 percent and require about

1.30(1.52

) kWh/kg of extruded product.

Table 9.5: Extrusion Distribution of Energy Consumption (Appendix F, Table F.10)

Energy Category Fossil-Thermal Electrical Other Fuels

Percent

87 (25

) % 7 (72

) % 6 (3

) %

Aluminum Processing

Theoretically, it is possible to extrude products without the need of additional heat treatments and

without loss of material. The minimum theoretical energy to extrude a product, in such a case, is

composed of only two components:

• the energy required to preheat the billet to extrusion temperature, and

• the energy required to deform the material through a die.

The hotter the material is, the lower the deformation energy required. The simpler the die, the lower is

the extrusion-energy requirement. The heat capacity equations needed to calculate the energy

requirements for heating pure aluminum are listed in Appendix K. The energy required to deform the

material through the die is highly dependent on the size and shape of the product and the die design.

Calculation of the minimum extrusion force is very complex and can only be estimated with

theoretical and empirical models. Typical formulae have the following simplified form:

F = A

(σ

/η) ε, where

ε, the strain, corresponds to the reduction area, ε = ln(A

is the mean stress for the strain,

is the original cross-sectional area, and

η is an efficiency factor.

The very large variations in alloy properties, particularly the infinite numbers of possible shapes,

make it impossible to calculate a theoretical minimum energy requirement. This value can only be

determined by analyzing a specific process, a specific extruded profile, and a specific alloy. The

perimeter of the profile and the radius of intersecting edges have a large influence on the force

required for extrusion. A rough approximation of a minimum energy value can be made by examining

the entire aluminum extrusion industry yield and energy values, and assuming an overall process-

heating efficiency and electric/hydraulic-system efficiency. This approach provides an estimate of the

minimum energy, 0.44 kWh/kg of aluminum, when efficiencies are assumed to be 50 percent for

heating and 75 percent for the electric/hydraulic system. These assumptions imply that overall

extrusion facilities operate at about 34 percent energy efficiency.

9.5 Shape Casting

Shape casting or the casting of engineered designs enables the production of simple and complex

parts that meet a wide variety of needs. The process produces parts weighing ounces to parts

weighing several tons. Figure 9.3 shows the unit operations of a typical aluminum shape casting

foundry. These operations vary significantly depending on the size of the operation, the processes

employed, the complexity of the parts, alloy compositions, and the type of castings being produced.

Aluminum Processing

Figure 9.3: Typical Aluminum Product Shape Casting Operations

The basic casting process consists of melting and alloying aluminum and pouring or injecting molten

metal into molds containing single or multiple cavities of the desired shape. The important casting

processes for engineered aluminum castings are pressure die, permanent mold, green and dry sand,

plaster, and investment casting.

In pressure die or die casting, metal is injected at pressures up to 10,000 psi into water-cooled steel

dies. Productivity rates are high and the process can be highly automated. While surface quality and

dimensional accuracy are excellent, the typical die casting contains a degree of internal unsoundness

associated with non-directional solidification, entrapped gasses and inclusions resulting from

turbulent metal flow, and the presence of air and lubricants in the die cavity.

Permanent mold or gravity die-castings are produced by introducing molten metal by gravity or

counter-gravity means into iron or steel molds. Productivity, surface quality and dimensional

accuracy is lower than in pressure die-casting. Internal soundness, depending on the extent to which

sound molten metal treatment for hydrogen elimination and oxide removal are practiced and the

principles of directional solidification are employed, can meet the most challenging quality standards.

Low-pressure casting is usually considered a variation of the permanent mold process even though

dry sand and plaster cast parts have been produced. In this process, molten metal is forced by the

application of pressure to rise through a tube into a mold mounted over the furnace. It has the

advantages of a significantly reduced gross to net weight ratio and correspondingly lower trimming

costs.

Green and dry sand casting can also yield high integrity parts. In green sand casting, sand, binders

such as clays, and moisture are blended to provide the molding medium. Patterns may include loose

pieces, wood models, cast match-plates, and molded styrofoam. For dry sand molding, air or thermal

setting chemicals coat sand particles so that the finished mold after curing offers superior surfaces,

dimensional accuracy, and shelf life.

Investment molds are produced by repetitive immersion of plastic, wax, or other low temperature

melting and volatile pattern material into ceramic slurries. After drying, the hardened casing

containing the pattern is heated to a temperature at which the pattern material is eliminated. Typically,

the mold is preheated before pouring and may be filled under vacuum. Investment castings offer

extremely fine finishes, thin walls, and excellent dimensional accuracy. Plaster molding offers the

same advantages and may be used for the production of thicker sections, large parts for which

investment is less suited.

Aluminum Processing

Molten aluminum is required as the basis for all foundry production. High volume casting operations

may acquire part or all of their metal requirements as molten metal delivered over-the-road in

insulated crucibles. Other foundries melt and process prealloyed ingot, RSI, and internal scrap

including gates and risers. All melting and melt processing technologies and considerations described

in Section 9.1 are applicable to foundry operations.

9.5.1 Energy Requirements for Shape Casting Aluminum

Nearly 2,413,000 metric tons of shape-cast products were produced in 2003. Shape casting has

typical yields of only 45 percent and requires about 2.56(2.64

) kWh/kg of cast product. The energy

consumed is almost exclusively related to furnace and heating operations (Appendix F, Table F.4).

Table 9.6: Shape Casting Distribution of Energy Consumption (Appendix F, Table F.10)

Energy Category Fossil-Thermal Electrical Other Fuels

Percent

100 (80

) % 0 (14

) % 0 (6

) %

The theoretical minimum energy requirement for shape casting can be calculated from the energy

required to go from room temperature to liquid metal plus some superheat value (Section 8.2). Pure

aluminum melts at 660°C. The minimum energy required to produce liquid aluminum at 660°C is

approximately 0.3 kWh/kg.

Alloy composition, superheat requirements, mold sprue, gates, runners, and riser systems, and post-

casting heat treatments vary by mold design and casting practices. A rough approximation of the

aluminum shape cast sector’s minimum energy requirement can be made by looking at the entire

sector yield and energy values (Appendix F, Table F.2), and assuming an overall process heating

efficiency and electric/hydraulic-system efficiency. The estimate of the minimum energy requirement

using this approach is 0.60 kWh/kg, when a 50 percent overall heating efficiency and a 75 percent

electric/hydraulic system efficiency are assumed. These assumptions imply that shape-casting

facilities operate at about 23 percent overall energy efficiency.

9.5.2 Technological Change in the Next Decade

Castings are among the most cost-effective and versatile solutions to part design and performance

challenges. The range of available alloys and properties provides combinations of manufacturability

and product characteristics for one-of-a-kind, prototype, limited, or high volume applications.

Castings are near-net-shape with the potential for precise integral internal passages and complex

shapes. Aluminum is cast in more alloys with a wider range of physical and mechanical properties by

more processes than any competing metal system.

Aluminum Processing

The most important trend affecting aluminum casting production is its continous growth in

automotive applications. The advantages of aluminum for many powertrain components including

transmission cases, oil pans, pistons, intake manifolds, cylinder heads, and engine blocks have been

reflected in its wide adoption.

Squeeze casting and semisolid forming have emerged as candidates for a new generation of process

capabilities producing heat-treatable, high integrity pressure die castings that utilize lower impurity

compositions, vacuum, and dry lubrication.

9.6 Thermal Treatments

A significant component of energy use in the aluminum industry is the heat treatment of metal and

products. The physical and mechanical properties of aluminum alloys in any product form can be

controllably altered by thermal treatment. Thermal treatments are used to soften the material and to

recrystallize the grain structure. Other aluminum alloys, principally those containing copper,

magnesium, silicon, and zinc can be thermally treated to significantly improve strength through the

dissolution and reprecipitation of soluble phases. Aluminum alloys are categorized as heat-treatable if

thermal treatments have significant hardening benefits or nonheat-treatable if the alloy is

unresponsive to thermal treatments for hardening purposes. Among the latter are alloys dependent on

work-hardening for strengthening and those whose properties are essentially fixed after solidification.

All product types including sheet, plate, foil, wire, rod, bar, extrusions, forgings, and castings are

produced in heat-treatable alloys. The majority of extrusions, forgings, and a large percentage of plate

and castings are heat-treated. Heat-treatment facilities are integral to larger operations. Commercial

firms also provide contracted heat-treatment services.

Annealing is performed at temperatures from 300°C to 500°C to reduce strength, improve formability

and ductility, lower residual stress levels, and improve dimensional stability in cast and wrought

products. Since electrical conductivity is adversely affected by elements retained in solution,

annealing is also used in electrical and electronic applications. Deep drawn sheet is normally

annealed. Intermediate annealing is usual practice in rolled product manufacture to permit subsequent

cold reductions. The final temper of nonheat-treatable rolled products may include annealing, partial

annealing, or stabilization treatments. Most annealing is a batch operation.

Heat-treatable aluminum alloys contain intermetallic metallurgical phases which can be dissolved at

elevated temperature (up to 550°C) and retained in solid solution by rapid quenching. Solution heat-

treatment is batch or continuous. Sheet and foil can be heat-treated continuously through accumulator

towers and plate, castings, and forgings by conveyer furnaces. Coiled sheet, plate, and extrusions are

more typically batch treated. In either case, the product must be held at solution temperature long

enough for complete solution to occur and for desirable changes in the shape or form of the insoluble

intermetallics that are present in the microstructure. The quench medium is water at room

Aluminum Processing

temperature. The retained metastable solid solution permits precipitation hardening at intermediate

temperatures (170°C to 300°C) for fully hardened, partially hardened, or over-aged conditions. Each

offers combinations of strengths, ductilities, toughness, stability, resistance to stress corrosion and

hardness not achievable through solidification or work hardening. Precipitation hardening furnaces

are also batch or continuous.

The high rate of solidification in many casting processes results in a degree of solution retention in

heat-treatable compositions permitting age or precipitation hardening to be employed without

solution heat-treatment. This method is extensively used in extrusions which can be press-quenched

by forced-air or water-mist to improve solution retention.

Recommended solution heat-treatment practices have been standardized to reflect worst-case

conditions. The cycle defines the minimum time at the required temperature for successfully treating

the part requiring the longest exposure. A safety factor is usually also applied to assure that reheat-

treatment will not be necessary. Practices are typically generic, applying to specific alloys and

tempers without regard for section thickness or degree of metallurgical refinement. Thinner wall

castings, forgings and extrusions generally respond to heat treatment more rapidly. Finer grain and

dendrite cell sizes are reflected in smaller, more dispersed, solute particles which can be more rapidly

dissolved. Solution heat treatment time can therefore be patterned to specific products and

manufacturing processes and the combination of finer metallurgical structures. Decreased variability

can result in substantially reduced cycle times and energy costs.

New heating technologies are being studied to reduce energy requirements through more rapid

heating to treatment temperature. Fluidized beds and infra-red heating can be used to shorten heat-up

times but do little to accelerate either the rate of solution or microstructural change once solution

temperatures are reached.

Another approach being investigated is the use of sensible heat to reduce energy requirements. While

there are metallurgical concerns, extrusions, castings, and forgings can be placed in heat-treatment

furnaces directly from the mold or die, thereby preserving the latent heat of the casting or final

forming operation.

9.6.1 Energy Requirements for Thermal Treatment

The theoretical energy required for all thermal treatments can be calculated from the specific heat or

heat capacity of the various aluminum alloys. For example; it requires 0.06 kWh/kg (95Btu/lb) to heat

A356 to its solution heat treatment temperature, 0.05 kWh/kg (80 Btu/lb) to reach annealing

temperature for 1100 alloy, and 0.02 kWh/kg (32Btu/lb) to reach precipitation hardening temperature

for alloy 2024.

Aluminum Processing

Time at temperature for each procedure varies depending on alloy and product form but can exceed

twelve hours. While most energy is consumed in raising the product and oven to temperature,

additional energy is required to maintain the temperature for the duration of the cycle. Sustaining

energy requirements are exclusively a function of furnace design and condition. Standard quench

temperatures include room temperature: 65°C (150°F), 80°C (180°F), and 100°C (212°F). Large

volumes of water must be heated and maintained at temperature by steam or other means for the latter

practices.

Endnotes

Aluminum Statistical Review for 2003 (Washington, D.C.: The Aluminum Association, Inc.,

2004), 7. The Aluminum Association’s statistical reviews are issued annually and assemble, in

one document, important detailed data from primary production, to markets for finished goods,

to the recovery of scrap. Both U.S. and world information is included.

Aluminum Industry: Vision Sustainable Solutions for a Dynamic World, (Washington, D.C.:

Aluminum Association, Inc., November 2001), 3. The Vision document identifies key needs of

the industry and outlines a comprehensive aluminum R&D agenda.

Production of Aluminium and Alumina, ed. A.R. Burkin (Chichester: John Wiley & Sons, 1987)

(p. 38). This book, published on behalf of the Society of Chemical Industry, is one of the only

texts to provide a comprehensive overview of the state-of-the-art of all science and technologies

associated with primary aluminum production in a single volume. It is a detailed, academic

volume that covers chemistry, thermochemistry, fluid dynamics, process dynamics, etc., from

both a theoretical and practical perspective.

Dietrich G. Altenpohl, Aluminum: Technology, Applications, and Environment; A Profile of a

Modern Metal, Sixth Edition (Washington, D.C.: The Aluminum Association, Inc. and the

Minerals, Metals and Materials Society, 1998). This book is described in its foreword as having

“global recognition as the definitive educational text and reference book for aluminum industry

participants, a broad range of aluminum fabricators and users, students, and the scientific,

engineering, and academic community.”

Aluminum Industry Vision, Sustainable Solutions for a Dynamic World, 22.

Robert D. Naranjo, Ehr-Ping Huang Fu, and Mike Gwyn, “Castings Drive Fuel Efficiency,”

Modern Casting Vol. 94 No. 9 (September 2004), pg. 20.

Life Cycle Inventory Report for the North American Aluminum Industry (Washington, D.C.:

Aluminum Association, Inc., 1998). This report is summarized in Appendix F. This report

provides information on a life cycle inventory study of the North American aluminum industry in

1995. The report is the result of extensive surveying and contains the best and most complete

industry performance information of any recent study. The report was produced in accordance

with International Organization for Standardization (ISO) procedures and was favorably peer-

reviewed by groups outside the industry.

Energy and Environmental Profile of the U.S. Aluminum Industry (Washington, D.C.: U.S.

Department of Energy Office of Industrial Technologies Program, July 1998). This report

reviews the energy and environmental characteristics of the key technologies used in the major

processes of the aluminum industry.

Aluminum Statistical Review for 2003, 49.

“Bauxite and Alumina Statistics and Information,” U.S.Geological Survey, Minerals

Information, Available: [http://minerals.usgs.gov/minerals/pubs/commodity/bauxite/].

Aluminum Statistical Review for 2003, 7.

Aluminum Statistical Review for 2003, 38, 42.

Life Cycle Inventory Report for the North American Aluminum Industry, E-3.

Technology Roadmap for Bauxite Residue Treatment and Utilization (Washington, D.C.: The

Aluminum Association, February 2000). This booklet contains a comprehensive discussion of

bauxite residue.

Production of Aluminium and Alumina, 38.

Energy and Environmental Profile of the U.S. Aluminum Industry, 29.

Life Cycle Inventory Report for the North American Aluminum Industry, E-8.

Alumina Technology Roadmap (Washington, D.C.: the Aluminum Association, Inc., November

2001). This document identifies performance goals and describes 25 research and development

areas that are a priority for the global alumina industry.

Paul J. Ellis, “Tutorial: Petroleum Coke Calcining and Uses of Calcined Petroleum Coke,” 13.

Speech given at AIChE 2000 Spring National Meeting, 3

International Conference on Refining

Processes, March, 2000.

Production of Aluminium and Alumina, 57.

Life Cycle Inventory Report for the North American Aluminum Industry, E-6.

Production of Aluminium and Alumina, 49.

Ellis, 13.

“The Manufacture of Carbon and Graphite,” Chapter in Industrial Graphite Engineering

Handbook, (Parma, OH: UCAR Carbon Company, Inc, 2001) 1-6.

Aluminum Statistical Review for 2003, 10-11.

Warren Haupin, “History of Electrical Consumption by Hall-Héroult Cells,” Chapter in Hall-

Héroult Centennial, ed. W. Peterson and R. Miller (New York: the Metallurgical Society of

AIME, 1986) 106. This article provides a detailed historical review of energy usage in the Hall-

Héroult process from the first cells to 1986.

R.P. Pawlek, “75 Years of Development of Aluminum Electrolysis Cells,”Aluminium, 75.9

(1999): 734-743.

Barry J. Welch, “Aluminum Production Paths in the New Millennium,” Journal on Metals51.5

(February 1999): 24-28.

Report of the American Society of Mechanical Engineers’ : Technical Working Group on Inert

Anode Technologies (Washington, D.C.: U.S. Department of Energy Office of Industrial

Technologies, July 1999) Appendix A-9: Thermodynamics of Electrochemical Reduction of

Aluminum, 3. This report reviews the literature and patents concerning inert anode technologies.

It presents a theoretical discussion of aluminum reduction and has input from over eleven experts

in the field. Carbon anode voltage equation E = 1.898 - 0.0005728 x T, where E is in V(dc) and T

is in degrees Kelvin.

Noel Jarrett, W.B. Frank, and Rudolf Keller, “Advances in Aluminum Smelting,” Metallurgical

Treatises AIME VI.93 (1981): 137.

Production of Aluminium and Alumina, 63.

U.S. Environmental Protection Agency, Office of Air and Radiation, Web site. Available:

[www.epa.gov/globalwarming/].

G. D. Brown, M. P. Taylor, G. J. Hardie, and R.W. Shaw, from Comalco Aluminum Ltd., “TiB

Coated Aluminum Reduction Cells: Status and Future Direction of Coated Cells in Comalco.”

Paper presented at the Queenstown Aluminum Smelting Conference, 26 Nov 1998.

Mark P. Taylor, Gregory J. Hardie, Fiona J. Stevens McFadden, and William Uru from Comalco

Aluminum Ltd., “Use of Refractory Hard Cathodes to Reduce Energy Consumption in

Aluminum Smelting.” Paper presented at the Second International Conference on Processing

Materials for Properties, TMS 2000 Technical Program.

Use of Refractory Hard Cathodes to Reduce Energy Consumption in Aluminum Smelting.

Larry Boxall, Arthur V. Cooke, and Wayne Hayden,“TiB

Cathode Material: Application in

Conventional VSS Cells,” Journal of Light Metals 36.11 (November 1984): 35-39.

Low Energy Aluminum Reduction Cell With Induced Bath Flow, U.S. Patent 4,602,990.

Report of the American Society of Mechanical Engineers’ : Technical Working Group on Inert

Anode Technologies, 28.

Thomas M. Leeuwen, “An Aluminum Revolution,” Desk Notes, (Credit Suisse First Boston

Corporation: 22 June 2000).

Report of the American Society of Mechanical Engineers’ : Technical Working Group on Inert

Anode Technologies, Appendix A-9. Inert anode voltage equation, E = 2.922 - 0.0005712 x T,

where E is in V(dc) and T is in degrees Kelvin.

Report of the American Society of Mechanical Engineers’ : Technical Working Group on Inert

Anode Technologies, Appendix A-9: Thermodynamics of Electrochemical Reduction of

Aluminum.

Inert Anode Roadmap: A Framework for Technology Development (Washington, D.C.: The

Aluminum Association, Inc. and the U.S Department of Energy Office of Industrial

Technologies Program, Feb. 1998), 1. This document describes the potential benefits,

performance targets, technical barriers and technology development steps associated with the

R&D and production of inert anodes.

D. Bhogeswara Rao, U.V. Choudry, T.E. Erstfeld, R.J. Williams, and Y.A. Chang, “Extraction

Processes for the Production of Aluminum, Titanium, Iron, Magnesium, and Oxygen from

Nonterrestrial Sources,” Chapter in Space Resources and Space Settlements (Moffett Field,

California: NASA Ames Research Center Summer Study, 1977) 15.

Warren Haupin, “Reflections on the Hall-Héroult Process,” Editorial in Molten Salts Bulletin, ed.

59 (Marseille: Ecole Polytechnique, Universitarire de Marseille).

Craig Brown, “Next Generation Vertical Electrode Cell,” Journal on Metals 53.5 (May 2001):

39-42.

Vianey Garcia-Osorio and B. Erik Ydstie (the Department of Chemical Engineering Carnegie

Mellon University) and Tor Lindstad (SINTEF MAterials Technology), “Dynamic Model for

Vapor Recovery in Carbothermic Aluminum Process.”

Kai Johansen, Jan A. Aune, Marshal Bruno, and Anders Schei, “Carbothermic Aluminum -

Alcoa and Elkem’s New Approach Based on Reactor Technology to Meet Process

Requirements.” Paper presented at the Sixth International Congress on Molten Slags, Fluxes,

and Salts, Stockholm, June 2000.

“Toth Aluminum Corporations’s Bipolar Clay-Aluminum Technology and Project,” Seminar

Edition. Paper presented at TMS annual meeting, New Orleans, 11 February 2001.

Life Cycle Inventory Report for the North American Aluminum, E-12.

Aluminum Statistical Review for 2003, 12.

“New Technology is a Breakthrough for Automotive Aluminum Recycling,” Aluminum Now

November/December 2000: 28.

“Melt Loss Reduction in Recycling Processes,” J.H.L. Van Linden, Katholieke Universiteit

Leuven, May, 1988.

“Aluminum Dross - Liability into Opportunity,” E.L. Rooy, Light Metal Age, June, 1995.

Aluminum Statistical Review for 2003, 19-23.

Metal Melting Practices and Procedure for Efficiency and Effectiveness, 20

Metal Melting Practices and Procedure for Efficiency and Effectiveness (California Cast Metals

Association: Summer 2001), sponsored by The California Energy Commission. This and its

companion booklet Foundry Energy Use Study and Conservation Manual provide detailed,

furnace design information and operational survey data on melting and furnace practices for non-

ferrous metals in the State of California.

Applications for Advanced Ceramics in Aluminum Production: Needs and Opportunities

(Washington, D.C.: The Aluminum Association, Inc., February 2001).

Shaun Hamer, Bruno Taraglio, and Chris Romanowski, “Continuous Casting and Rolling of

Aluminum: Analysis of Capacity, Product Ranges, and Technology,” Light Metal Age 60.9-10

(October 2002): 6.

Alumina

Aluminum

Anode

Anode-Cathode Distance

(ACD)

Anode Effect

Bath

Bauxite

Bayer Process

Calcining

Carbon Dioxide Equivalents

(CDE)

Carbon Equivalents (CE)

Carbothermic Reduction

Glossary

An oxide of aluminum (Al

) and the compound from which aluminum metal is

commercially obtained.

A versatile, silvery-white metal. When exposed to the atmosphere, aluminum

rapidly forms an oxide film that prevents it from reacting with air and water. This

gives it exceptional corrosion-resistant properties. Aluminum is not found in

nature as a free metal like gold, but is chemically bound to other elements.

Aluminum is the most abundant metal in the earth’s crust (8.1 percent).

Atomic Number ... 13

Atomic Mass........ 26.982

Melting Point ...... 993.52

Boiling Point ....... 2698

A positively charged mass or surface that attracts negatively charged ions

(anions). The anode used in the Hall-Héroult Process is composed of carbon. The

oxygen containing anions react on the anode surface, releasing oxygen that

consumes the carbon to form carbon dioxide.

The geometric linear distance between the anode and the cathode is a critical

measurement in an electrolytic cell. This distance affects the voltage and energy

requirement of a cell.

An aluminum-industry idiom used to describe a process upset where the anode

reaction shifts from producing oxygen to fluorine, and the cell voltage increases.

Anode effects are primarily the result of having insufficient alumina dissolved in

the bath and available at the anode for reduction.

An aluminum-industry idiom referring to the cryolite-based electrolyte pool in the

reduction cell.

A prime source of alumina, found as a collection of small, reddish-brown nodules

in a light brown, earthy matrix. Commercial bauxite ore contains 30 to 60 weight

percent of alumina.

A process developed by Karl Bayer in 1888 that refines bauxite ore into alumina

grains. It is the process currently in use worldwide.

The process of heating a material to a sufficiently high temperature to drive off

volatile components or to oxidize the material without fusing it. The aluminum

industry uses calcining in the Bayer Process to produce alumina and to prepare

coke for anodes.

The preferred unit of measure used to compare the impact of different

Greenhouse Gases. It is calculated by multiplying the quantity of a Greenhouse

Gas emission by the Global Warming Potential (GWP) of the gas. The results are

commonly expressed in terms of a million metric tons of carbon dioxide

equivalent (10

TCDE).

A unit of measure used to compare the impact of different Greenhouse Gases

(GHG). It is calculated by multiplying the Carbon Dioxide Equivalent (CDE) by

12/44, the mass ratio of carbon to carbon dioxide.

An alternative process to electrolytic reduction. The carbothermic process reduces

alumina in a high-temperature furnace with carbon.

Castings Metal objects which are cast into a shape by pouring or injecting molten/liquid

metal into a mold. This report divides castings into ingot and shape categories.

Ingot castings are produced in molds of very simple cross-section and shape

castings are complex structures.

Cathode A negatively charged surface that attracts positively charged ions (cations). The

cathode surface in the Hall-Héroult Process is the molten aluminum pad, which

rests directly on the cell’s carbon lining. The aluminum containing cations reacts

on the cathode surface, releasing the aluminum as free metal.

Chloride Reduction An alternative process to alumina electrolytic reduction in which aluminum

chloride is used as the feed to the reduction cell.

Coke A carbon product of the crude oil refining industry. Green or raw coke contains 8

to 10 percent moisture and 5 to 15 percent volatile organic materials. Coke is

calcined in thermal kilns to remove moisture and volatile organic materials.

Cryolite Na

AlF

, a mineral that when molten dissolves alumina to form aluminum and

oxide ions. It is the main component used in the electrolyte bath for aluminum

production.

Dross The material that forms on the surface of molten aluminum as it is held in a

furnace. It is composed of impurities that have surfaced as a result of gas fluxing,

oxidized aluminum that is the result of molten aluminum exposure to the furnaces

atmosphere and aluminum that becomes entrapped in the surface material. Dross

is periodically skimmed off the surface of molten aluminum and processed to

recover its aluminum content.

Dusting An aluminum industry idiom used to describe fine carbon anode particles that are

lost in the electrolyte bath or atmosphere during electrolytic reduction. Dusting

results in a loss of productivity.

Electrolysis An electrochemical process in which the charged species in an electrolyte are

attracted to electrodes where they react with the electrons of the electrical current.

Positively charged ions migrate to the cathode and negatively charged ions

migrate to the anode.

Electrolyte A nonmetallic electrical conductor in which current is carried by the movement of

ions.

Extrusion The process of forcing the metal ingot (or billet) to flow through a die to create a

new cross-section.

Feedstock Energy

These values represent the energy inherent in a fuel that is used as material. For

example, aluminum production uses coke as the raw material in carbon anodes.

The energy contribution of a feedstock is expressed in terms of calorific or fuel

value plus the tacit/process energy used to produce the feedstock.

Global Warming Potential

(GWP)

Greenhouse gases differ in their abilities to trap heat. Global Warming Potential is

used to express the greenhouse effect of different gases in a comparable way. The

heat-trapping ability of one metric ton of CO

is the common standard, and

emissions are expressed in terms of a million metric tons of CO

equivalent or 10

TCDE.

Greenhouse Gases (GHG)

Atmospheric gases that contribute to climate change by increasing the ability of

the atmosphere to trap heat.

Hall-Héroult Process

Ingot

Kilowatt-hour (kWh)

Life Cycle Assessment (LCA)

Onsite Energy

Pad

Polarization

Pot

Potline

Potlining

Primary Aluminum

Quad

Red Mud

Reduction Cell

An electrolytic process for reduction of alumina, developed independently by

Charles Martin Hall and Paul Lewis Toussaint Héroult in 1886. This process is

commonly referred to using both names, the Hall-Héroult process. It is the pro-

cess used worldwide for commercial aluminum production.

Ingot as used in this report describes an aluminum casting of simple shape. It

includes billets, pigs, sows, T-bar and other simple cast semifinished shapes.

A unit of energy.

An internationally recognized analysis model of a product’s impact on energy,

environment, economic, and social values. LCA extends from “cradle-to-grave”:

from material acquisition and production, through manufacturing, product use

and maintenance, and finally, through the end of the product’s life in disposal or

recycling. The LCA is particularly useful in ensuring that benefits derived in one

area do not shift the impact burden to other places within a product’s life cycle.

The energy used within a facility. This is sometimes called “primary energy.”

Electrical onsite energy is the kilowatt hours used and does not include the “sec-

ondary energy” required for generation and transmission of electricity. Fuel onsite

energy use is based on the calorific heating value of the fuel and does not include

the “secondary energy” required to produce and transport the fuel.

An aluminum industry idiom used to describe the body of molten aluminum that

accumulates within the Hall-Héroult electrolytic cell.

The nonuniform concentration gradients that form near electrodes during the

reduction process. The reactions occurring at the anode and the cathode create

localized conditions that are different from the bulk of the bath. The reactions

deplete the supply of reactants and increase the quantity of products. Additionally,

in aluminum electrolysis, gas is generated at the anode which lowers the effective

bath conductivity. An electric overpotential is required to overcome the effects of

polarization.

An aluminum industry idiom used to describe an electrolytic cell. The term was

derived from the shape of the first cells.

An aluminum industry idiom that describes the arrangement of a long row of

interconnected electrolytic cells (pots).

An aluminum industry idiom that describes the refractory and carbon materials

used to line the interior of the cell (pot).

Refers to aluminum metal produced directly from alumina feedstock by chemical

reduction.

A common abbreviation for a quadrillion Btu. (1 quad = 10

Btu.)

The residue of insoluble materials that results from extracting alumina from

bauxite ore. It is also referred to as “bauxite residue.”

A container holding single or multiple anodes, cathodes and an electrolytic bath

used for reducing a material.

Reverberatory Furnace The most commonly used furnace type in the aluminum industry. The furnace is

box-shaped and consists of a steel shell with refractory lining. Fuel is fired

directly into the box either from the roof, or more typically, from the sidewall.

Heat is transferred to the molten metal with convection and radiation.

Rolling A process that results in the reduction of the cross-sectional area of a metal shape

as it is passed through rotating rolls.

Secondary Aluminum Aluminum metal that is produced from recycled aluminum products and wastes.

Tacit Energy A term used to describe an energy value that equals the combination of onsite

energy (“primary energy”) consumption, the process energy required to produce

and transmit/transport the energy source (“secondary energy”), and feedstock

energy (energy inherent in fuels used as materials). This report uses the

superscript “tf” to denote any value that includes the tacit and feedstock energy

contributions. The report does not include the energy used to make the equipment

or buildings that house the process steps (“tertiary energy”).

Urban Mining A term that describes the large source of aluminum available through urban

recycling programs as compared to bauxite mining.

Value Chain Analysis A method that captures the energy and material inputs and outputs of each

processing step (link) and builds the cumulative value for each product along the

chain. A value chain analysis or “cradle to shipping dock” analysis is an integral

part of a Life Cycle Analysis.

Appendix A: Summary of Production and Energy Data for the U.S.

Aluminum Industry

The following tables summarize the U.S. aluminum industry production and energy data developed in

this report. Energy data are based on requirements to produce a kilogram of aluminum. Process

energy is a direct measure of the energy used within a processing facility, the onsite energy. Gross

Energy

is a tacit measure of the total energy consumed and consists of the process energy plus the

energy associated with the generation and transmission of electricity, the energy content of fuel

products that are used as materials (e.g., carbon used as anodes) and the energy required to produce

fuels.

Table A.1: Theoretical, Process, and Gross Energy Requirements

Ratio of

Material to

Primary

Aluminum

Produced

kg/kg of

Aluminum

Theoretical

Minimum

Energy

Requirement

kWh/kg of

Aluminum

Process

Energy

Required

kWh/kg of

Aluminum

Overall

Process

Energy

Efficiency

percent

Gross

Energy

Required

kWh/kg of

Aluminum

Overall

Gross

Energy

Efficiency

percent

Bauxite Mining 5.10 0.00 0.32 0% 0.34 0%

Alumina Refining 1.93 0.27 7.27 4% 7.90 3%

Anode Production 0.45 4.40 5.71 77% 5.96 74%

Aluminum Smelting 1.00 5.99 15.58 38% 46.54 13%

Primary Casting 0.33 1.01 33% 1.46 23%

Secondary Casting 0.33 2.50 13% 2.81 12%

Rolling 0.32 0.63 51% 1.26 25%

Extrusion 0.44 1.30 34% 1.52 29%

Shape Casting 0.33 2.56 13% 2.64 13%

Appendix A: Summary of Production and Energy Data for the U.S. Aluminum Industry

Table A.2: United States Total Energy Requirements and Potential Savings

U.S. Theoretical U.S. Potential

Total U.S. Potential

Annual Minimum Process Process U.S.

Gross Gross U.S.

Production Energy Energy Energy

Energy

2003 Requirement Required Savings

Required Savings

metric tons

kWh (10

)

/ yr

kWh (10

)

/ yr

kWh (10

)

/ yr

kWh (10

)

/ yr

kWh (10

)

/ yr

Bauxite Mining*

Alumina Refining* 2,661,500 0.37 10.02 9.65 10.89 10.52

Anode Production 1,230,000 12.12 15.75 3.63 16.45 4.33

Aluminum Smelting 2,758,000 16.52 42.97 26.46 128.36 111.84

Primary Casting 2,704,000 0.90 2.73 1.83 3.94 3.04

Secondary Casting 2,820,000 0.94 7.05 6.11 7.93 6.99

Rolling 4,842,600 1.55 3.04 1.49 6.08 4.53

Extrusion 1,826,000 0.80 2.37 1.57 2.77 1.97

Shape Casting 2,413,000 0.80 6.17 5.36 6.37 5.56

Total 34.00 90.10 56.10 182.77 148.78

* Bauxite is no longer mined in the United States in quantities large enough for commercial

aluminum production. Bauxite is imported and refined to alumina in the United States. However, the

United States imported 53% of the alumina needed to produce aluminum in 2003 (see Appendix H).

Appendix B: Energy Intensity of Materials Produced in the United States

Aluminum metal is the most energy-intensive major product manufactured in the United States.

Aluminum, by several measures, is one of the most energy-intensive (Btu/lb) materials produced,

ranking at the top among the major products of the United States. Table B.1 shows a comparison of the

onsite energy requirements for several major products manufactured in the United States. These values

do not include the energy content of fuels used as materials or the generation and transmission losses

associated with electricity production.

Table B.1: Energy Requirements to Produce Materials in the United States (2002)

a b c a b c

Btu/yr Btu/lb lb/yr

Data Sources

†

Paper & Paper Board 2.75E+15 15,590 1.76E+11 MECS a/c MECS

Gasoline 2.41E+15 2,659 9.07E+11 b*c Drexel EIA(2)

Iron & Steel 1.79E+15 8,700 2.06E+11 b*c Steel, pg 23 Steel(A), pg 1

Ethylene 4.22E+14 8,107 5.21E+10 b*c E&E, pg 28 ACC, pg 31

Aluminum (primary ingot) 7.22E+14 44,711 1.62E+10 b*c

Appendix F,

Table F.1

Appendix G

Distillate 3.63E+14 990 3.67E+11 b*c Drexel EIA(2)

Ammonia 3.53E+14 12,150 2.90E+10 b*c E&E, pg 32 ACC, pg 31

Propylene 4.30E+13 1,351 3.18E+10 b*c E&E, pg 28 ACC, pg 31

Jet Fuel 1.46E+14 990 1.47E+11 b*c Drexel EIA(2)

Coal 1.29E+14 60 2.14E+12 b*c EIA(1) EIA(1)

Benzene 2.03E+13 1,255 1.62E+10 b*c E&E, pg 30 ACC, pg 31

†

The data sources in a,b and c columns indicate the source of values in the corresponding data columns. For

key to the abbreviations, see Sources for Tables B.1 and B.2 on page 101.

The data in Table B.2 are the onsite process energy requirements to produce the corresponding products

plus, the energy content of fuels used as materials, and the generation and transmission losses associated

with electricity production. Examples of the energy content of fuels used as materials (feedstock energy)

are: 22,681 Btu/lb is the feedstock energy for ethylene; wood products are commonly assumed to have

no feedstock value since they are renewable resources; and petroleum calcined coke used as a raw

material in aluminum production has a feedstock energy of 15,250 Btu/lb.

100

Appendix B: Energy Intensity of Materials Produced in the United States

Table B.2: Gross Energy Requirements to Produce Materials in the United States (2002)

a b c a b c

Btu/yr Btu/lb lb/yr

Data Sources

†

Paper & Paper Board 2.75E+15 15,590 1.76E+11 MECS a/c MECS

Gasoline 2.41E+15 2,659 9.07E+11 b*c Drexel EIA(2)

Iron & Steel 1.79E+15 8,700 2.06E+11 b*c Steel, pg 23

Steel(A),

pg 1

Ethylene 1.74E+15 33,470 5.21E+10 E&E, pg 28 a/c ACC, pg 31

Propylene 8.90E+14 27,978 3.18E+10 E&E, pg 28 a/c ACC, pg 31

Ammonia 6.95E+14 23,917 2.90E+10 E&E, pg 32 a/c ACC, pg 31

Benzene 5.38E+14 33,305 1.62E+10 E&E, pg 30 a/c ACC, pg 31

Aluminum (primary ingot) 5.59E+14 94,030 5.95E+09 a*b

Appendix F

Table F.1

Appendix G

Distillate 3.63E+14 990 3.67E+11 a*b Drexel EIA(2)

Coal 1.29E+14 60 2.14E+12 b*c EIA(1) EIA(1)

Jet Fuel 1.46E+14 990 1.47E+11 b*c Drexel EIA(2)

†

The data sources in a,b and c columns indicate the source of values in the corresponding data columns.

For key to the abbreviations, see Sources for Tables B.1 and B.2 below.

Sources for Tables B.1 and B.2

ACC: Guide to The Business of Chemistry, American Chemical Council, 2003

Drexel: Energy Analysis of 108 Processes, Harry Brown, Drexel University, 1996

E&E: Energy and Environmental Profile of the U.S. Chemical Industry, May 2000, DOE-OIT

EIA(1): Coal Industry Annual 2003, Energy Information Agency, DOE

EIA(2): Petroleum Annual 2003, Energy Information Agency, DOE

MECS: Manufacturing Energy Consumption Survey, Energy Information Agency, DOE

Steel : Energy Use in U.S. Steel Industry: A Historical Perspective and Future Opportunities,

Dr. J. Stubbles, Sept 2000

Steel(A): Steel Industry of the Future, FY 2004 Annual Report, DOE

101

102

Appendix C: Energy Values for Energy Sources and Materials

Calorific energy values are the energy content inherent to the material. Except for pure materials, e.g., propane, these values vary depending on the raw

materials used and the final products formulations.

Process energy is a measure of the energy required to manufacture the material. Process Energy values are also variable and depend on equipment

efficiency estimates and system boundaries. These values for crude oil-derived products are variable and depend on the specific crude processed, refinery

configuration, local product specifications and refinery efficiency. Process energy values from the California Energy Commission (CEC) include the

energy for processing, as well as factors for raw material, e.g. delivery of crude to refiners and transporting fuels to their point of use.

Tacit or gross energy is the sum of the calorific and process energy values.

Table C.1: Energy Values for Energy Sources and Materials

Input

Unit

Calorific Energy Values

"Primary Energy"

Process Energy Values

"Secondary Energy"

Tacit

Energy

Input

Unit

Btu per Data

Input Unit Source

Btu per Common Unit

Btu per Data

Input Unit Source

Btu per Common Unit

Required to Produce

Btu per

Input Unit

FUELS

†

Fuel Oil, medium

Fuel Oil, light

Diesel

Kerosene

Gasoline

Natural Gas

Bituminous/Sub

Calcined Coke

Pitch

Green Coke

Propane

Coal

139,000 Btu/gal 43,380

cogeneration.net

150,000 Btu/gal 46,813

cogeneration.net

5,670,000 Btu/bbl 35,667 EIA

5,670,000 Btu/bbl 35,667 AEO2004

5,198,000 Btu/bbl 32,698 AEO2004

1,027

Btu/ft

36,268 AEO2004

11,110 Btu/lb 24,493 CIA2000

15,250 Btu/lb 33,620 MidCon

6,065,000 Btu/bbl 38,152 E&E, pg 52

14,200 Btu/lb 31,305 MidCon

3,824,000 Btu/bbl 24,055 EGGUS

10,240 Btu/lb 22,574 AEO2004

5,000 Btu/gal 1,321 Drexel

5,200 Btu/gal 1,374 CEC

12,000 Btu/gal 3,170 CEC

Btu/ft

1,059

60 Btu/lb 132 E&E, Mining

179 Btu/lb 395 E&E, pg 52

18 Btu/lb 40 E&E, pg 52

Btu/lb 500 Drexel

Btu/lb

60 Btu/lb 132

44,700

48,130

37,040

35,870

37,330

24,630

34,010

38,190

31,810

24,050

22,710

kWh

ELECTRICITY

Electric

Avg U.S. Electric*

Primary Al Electric**

Coal-Fired Electric

kWh

1 kWh 3412 ISO

Hydroelectric Utility 0 ISO

Average U.S. Grid 6,856 tacit Btu

Primary Al Electric** 4,208 tacit Btu

Coal-Fired Utility 6,976 tacit Btu

3,412

10,270

7,620

10,390

†

Fuels and fuels used as materails

* Tacit Btu accounts for generation and transmission energy losses for average U.S. generation (see Appendix D, Table D.4)

Tacit Btu for electrolysis and anode manufacture is lower then general electricity because of the large 39.4% hydro component

Sources for Table C.1

AEO2004 - Annual Energy Outlook 2004, Energy Information Agency, Jan 2004, pg 262

CEC - California Energy Commission

CIA2000 - Coal Industry Annual 2000, Energy Information Administration, Jan 2002, pg 284. This report uses an average value for bituminous and sub-bitumenous coals.

Drexel - data derived from Drexel University study

E&E - the "Energy and Environmental Profile" for Aluminum and Mining (documents from DOE, ITP)

EIA - Emissions of Greenhouse Gases in the United States 1987-1992, Energy Information Agency, Oct 1994 (Appendix A)

ISO - Independent System Operators

EGGUS - Emissions of Greenhouse Gases in the United States 1987-1992, Energy Information Agency, Oct 1994 (Appendix A)

MidCon - data from Mid-Continent Coal & Coke Co.

The process ("secondary energy") energy values contribute about 3% additional energy to the total carbon-based fuel energy of the U.S. aluminum

industry. This includes the energy expended worldwide for the production of aluminum in the United States. The United States does not mine bauxite, but

refines it to supply 47% of the alumina needed in 2003 and imported the remaining 53% of alumina required for aluminum production.

Table C.2: kWh/yr of Fuels Consumed Worldwide for U.S. Aluminum Processing Excluding Electricity and Coke Feedstock Energy

Mining Refining Anode w/ Electrolysis Primary Secondary Hot Cold Extrusion Shape TOTAL

Feedstock Casting Casting Rolling Rolling Casting

No Process

(

"secondar

ener

)

inputs for fuels

(

kWh/

)

8.86E+08 1.95E+10 1.35E+09 5.00E+08 2.16E+09 6.73E+09 8.62E+08 6.93E+08 2.20E+09 6.16E+09

4.10E+10

With Process

(

"secondar

ener

)

inputs for fuels

(

kWh/

)

9.21E+08 2.00E+10 1.39E+09 5.14E+08 2.22E+09 6.95E+09 8.87E+08 7.17E+08 2.26E+09 6.34E+09

4.22E+10

Percent Process

(

"secondar

ener

)

fuel contributio

4% 3% 3% 3% 3% 3% 3% 3% 3% 3%

Tacit electric energy conversion factors vary significantly depending on the fuel source used to generate electricity. The large variation in electric tacit

conversion factors creates the need for careful analysis when comparing different studies. A completely coal-fired electric-based smelting operation

requires 2.5 times greater tacit energy than a completely hydroelectric smelting operation (Appendix D, Table D.1).

Table C.3: Impact of Electric Tacit Conversion Factors on kWh/yr Consumed Worldwide for U.S. Aluminum Production

Mining Refining Anode w/ Electrolysis Primary Secondary Hot Cold Extrusion Shape

Feedstock Casting Casting Rolling Rolling Casting

TOTAL

Average U.S. Aluminum Processing Facilities

10,270 Btu of energy required to produce 1 kWh of onsite electrical power for all U.S. aluminum facility operations.

kWh/yr 9.38E+08 2.18E+10 1.64E+10 1.28E+11 3.94E+09 7.93E+09 2.82E+09 3.26E+09 2.77E+09 6.37E+09

Percent increase relative to a facilities with 100% hydroelectric energy resource

% 4% 9% 4% 199% 44% 12% 87% 112% 17% 3%

1.95E+11

93%

U.S. Aluminum Processing Facilities with 100% electrical power generated by hydroelectric utilities

3,412 Btu of energy required to produce 1 kWh of onsite electrical power from hydroelectric resources for all U.S. aluminum facilitie

kWh/yr 8.99E+08 2.00E+10 1.57E+10 4.30E+10 2.73E+09 7.05E+09 1.50E+09 1.54E+09 2.37E+09 6.17E+09 1.01E+11

U.S. Aluminum Processing Facilities with 100% electrical power generated by coal-fired utilities

10,390 Btu of energy required to produce 1 kWh of onsite electrical power from coal resources for all U.S. aluminum facilities

kWh/yr 9.38E+08 2.18E+10 1.65E+10 1.33E+11 4.00E+09 7.97E+09 2.90E+09 3.36E+09 2.79E+09 6.37E+09

Percent increase relative to a facilities with 100% hydroelectric energy resource

% 4% 9% 5% 211% 47% 13% 93% 119% 18% 3%

2.00E+11

98%

Appendix C: Energy Values for Energy Sources and Materials

103

Appendix D: Hydroelectric Distribution and Electrical Energy Values

Significant energy is consumed in the generation and transmission of electricity. Tacit electric energy conversion factors

include the energy associated with production, processing and distribution of the "primary-energy" sources used in the

production of electricity. Tacit values vary significantly depending on the source of energy.

In the United States, 39.4% of aluminum smelting capacity utilizes hydroelectric power.

Table D.1: Electric Tacit Energy and Emission Data for Fuels Used for Aluminum Producion

Electrical Energy

Source

U. S. Primary Aluminum Capacity

metric tons %

Heat Rate

(2003)

Btu/kWh

Carbon Emission

Coefficient

Mt/Qbtu

Hydro

Coal

Oil

Natural Gas

Nuclear

1,633,696 39.4%

2,415,826 58.2%

8,245 0.2%

31,120 0.8%

60,112 1.4%

3,412

10,388

10,203

8,021

10,440

21.25

19.08

12.50

Total 4,149,000 100.0%

Weighted Averages based on Aluminum Capacity

Average US Grid

7,624

10,268

12.51

13.56

The distribution of electrical sources for U.S. Primary Aluminum capacity is obtained by

subtracting the Canadian capacity of 2.827 million metric tons (2002), which is 100%

hydroelectric, from the North American Totals presented in Table D.2.

Heat Rate values are derived using the Primary energy values in Appendix C, Table C.1 and the

Net Generation and Consumption values listed in the "Annual Energy Review 2003," Energy

Information Administration, pg 224.

Carbon Emission Coefficient values are derived using the Heat Rate Values in this table and the

Electric Power Industrial Sector Carbon Dioxide Emissions by Fuel Input for Year 2000 listed in

the "Emissions of Greenhouse Gases in the United States 2000," Energy Information Agency,

November 2001, page 23.

Appendix D, Table D.4

This Report Uses

10,270 Tacit Btu per kWh for all electricity consumed in U.S. aluminum

processing operations. This value overstates the actual aluminum related tacit

values (which is 7,620 Btu/kWh); but it is m

ore useful for comparing the

aluminum industry to other U.S. manufacturing operations.

104

Appendix D: Hydroelectric Distribution and Electrical Energy Values

Electrical Power Used In Primary Aluminium Production Worldwide

Table D.2: Energy Sources of Electrical Power in 2002 (electrical power used in gigawatt hours)

Electric

Energy

Source

Africa

GWh

North America

GWh %

Latin

America

GWh

Asia

GWh

Europe

GWh

Oceania

GWh

Total

GWh

Grand

Total

Hydro 6,444 50,312 64% 32,207 3,030 29,969 7,380 129,342 50%

Coal 13,443 27,248 35% 0 10,080 15,773 24,120 90,664 35%

Oil 0 93 0% 0 109 1,279 0 1,481 1%

Natural Gas 38 351 0% 1,343 16,336 6,199 0 24,267 9%

Nuclear 189 678 1% 0 0 12,672 0 13,539 5%

Total 20,114 78,682 100% 33,550 29,555 65,892 31,500 259,293 100%

Table D.3: Sources of Supply of Electrical Power in 2002 (electrical power used in gigawatt hours)

Electric Source

of Supply

Africa North America

Latin

America

Asia Europe Oceania Total

Grand

Total

GWh GWh % GWh GWh GWh GWh GWh %

Self-Generated 0 27,702 35% 4,386 28,228 9,193 1,244 70,753 27%

Purchased - Grid 20,114 39,015 50% 28,863 1,327 56,698 23,237 169,254 65%

Purchased -

Other

0 11,965 15% 301 0 1 7,019 19,286 7%

Total 20,114 78,682 100% 33,550 29,555 65,892 31,500 259,293 100%

Self-Generated

Other Purposes

0 351 253 2,662 1 0 3,267

Tables D.2 and D.3 Source:

International Aluminum Institute,

New Zealand House, Haymarket,

London SW1Y 4TE,

United Kingdom

http://www.world-aluminium.org/iai/stats

105

Appendix D: Hydroelectric Distribution and Electrical Energy Values

Significant energy is consumed in the generation and transmission of electricity. Tacit electric energy

conversion factors (Btu/kWh) include the energy associated with production, processing and

distribution of the "primary-energy" sources used in the production of electricity. Tacit values vary

significantly depending on the source of energy to produce electric power.

Table D.4: Average U.S. Grid Connection Tacit Energy

billion kWh %

(a)

U.S. Electricity Net

Generation in 2003

Units Consumed

(b)

Heat Content

(c)

Energy

Consume

Btu

(d)

Btu/kWh

(d) y (a)

Coal

Petroleum

1,970.3 51% 1,014.3 X 10

short tons 20,620,000 Btu/ton 3 20,468,000 8 10,388

30.3 X 10

bbl Distillate 5,825,000 Btu/gal 3 176,498 9

142.6 X 10

bbl Residual 6,287,000 Btu/bbl 3 896,526 9

3.4 X 10

bbl Other Liquids 5,670,000 Btu/bbl 5,3 19,278 9

6.4 X 10

short tons Coke 14,200 Btu/lb 6 181,760 9

Sub-Total 118.3 3% 1,207,000 8 10,203

Natural Gas 629.2 16% 5,380 X 10

1,019 Btu/ft

3 5,047,000 8 8,021

Nuclear 763.7 20% 10,000 Btu/kWh

7,973,000 8 10,440

Hydroelectric 275.0 7% 9,578 Btu/kWh

2,634,000 8 9,578

Renewable 84.2 2% 957 X 10

Btu 896,000 8 10,641

Other 5.1 0% 27 X 10

Btu 22,000 8 4,314

Sub-Totals

Transmission

3,845.8 100% 38,247,000 9,945

Losses 1,243,030

TOTAL

3,845.

100% 39,490,03

10,26

AER2003 - Annual Energy Review 2003. Table 8.2a Electricity Net Generation, page 224

AER2003 - Annual Energy Review 2003. Table 8.5a Consuption of Combustable Fuels for

Electricity Generation, page 238

AEO2004- Annual Energy Outlook 2004, Energy Information Agency, Jan 2004 pg 262

Assumes that a substitue for hydropower is equivalent to the average of the other sources of

electricity

Assumes diesel fuel is the majority of other fuels

Data from Mid-Continent Coal & Coke Co. for green petroleum coke

Data from Nuclear Energy Institute

AER2003 - Annual Energy Review 2003. Table 2.1f Electric Power Sector Energy

Consumption, page 43

106

107

Appendix E: Emission Data and Calculations

Table E.1 presents the carbon dioxide equivalent emission values for the fuels and materials associated with the production of

aluminum metal and aluminum products.

Table E.1: Carbon Dioxide Equivalent Emission Coefficients for Fuels Associated with Aluminum Production

FUEL

Inpu

Btu pe

Carbon Carbon Percent

PI Densit

Million

Unit input uni

Emission Emission Carbon Gravit

(EGGUS) Btu/bbl

(Appendix F) Coefficient Coefficient

(EGGUS) (EGGUS) (EGGUS)

Source

Mt/Qbt

* Pounds/gal

Emissions

per input unit

Kilogram CDE

†

FUEL

Fuel Oil, Heavy (#6)

Fuel Oil, medium

Fuel Oil, light (#2)

Diesel

Kerosene

Gasoline

Natural Gas

Bituminous/Sub

Calcined Coke

Pitch

Green Coke

Propane

Coal

kg 21.49 EGGUS 85.7 17.0 6.287

kg 44,700 20.72 a 3.40E+00

kg 48,130 19.95 b 86.3 33.9 7.064 5.825 3.52E+00

L 37,040 19.95 EGGUS 86.3 35.5 7.064 5.825 2.71E+00

L 37,040 19.72 EGG2000 86.1 41.4 5.670 2.68E+00

L 35,870 19.34 EGG2000 86.6 58.6 5.253 2.54E+00

37,330 14.47 EGG2000 1.98E+00

kg 24,630 25.81 EGGUS n/a 2.33E+00

kg 34,010 27.85 c 3.47E+00

kg 38,190 20.62 d 85.8 25.6 2.89E+00

kg 31,810 27.85 EGG2000 92.3 n/a 9.543 6.024 3.25E+00

L 24,050 17.20 EGGUS 1.52E+00

kg 22,710 25.76 EGG2000 n/a 2.15E+00

Fuel Oil, Heavy (#6)

Fuel Oil, medium

Fuel Oil, light

Diesel

Kerosene

Gasoline

Natural Gas

Bituminous/Sub

Calcined Coke

Pitch

Green Coke

Propane

Coal

Electricity

Hydro Electric

Average U.S. Electric

Primary Al Electric

Coal-Fired Electric

kWh 3,412 0.0 EGG2000 0.00E+00

kWh 10,270 13.56 EGG2000 5.11E-01

kWh 7,620 12.51 e 3.49E-01

kWh 10,390 21.25 EGG2000 8.10E-01

Hydro Electric

Average U.S. Electric

Primary Al Electric

Coal-Fired Electric

Mt/Qbtu - million metric tons per Quadrillion Btu (10

metric tons / 10

Btu)

† CDE - Carbon Dioxide Equivalent

SOURCES

EGG2000 Emission of Greenhouse Gases in the United States 2000, Energy Information Agency, Nov 2001, page 140

EGGUS Emissions of Greenhouse Gases in the United States 1987-1992, Energy Information Agency, Oct 1994, Appendix A

a Medium Fuel Oil values are the average of the light and heavy fuel oil values

b Diesel and light fuel oil is assumed to have the same carbon coefficient in EGGUS

c Green coke and Calcined coke are assumed to have the same carbon coefficient

d Pitch is assumed to have the same carbon coefficient as asphalt reported in EGGUS

e Primary Aluminum Carbon Dioxide Equivalent are based on the industry's mix of fuels (Appendix D, Table D.1)

Table E.2 uses the units of Energy Input from Table F.1 and calculates the CDE emissions for each Energy Input based on the

CDE values for fuels presented in Table E.1.

Table E.2: Carbon Dioxide Equivalent (CDE) Emissions Associated with Primary Aluminum Production

Input/Output Refining Anode ElectrolysisMining Primary Aluminum Production

Rich Soil kg

Bauxite kg

Alumina kg

Calcined Coke kg

Pitch kg

Green Coke kg

Anode kg

Fluorides kg

Aluminum kg

1,150

1,000 2,640

1,000

820

231

1,000

1,930

18.6

446

1,000

Cumulative CDE emissions

from manufacturing the raw

materials required to

produce aluminum

Total CDE emissions from

manufacturing the raw

materials and electrolytic

reduction required to

produce aluminum

Ratio to Aluminum

Fuel Oil, medium kg

Fuel Oil, light (#2) kg

Diesel L

Kerosene L

Gasoline L

Natural Gas

Bituminous/Sub kg

Calcined Coke kg

Pitch kg

Green Coke kg

Propane L

Coal kg

Electric kWh

Total kg CDE

Manufacturing portion kg CDE

Feedstock portion CDE

kg CDE / kg Al

Total Material Production in the U.

metric tons

TOTAL per 1000 kg

Energy Inputs per 1,000 kg

units kg CDE

1.160 3.94E+00

4.370 1.18E+01

0.274 6.97E-01

0.4 2.04E-01

1.67E+01

0.085

S. for Aluminum Manufacturing

Bauxite

5.10

units kg CDE

93.200 3.17E+02

1.670 4.52E+00

0.024 6.05E-02

225 4.46E+02

8.59 2.00E+01

0.00134 4.65E-03

109 5.57E+01

8.42E+02

1.626

3,488,400

Alumina

1.93

units kg CDE

3.790 1.29E+01

0.814 2.87E+00

0.110 2.98E-01

0.046 1.17E-01

97.1 1.92E+02

820 2.85E+03

231 6.67E+02

85 2.76E+02

0.136 2.06E-01

266 9.30E+01

VOC 4.99E-01

4.09E+03

3.02E+02

3.79E+03

1.83

1,230,000

Anode

0.45

units kg CDE

4.310 1.52E+01

1.840 4.99E+00

0.285 7.25E-01

7.63 1.51E+01

2.72 4.13E+00

15,400 5.38E+03

5.42E+03

5.42

2,758,000

Aluminum

1.00

units/kg Al kg CDE/kg Al

187.48 6.37E+02

0.36 1.28E+00

25.54 6.92E+01

- 0.00E+00

1.46 3.72E+00

477.56 9.46E+02

16.58 3.86E+01

365.72 1.27E+03

103.03 2.97E+02

37.91 1.23E+02

0.06 9.20E-02

1.57E+04 1.50E+02

3.54E+03

1.85E+03

1.69E+03

3.54

units/kg Al kg CDE/kg Al

187.477 6.37E+02

4.673 1.65E+01

27.378 7.42E+01

- 0.00E+00

1.748 4.45E+00

485.187 9.61E+02

16.579 3.86E+01

365.723 1.27E+03

103.026 2.97E+02

37.910 1.23E+02

2.781 4.22E+00

31,066.00 5.53E+03

8.96E+03

7.18E+03

1.69E+03

8.96

Total U.S. Production Carbon Dioxide Emissions

metric tons CDE

Total Material Production Worldwid

metric tons

Total Worldwide Emissions Based

metric tons CDE

kg CDE / kg Al

e for Use in the U.S. for Alumin

14,053,000

on U.S. Production

234,418

2,938,677

um Manufacturing

5,323,000

4,484,170

5,034,490

1,230,000

5,034,490

14,953,252

2,758,000

14,953,252

7,917,030

12,094,238

4.39

22,870,282

27,047,491

9.81

Appendix E: Emission Data and Calculations

108

Table E.3 uses the units of Energy Input from Table F.2 and calculates the CDE emissions for each Energy Input based on the

CDE values for fuels presented in Table E.1

Table E.3: Carbon Dioxide Equivalent (CDE) Emissions Associated with Aluminum Production

Appendix E: Emission Data and Calculations

109

Input/Output Casting Casting Extrusion Shape CastingHot Rolling Cold Rolling

Primary Ingot Secondary Ingot

Electrolysis Metal kg

Alloy Additives kg

Scrap kg

Primary Ingot kg

Secondary Ingot kg

Product kg

1,020

Yield:

1,000 98%

676

346

Yield:

1,000 96%

498

723 Yield:

1,000 82%

474

722 Yield:

1,000 84%

559

885 Yield:

1,000 69%

2,200 Yield:

1,000 45%

Energy Inputs per 1,000 kg

Fuel Oil, medium kg

Fuel Oil, light (#2) kg

Diesel L

Kerosene L

Gasoline L

Natural Gas

Bituminous/Sub kg

Calcined Coke kg

Pitch kg

Green Coke kg

Propane L

Coal kg

Electric kWh

TOTAL per 1000 kg

Total Material Production in the

metric tons

Total U.S. Production Carbon Di

metric tons CDE

kg of Carbon Dioxide Equivalent

Total kg CDE

kg CDE/kg Al

units kg CDE

17.4 61

0.184

0.075 0

51.8 103

0.798 1

211 73.74

2.39E+02

U.S. for Aluminum Manufac

2,704,000

oxide Emissions

646,233

per kilogram of Product

10.05

units kg CDE

42.3 149

31.464 85

12.757 32

126.0 250

1.941 3

115.000 58.73

5.78E+02

turing

2,820,000

1,629,782

0.58

units kg CDE

0.040 0.109

0.007 0

33.4 66

0.034 0

265 135.32

2.02E+02

2,421,300

488,308

0.20

units kg CDE

0.038 0

2.150 5

24.8 49

0.238 0

349 178.22

2.33E+02

2,421,300

564,819

0.23

units kg CDE

0.002 0

0.417 1

0.028 0

0.044 0

103.0 204

4.460 7

11.200 24

93 47.49

2.84E+02

1,826,000

517,866

0.28

units kg CDE

240.0 475

4 2.06

4.77E+02

2,413,000

1,151,974

0.48

e E.

prov

es compar

son o

e ore-to-meta

l C

on D

e Equ

ent

)

ons

or a mo

ern Ha

-H

rou

t ce

the CDE emissions for improved Hall-Héroult and alternative technologies.

Table E.4: Carbon Dioxide Equivalent Emissions Associated with New Aluminum Production Technologies

kWh/kg Al

kgCO

/kg Al

Typical modern Hall-Héroult

cell operating at 95% current

efficiency and with an

ACD = 4.5 cm

kWh/kg Al

kgCO

/kg Al

Typical modern Hall-Héroult

cell operating at 95% current

efficiency, retrofitted with a

sloped and wetted cathode

surface, aluminum sump, and

a reduced ACD,

ACD = 2.0 cm

kWh/kg Al

kgCO

/kg Al

Inert anode operating at 95%

current efficiency with oxygen

polarization differences and

using wetted cathode

technology, ACD = 2.0 cm

kWh/kg Al

kgCO

/kg Al

Carbothermic Reduction

with a reaction efficiency of

95% and a furnace

efficiency of 85% with

electric furnace operating

on the Average U.S. grid

kWh/kg Al

kgCO

/kg Al

Chloride Reduction of

kaolinite clays, electrolysis

current efficiency 95%,

reaction efficiency 95% and

heating efficiency 85%,

onsite electric furnace and

electrolysis cell operating on

the Average U.S. grid

Mineral Material

Tacit Energy Required

8.24 1.71 8.24 1.71 8.24 1.71 8.24 1.71 8.84 1.84

Onsite Reaction Energy Requirements

Reaction Thermal

Furnace Losses

Reaction Electrolysis

Cell Ohmic

TOTAL Reaction Energy

Anode Related Emissions

Anode Manufacturing

Anode Use

Anode Reaction

Anode Excess (Air Burning)

Total Anode

Process Related Emissions

Carbon Reactant

Perfluorocarbon

3.76

1.32

10.67

3.73

14.43

5.04

0.61

0.13

0.33 kg

1.22

0.11 kg

0.30

1.66

2.20

3.76

1.32

7.62

2.66

11.38

3.98

0.61

0.13

1.22

0.30

1.66

0.55

6.90

2.41

6.20

2.17

13.11

4.58

0.77

0.27

0.00

7.71

3.94

1.93

0.98

9.63

4.92

0.66 kg

2.45

0.00

-1.90

-0.97

0.40

0.20

6.48

3.31

2.93

1.49

7.91

4.04

0.89 kg

3.27

0.00

TOTAL CO

Emissions

Mineral Material

Reactions

Carbon

Process

TOTAL

1.71

5.04

1.66

2.20

10.62

1.71

3.98

1.66

0.55

7.90

1.71

4.58

0.00

6.29

1.71

4.92

2.45

0.00

9.08

1.84

4.04

3.27

0.00

9.15

Appendix E: Emission Data and Calculations

110

- - - -

111

Appendix F: U.S. Energy Use by Aluminum Processing Area

The energy input units and raw material quantities used in Tables F.1 and F.2 are from the Life Cycle Inventory Report for the North American Aluminum

Industry published by the Aluminum Association in November, 1998. The report is the result of extensive surveying and contains the best and most

complete industry performance information of any recent study. The report is consistent with ISO procedures, and was favorably “peer” reviewed by groups

outside the industry. The Aluminum Association has provided the authors with permission to use these data.

The conversion of input units to common energy units are based on the values presented in Appendix C, Table C.1.

Table F.1: Materials and Energy Associated with Primary Metal Electrolysis

Input/Output Mining Refining Anode Electrolysis Primary Aluminum

Rich Soil kg

1,150

Bauxite kg

1,000 2,640

Cumulative energy

Total energy requirement

Alumina kg 1,000

1,930

requirement for

for manufacturing the raw

Calcined Coke kg

820

manufacturing the raw

materials and electrolytic

Pitch kg

231

materials required to

reduction required to

Green Coke kg

produce aluminum

Anode kg

1,000 446

Fluorides kg

18.6

Aluminum kg

1,000

Bauxite Alumina Anode Aluminum

Ratio to aluminum

5.10 1.93 0.45 1.00

Energy Inputs per 1,000 kg

units Btu units Btu units Btu units Btu units/ kg Al Btu/kg Al units/kg Al Btu/kg Al

of Material Produced

Fuel Oil, medium kg 1.160 50,321 93.200 4,043,057 3.790 164,412 187.48 8,132,824 187.477 8,132,824

Fuel Oil, light kg

0.814 38,106 4.310 201,766 0.36 16,995 4.673 218,761

Diesel L

4.370 155,865 1.670 59,564 0.110 3,923 1.840 65,627 25.54 910,873 27.378 976,501

Kerosene L

Gasoline L

0.274 8,959 0.024 778 0.046 1,504 0.285 9,319 1.46 47,822 1.748 57,141

Natural Gas

225 8,160,344

97.1 3,521,642 7.63 276,726 477.56 17,320,117 485.187 17,596,843

Bituminous/Sub kg

8.59 210,396 16.58 406,064 16.579 406,064

Coke kg

0.00134 45 0.003 87 0.003 87

Propane L

0.136 3,271 2.72 65,429 0.06 1,459 2.781 66,888

Coal kg

Electric kWh

0.4 1,365 109 371,908 266 907,592 15,400 52,544,800 331.04 1,129,522 15,731.04 53,674,322

TACIT electric kWh 1.2 328 801 46,353 1,130 47,483

TOTAL ENERGY INPUTS (Energy per kg of Material Output)

Note: The onsite or "primary" electrical energy

kWh/kg

Btu/kg 217 12,846 4,640 53,164

requirement to produce aluminum from ore accounts for

0.06 3.76 1.36 15.58

65% of the total energy requirement

TACIT kWh/kg 0.07 4.09 1.93 46.54

Ratio Adjusted ENERGY INPUTS (Energy per kg of Aluminum)

Btu/kg 1,103 24,793 2,070 53,164 27,966 81,129

kWh/kg

0.32 7.27 0.61 15.58 8.20 23.78

TACIT kWh/kg 0.34 7.90 0.86 46.54 9.10 55.64

Feedstock Energy per Kg Al (Energy Inherent in Fuels used as Materials) Feedstock Ratio adjusted

Btu/kg 17,414 98,544

kWh/kg 5.10 28.88

TACIT kWh/kg Total Tacit (process 0.80 + feedstock 5.10) 6.02 14.26 60.80

Appendix F: U.S. Energy Use by Aluminum Processing Area

Table F.2: Materials and Energy Associated with Aluminum Manufacturing Operations

Input/Output Casting

Primary Ingot

Casting

Secondary Ingot

Hot Rolling Cold Rolling Extrusion Shape Casting

Electrolysis Metal kg

Alloy Additives kg

Scrap kg

Primary Ingot kg

Secondary Ingot kg

Product kg

1,020

Yield:

1,000 98%

676

346

Yield:

1,000 96%

498

723 Yield:

1,000 82%

474

722 Yield:

1,000 84%

559

885 Yield:

1,000 69%

2,200 Yield:

1,000 45%

Energy Inputs per 1,000 kg

Fuel Oil, medium kg

Fuel Oil, light kg

Diesel L

Kerosene L

Gasoline L

Natural Gas m3

Bituminous/Sub kg

Coke kg

Propane L

Anthracite kg

Electric kWh

units Btu

17.4 814,554

0.184 6,563

0.075 2,439

51.8 1,878,693

0.798 19,196

211 719,932

units Btu

42.3 1,981,346

31.464 1,122,230

12.757 417,115

126.0 4,569,793

1.941 46,692

115 392,380

units Btu

0.040 1,441

0.007 249

0.007 213

33.4 1,211,358

0.034 818

265 904,180

units Btu

0.038 1,362

2.150 70,301

24.8 899,451

0.238 5,725

349 1,190,788

units Btu

0.002 109

0.417 14,873

0.028 988

0.044 1,442

103.000 3,735,624

4.460 107,285

11.200 252,829

93 317,316

units Btu

240.000 8,704,367

4 13,750

TOTAL ENERGY INPUTS (Energy per kg of Output)

Btu/kg

kWh/kg 1.01

3,441

2.50

8,530

0.62

2,118

0.64

2,168 4,430

1.30

8,718

2.56

TACIT ENERGY INPUTS (Energy per kg of Output)

Btu/kg

kWh/kg

4,967

1.46

9,591

2.81 1.16

3,971

1.35

4,594

1.52

5,180

2.64

9,001

112

Appendix F: U.S. Energy Use by Aluminum Processing Area

113

Table F.3 lists the theoretical minimum energy requirements to produce raw materials and aluminum products. It also shows the process efficiency for

each operation. Mining, refining, anode manufacturing and electrolysis minimum energy values are based on the net chemical changes that result from

these processes. Primary casting, secondary casting, and shape casting minimum energy values are based on the energy required to produce molten pure

aluminum at 775 °C. The minimum energy requirements for rolling and extrusion operations are estimated from their yield and onsite energy

consumption values in Table F.2, and from an overall assumed electric and hydraulic system efficiencies of 75% and thermal furnace and heating

efficiencies of 50%.

Table F.3: Theoretical Minimum Energy Requirements to Produce Raw Materials and Aluminum Products

Mining Refining Anode

w/ Feedstock

Electrolysis Primary

Casting

Secondary

Casting

Hot

Rolling

Cold

Rolling

Extrusion Shape

Casting

kWh/kg

Ratioed to Al kWh/kg Al

Current Practice Onsite

Process Efficiency %

0.14

0.27

12.97

4.40

77%

5.99

38%

0.33

33%

0.33

13%

0.31

50%

0.33

52%

0.44

34%

0.33

13%

Table F.4 lists U.S. production quantities, and energy and tacit energy consumption associated with producing aluminum products within the United

States. The United States does not consume energy to produce metallurgical bauxite, and consumes only 79% of the energy required for alumina. This

report distinguishes between the worldwide energy values and the United States values in order to measure the impact of market or process changes to the

energy demands within the United States. Table F.9 lists the worldwide values. The theoretical magnitude of the potential U.S. energy savings can be

measured by subtracting the theoretical energy requirements from the actual energy consumption numbers.

Table F.4: U.S. Energy Use in the Production of Domestic Aluminum

Mining Refining Anode

w/ Feedstock

Electrolysis Primary

Casting

Secondary

Casting

Hot

Rolling *

Cold

Rolling *

Extrusion * Shape

Casting

TOTAL

ENERGY

Material Production within the United States

metric tons (2003) 0 3,488,400 1,230,000 2,758,000 2,704,000 2,820,000 2,421,300 2,421,300 1,826,000 2,413,000

On-Site Energy Consumed in Production for the U.S. Aluminum Industry

kWh/kg

ratioed* to Al kWh/kg Al

Total kWh/yr

Proportion of U.S. Energy Used

3.76

7.27

1.31E+10

14%

12.80

5.71

1.57E+10

17%

15.58

4.30E+10

46%

1.01

2.73E+09

80%

2.50 0.62 0.64

7.05E+09 1.50E+09 1.54E+09

8% 2% 2%

Total U.S. Primary proportion

1.30

2.37E+09

2.56

6.17E+09

9.32E+10

100%

Total TACIT Energy Consumed in Production for the U.S. Aluminum Industry

kWh/kg

ratioed* to Al kWh/kg Al

Total kWh/yr

Proportion of U.S. Energy Used

4.09

7.90

1.43E+10

7.7%

13.37

5.96

1.64E+10

8.8%

46.54

1.28E+11

69.0%

1.46

3.94E+09

2.1%

87.6%

2.81 1.16 1.35

7.93E+09 2.82E+09 3.26E+09

4.3% 1.5% 1.8%

Total U.S. Primary proportion

1.52

2.77E+09

1.5%

2.64

6.37E+09

3.4%

1.86E+11

100%

Theoretical Magnitude of Opportunities for U.S. Energy Savings

kWh/yr

Tacit kWh/yr

1.26E+10

1.38E+10

-1.99E+08

4.98E+08

2.65E+10

1.12E+11

1.83E+09

3.04E+09

6.11E+09

6.99E+09

7.56E+08

2.07E+09

7.36E+08

2.46E+09

1.57E+09

1.97E+09

5.36E+09

5.56E+09

5.53E+10

1.48E+11

* Data is from 2000. These are the latest numbers available that are specific to the U.S. production. After 2000, the numbers are based on North American data and not specifically the U.S. data.

Table F.5 shows the total energy values consumed by the U.S. aluminum industry and the energy savings possible if:

a it was possible to reduce energy consumption to the theoretical limit,

b. energy were reduced by 30%, or

c. electrolysis electrical energy were reduced from 15 to 11 kWh/kg of aluminum

Table F.5: Total U.S. Aluminum Industry Energy Consumption and Potential Savings

U.S. Aluminum Industr

kWh/y

Quads/y

MW Households bbl crude

per yea

Total Energy Use on-site 9.32E+10 0.32 10,600 10,600,000 54,900,000

tacit 1.86E+11 0.64 21,300 21,300,000 109,600,000

Theoretical Requirement on-site 3.79E+10 0.13 4,300 4,300,000 22,400,000

tacit 3.79E+10 0.13 4,400 4,400,000 22,400,000

Energy Efficiency on-site 41%

tacit 20%

a. Theoretical Opportunity

on-site 5.53E+10 0.19 6,300 6,300,000 32,600,000

tacit 1.48E+11 0.51 16,900 16,900,000 87,200,000

b. Practical Goal (Vision)

on-site 2.80E+10 0.10 3,180 3,200 16,500,000

tacit 5.58E+10 0.19 6,390 6,400 32,900,000

c. Electrolysis Savings of 4 kWh/kg

on-site 1.10E+10 0.04 1,259 1,300 6,500,000

tacit 2.46E+10 0.08 2,813 2,900 14,500,000

Table F.6 shows 94% reduction in energy consumption associated with secondary metal production. Recycling aluminum essentially recaptures all the energy

associated with mining, refining and smelting.

Table F.6: Energy Impact of Recycling

Energy Saved with Recycling

(Tacit Energy Values)

62.2 kWh/kg to produce Primary Metal Ingot

2.8 kWh/kg to produce Secondary Metal Ingot

4.5% Secondary to Primary Energy

2,820,000 kg of Secondary Metal produced in 2003

1.67E+11 kWh/yr Energy Saved in 2003

0.57 Quads Saved per Year 2003

19,100 MW SAVED 2003

Appendix F: U.S. Energy Use by Aluminum Processing Area

114

Table F.7 shows the electric energy consumption of the U.S. aluminum industry and compares it to the total electric energy produced in the United States.

Table F.7: U.S. Electric On-site Energy Consumption in the Aluminum Industry

Appendix F: U.S. Energy Use by Aluminum Processing Area

115

Refining Anode Electrolysis Primary Secondary Hot Cold Extrusion Shape TOTAL

w/ Feedstoc

Casting Casting Rolling Rolling Casting ELECTRIC

ENERGY

kWh/kg

kWh/yr

metric tons (2003)

3,488,400 1,230,000 2,758,000 2,704,000 2,820,000 2,421,300 2,421,300 1,826,000 2,413,000

0.109 0.266 15.400 0.211 0.115 0.265 0.349 0.093 0.004

3.80E+08 3.27E+08 4.25E+10 5.71E+08 3.24E+08 6.42E+08 8.45E+08 1.70E+08 9.72E+06 4.57E+10

quad 0.16

Total U.S. Al Industry Electrical Use: 4.57E+10 kWh/yr

on-site 5,222 MW

5,222,000 households Net U.S. Generation: 3.858E+12 kWh/yr (Source: eia.doe.gov/cneaf/electricity)

tacit 54,000,000 bbl crude

luminum's % of U.S. Net: 1.2

Electrolysis and process heating are the two most energy-intensive process areas of the aluminum industry. Electrolysis accounts for 46% of all onsite and

69% of all tacit energy use. Process heating accounts for 27% of all onsite and 13% of all tacit energy use in the industry.

Table F.8: Smelting and Heating Fractions of Total U.S. Aluminum Industry Energy Consumed

Smelting Fraction of Total U.S. Aluminum Industry Energy Consumed

Electrolysis

Total

Industr

Percent

of Industr

Onsite kWh/yr 4.30E+10

Tacit kWh/yr 1.28E+11

9.32E+10

1.86E+11

46%

Aluminum Metal Process Heating/Melting Fraction of Total U.S. Aluminum Industry Energy Consumed

Casting Extrusion Shape Total

Hot Cold Casting Process

Heating

Rolling

Total

Industr

Percent

of Industr

Btu/kg Al 9,244 1,212 905 4,096 8,704 2.42E+04

kWh/kg Al 2.71 0.36 0.27 1.20 2.55 7.08E+00

Onsite kWh/yr 1.50E+10 8.60E+08 6.42E+08 2.19E+09 6.16E+09 2.48E+10

Tacit

Note: The tacit increase for process heating is negligible

9.32E+10

1.86E+11

Table F.9 lists the worldwide production quantities, energy, and tacit energy consumption associated with producing aluminum products within the United

States. These values provide a full measure of the energy associated with producing aluminum and aluminum products. The actual energy consumed within

the United States is lower since no aluminum metallurgical bauxite is mined in the United States, and only 47% of the alumina required is refined in the

United States.

Table F.9: Total Worldwide Production and Energy Consumption Associated with Producing Aluminum in the United States

* ratios to primary aluminum

Mining*

5.10

Refining*

1.93

Anode*

w/ Feedstock

0.45

Electrolysis*

1.00

Primary

Casting

Secondary

Casting

Hot

Rolling

†

Cold

Rolling

†

Extrusion

†

Shape

Casting

TOTAL

ENERGY

Total Worldwide Production Required to Produce Aluminum in the United States

metric tons (2003) 14,053,000 5,323,000 1,230,000 2,758,000 2,704,000 2,820,000 2,421,300 2,421,300 1,826,000 2,413,000

Total On-Site Energy Consumed in Worldwide Production for U.S. Aluminum Industry

kWh/kg

ratioed* to Al kWh/kg Al

Total kWh/yr

% Relative to Electrolysis*

Proportion of Energy Used

0.06

0.32

8.92E+08

2.1%

0.9%

3.76

7.27

2.00E+10

46.6%

19.8%

12.80

5.71

1.57E+10

36.6%

15.6%

15.6

4.30E+10

100.0%

42.5%

1.01

2.73E+09

6.3%

2.7%

82%

2.50 0.62 0.64

7.05E+09 1.50E+09 1.54E+09

16.4% 3.5% 3.6%

7.0% 1.5% 1.5%

Total Primary proportion

1.30

2.37E+09

5.5%

2.3%

2.56

6.17E+09

14.3%

6.1%

1.01E+11

100.0%

Total TACIT Energy Consumed in Worldwide Production for the U.S. Aluminum Industry

kWh/kg

ratioed* to Al kWh/kg Al

Total kWh/yr

% relative to Electrolysis

Proportion of Energy Used

0.07

0.34

9.38E+08

0.7%

0.5%

4.09

7.90

2.18E+10

17.0%

11.2%

13.37

5.96

1.64E+10

12.8%

8.5%

46.5

1.28E+11

100.0%

66.0%

1.5

3.94E+09

3.1%

2.0%

88%

2.8 1.2 1.3

7.93E+09 2.82E+09 3.26E+09

6.2% 2.2% 2.5%

4.1% 1.4% 1.7%

Total Primary proportion

1.5

2.77E+09

2.2%

1.4%

2.6

6.37E+09

5.0%

3.3%

1.95E+11

100.0%

† Data is from 2000. This is the last year numbers specific to U.S. production are available. After 2000, the numbers are based on North American data, and not specifically U.S. data.

Table F.10 divides energy uses in aluminum casting and semi-fabrication manufacturing operations into three areas: electric, heating and miscellaneous fuels.

The following percentages of energy use are derived from the energy values listed in Table F.2. Fuel oils, natural gas and propane are assumed to be

associated with heating (thermal operations). Miscellaneous fuels include all other fuels and it is assumed that these are fuels used in non-heating operations,

such as transportation and collection of materials. (Note: rounding causes some totals to differ from 100%)

Table F.10: Percent Break-Up of Different Types of Energy Used in Aluminum Production Operations

Primary Secondary Hot Cold Extrusion Shape

Casting Casting Rolling Rolling Casting

Onsite: % onsite electric

% heating

miscellaneous fuels

21% 5% 43% 55% 7% 0%

79% 77% 57% 42% 87% 100%

0% 18% 0% 3% 6% 0%

Tacit: % tacit electric

% heating

miscellaneous fuels

36% 9% 62% 72% 72% 14%

63% 73% 38% 25% 25% 80%

0% 17% 0% 3% 3% 6%

Appendix F: U.S. Energy Use by Aluminum Processing Area

116

Appendix G: U.S. Primary, Secondary and Imported Aluminum

Quantities, 1960-2003

Table G.1: U.S. Supply of Aluminum from 1960 to 2003 (in Thousand Metric Dry Tons)

Year

U.S. Primary

Aluminum

U.S. Secondary

Aluminum

U.S. Aluminum

Imports

U.S. Total

Aluminum

1960 1,828 401 178 2,407

1961 1,727 445 230 2,402

1962 1,921 533 341 2,795

1963 2,098 601 419 3,118

1964 2,316 648 409 3,373

1965 2,499 774 543 3,816

1966 2,693 833 591 4,117

1967 2,966 821 466 4,253

1968 2,953 935 684 4,572

1969 3,441 1,067 484 4,992

1970 3,607 937 406 4,950

1971 3,561 1,004 582 5,147

1972 3,740 1,022 684 5,446

1973 4,109 1,127 523 5,759

1974 4,448 1,163 511 6,122

1975 3,519 1,121 453 5,093

1976 3,857 1,334 608 5,799

1977 4,117 1,456 683 6,256

1978 4,358 1,518 907 6,783

1979 4,557 1,612 711 6,880

1980 4,653 1,577 603 6,833

1981 4,489 1,790 782 7,061

1982 3,274 1,666 823 5,763

1983 3,353 1,773 1,023 6,149

1984 4,099 1,760 1,376 7,235

1985 3,500

1,762 1,332 6,594

1986 3,039 1,773 1,843 6,655

1987 3,347 1,986 1,702 7,035

1988 3,945 2,122 1,467 7,534

1989 4,030 2,054 1,353 7,437

1990 4,048 2,393 1,421 7,862

1991 4,121 2,286 1,398 7,805

117

Appendix G: U.S. Primary, Secondary and Imported Aluminum Quantities, 1960-2003

Table G.1: U.S. Supply of Aluminum from 1960 to 2003 (in Thousand Metric Dry Tons) (Continued)

Year

U.S. Primary

Aluminum

U.S. Secondary

Aluminum

U.S. Aluminum

Imports

U.S. Total

Aluminum

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

4,042

3,695

3,299

3,375

3,577

3,603

3,713

3,779

3,668

2,637

2,705

2,704

2,756

2,944

3,086

3,188

3,307

3,547

3,442

3,695

3,450

2,970

2,927

2,820

1,573

2,327

3,136

2,701

2,572

2,804

3,264

3,680

3,580

3,487

3,947

4,068

8,371

8,966

9,521

9,264

9,456

9,954

10,419

11,154

10,698

9,094

9,579

9,592

Average Growth Rate 1993 through 2003

-2.5% -0.4% 5.1% 0.6%

Percent of Total Supply (2003)

28% 29% 42% 100%

Source: Aluminum Statistical Review for 2003, The Aluminum Association, 2004, p. 7

The growth rates utilized in this report are based on a linear regression of data over ten year period

from 1993 through 2003. Linearizing the data provides a better value for average growth over the

period than a simple comparison of 1993 to 2003.

Primary Secondary Imports Supply

Linear regression for 1993 through 2003

slope

intercept

-83.5 -13.4 156.4

170,211 30,026 -309,163

59.4

-108,927

Linearized values

1993

2003

3,759 3,283 2,452

2,924 3,149 4,015

9,494

10,088

Linearized growth from 1993 to 2003

Avg Growth -2.5% -0.4% 5.1%

U.S. Aluminum Production Growth (Primary & Secondary Combined)

0.6%

-1.47%

118

Appendix H: U.S. Bauxite and Alumina Quantities, 1960-2003

Table H.1: U.S. Supply of Bauxite and Alumina from 1960 to 2003 (in Thousand Metric Dry Tons)

A B C D E F G H I

Year

Imports

of Bauxite

U.S.

Mined

Bauxite

Exports of

Bauxite

Net

Bauxite

Supply

Estimated

Alumina

Production

from

Bauxite

Imports

Alumina

Exports

Alumina

Estimated

Net

Alumina

for

Electrolysis

1960 8,879 2,030 29 10,880 4,651 80 4,266

1961 9,354 1,248 153 10,449 4,467 171 4,191

1962 10,745 1,391 263 11,873 5,076 158 4,726

1963 9,408 1,549 206 10,751 4,596 173 4,309

1964 10,518 1,627 283 11,862 5,071 191 4,755

1965 11,601 1,681 149 13,133 5,614 206 290 4,969

1966 11,928 1,825 63 13,690 5,852 443 290 5,420

1967 12,010 1,681 2 13,689 5,852 865 499 5,633

1968 11,359 1,692 7 13,044 5,576 1,190 780 5,429

1969 12,355 1,873 5 14,223 6,080 1,730 885 6,317

1970 13,039 2,115 3 15,151 6,477 2,340 998 7,171

1971 12,837 2,020 35 14,822 6,336 2,190 980 6,913

1972 12,803 1,841 29 14,615 6,248 2,590 797 7,416

1973 13,618 1,909 12 15,515 6,633 3,090 694 8,365

1974 15,216 1,980 16 17,180 7,344 3,290 927 8,973

1975 11,714 1,801 20 13,495 5,769 3,180 933 7,439

1976 12,749 1,989 15 14,723 6,294 3,290 1,050 7,905

1977 12,989 2,013 26 14,976 6,402 3,760 857 8,665

1978 13,847 1,669 13 15,503 6,628 3,970 878 9,057

1979 13,780 1,821 15 15,586 6,663 3,840 849 8,988

1980 14,087 1,559 21 15,625 6,680 4,360 1,140 9,232

1981 12,802 1,510 20 14,292 6,110 3,980 730 8,749

1982 10,122 732 49 10,805 4,619 3,180 590 6,747

1983 7,601 679 74 8,206 3,508 4,030 602 6,585

1984 9,435 856 82 10,209 4,364 4,290 648 7,570

1985 7,158 674 56 7,776 3,324 3,830 316 6,506

1986 6,456 510 69 6,897 2,948 3,600 487 5,767

1987 9,156 576 201 9,531 4,075 4,070 1,130 6,607

1988 9,944 588 63 10,469 4,475 4,630 1,040 7,618

1989 10,893 W 44 10,849 4,638 4,310 1,330 7,154

119

Appendix H: U.S. Bauxite and Alumina Quantities, 1960-2003

Table H.1: U.S. Supply of Bauxite and Alumina from 1960 to 2003 (in Thousand Metric Dry Tons) (Continued)

A B C D E F G H I

Year

Imports

of Bauxite

U.S.

Mined

Bauxite

Exports of

Bauxite

Net

Bauxite

Supply

Estimated

Alumina

Production

from

Bauxite

Imports

Alumina

Exports

Alumina

Estimated

Net

Alumina

for

Electrolysis

1990 12,142 W 53 12,089 5,168 4,070 1,260 7,461

1991 12,300 W 58 12,242 5,233 4,590 1,350 7,950

1992 11,400 W 68 11,332 4,844 4,700 1,140 7,920

1993 11,900 W 92 11,808 5,048 3,940 1,240 7,243

1994 11,200 W 137 11,063 4,729 3,120 1,040 6,336

1995 10,800 W 120 10,680 4,566 4,000 1,040 7,069

1996 10,700 W 154 10,546 4,508 4,320 918 7,460

1997 11,300 W 97 11,203 4,789 3,830 1,270 6,870

1998 11,600 W 108 11,492 4,913 4,050 1,280 7,192

1999 10,400 W 168 10,232 4,374 3,810 1,230 6,517

2000 9,030 W 147 8,883 3,797 3,820 1,090 6,148

2001 8,670 W 88 8,582 3,669 3,100 1,250 5,152

2002 7,710 W 52 7,658 3,274 3,010 1,270 4,686

2003 8,300 W 140 8,160 3,488 2,300 1,000 4,310

W - data withheld

Source: United States Geological Survey, Minerals Information, Statistical Compendium

Note: 3,488,000 metric tons or 79% of the 4,310,000 metric tons of alumina needed to produce

aluminum in the United States was refined in the United States in 2003.

Calculations

Column F is calculated based on USGS data. The average alumina content for bauxite is 45%, and

95% of bauxite is converted to alumina. The calculation is F = 0.95 x 0.45 x E. The numbers produced

in Column F may be slightly over-estimating since slightly less than 100% of the alumina content of

bauxite is actually recovered.

Column I is calculated based on USGS data. 90% of alumina produced is metallurgical alumina used

in aluminum production. The calculation is I = 0.9 x ( F + G - H).

120

Appendix I: Energy Requirements for Carbon Anodes

The following table shows the onsite, process and feedstock energy values that are part of making a carbon electrode (anode)

for reduction of alumina to aluminum. Pitch and coke require process energy for their manufacture, and they have a fuel or

feedstock energy value that must be accounted for in order to fully evaluate the energy associated with producing aluminum.

Table I.1: Energy Associated with Aluminum Industry Carbon Anode Manufacturing

Mass of Material

Input

kg/1000 kg Anode

Material Input

Energy

Btu/kg

kWh per kg

Anode

kWh per kg

Aluminum

Pitch

Mass 231

Feedstock Energy 8,813 2.58 1.15

Process Energy

Calcined Coke

9 0.003 0.001

Mass 820

Feedstock Energy 27,569 8.08 3.60

Process Energy

Green Coke

395 0.12 0.05

Mass 85

Feedstock Energy 2,664 0.78 0.35

Process Energy

Carbon Anode Baking

43 0.01 0.01

Process Energy

TOTAL Raw Materials and Energy

1.93 0.86

Mass (kg/1000 kg Anode) 1,136

Feedstock Energy (kWh) 39,046 11.44 5.10

Process Energy (kWh) 446 2.06 0.92

TOTAL Tacit Energy Input 13.50 6.02

Values from Appendix F, Table F.1

Values from Appendix C, Table C.1

Table I.2: Onsite and Tacit Energy Associated with Carbon Anode Production

kWh per kg of Anode

kWh per kg of Aluminum

Pitch

Coke

Baking

Total

0.003 2.58

0.13 8.99

1.36 1.93

1.49 13.50

0.001 1.15

0.06 4.01

0.61 0.86

0.66 6.02

121

Appendix J: Theoretical Energy Data and Calculations

The theoretical minimum energy requirement for producing any chemical is determined based on the net chemical reaction used

to produce the product. It is defined as the energy required to synthesize a substance in its standard state from substances also in

their standard states. It can be calculated by summing the reaction energies of the products minus the energies of the reactants.

This report calculates the theoretical minimum energy by assuming the reactants enter and the byproducts leave the system at

room temperature and that molten aluminum leaves the system at 960 ºC. This report has chosen 960 ºC (1233 ºK) as the molten

metal temperature. This value is an approximation of the average operating cell temperatures of industrial cells.

Some studies assume that the gases evolved during reduction leave the system at the molten metal temperature. In these studies,

the theoretical minimum is 2.5% to 3% higher than the numbers calculated here. The additional energy for heating of the

emissions is shown in the individual tables. Theoretically, it is possible to capture all the energy associated with these gaseous

emissions. However, there is currently no available, economic means of recovering this energy. In practice, the emission gas

stream is diluted with air in the cells to lower the temperature that the cell hoods and ducts are exposed to. The emission gas is

collected from hundreds of hooded pots and treated before release to the atmosphere. Only a very small portion of the heat is

actually absorbed and returned to the system.

The minimum theoretical energy requirement for aluminum production requires the evaluation of three energy factors: the

energy required to drive the reduction reaction forward, the energy required to maintain the system at constant pressure and

temperature, and the energy required to change the temperature of the reactant and/or product. The thermodynamics and

chemical equilibrium of reactions are described by the equation ∆G = ∆H - T ∆S and the numeric values are given in Table J.6.

The energy required to drive the reaction forward is the Gibbs free energy value (∆G). The energy required to maintain system

equilibrium is the difference between the heat of reaction (∆H) and the Gibbs free energy value (∆G), which equals the entropy

term (T∆S). Since the Gibbs free energy requirement is less than the heat of reaction for alumina reduction, additional energy

must be added to the system to maintain the system temperature. Otherwise, the system would cool as the reaction proceedes.

(Reduction cells operate at atmospheric conditions and no pressure change results during reduction.) The numeric values for G,

H and S are given in Table J.6.

A detailed discussion of the theoretical requirements is made in Current and Energy Efficiency of Hall-Héroult Cells - Past,

Present and Future, by Warren Haupin and William Frank, published in Light Metal Age, June 2002.

122

Table J.1: Theoretical Minimum Energy for Hall Hé

oult Carbon Anode System

(products

–

Reactants Products Reaction Thermodynamics at 298

reactants)

Temp Temp cal/gm mole Al

( 2 Al

+ 3 C ) to ( 4 Al +

3 CO

)

= Net

2 Al

4 Al

delta G -756,358 0 0 -282,779 118,395

25°C

960q

delta H -801,000 0 0 -282,155 129,711

Carbon delta S 24.3 4.1 27.1 153.2 11,311

Anode Process Theoretical Minimum Energy Requirements

kWh/kg

3 C

3 CO

2Cell Electrolytic Work Requirement (delta G) 5.11

25°C

Thermal Energy for Temperature Maintenance (delta H - delta G) 0.49

Thermal Energy for Al at 960°C 0.39

Theoretical Minimum 5.99

Thermal Energy for CO

at 960°C

0.17

Note: Thermodynamic values for G, H and S are from Table J.6. Heat capacity data are from Table J.9 and Appendix K

2 Al

960°C

25°C

Table J.2: Theoretical Minimum Energy for Hall Héroult Inert Anode System

(products

–

Reaction Thermodynamics at 298

reactants)

Reactants Products cal/gm mole Al

Temp Temp

( Al

) to ( 2 Al +

1.5 O

)

= Net

delta G -378,179 0 0 189,089

delta H -400,500 0 0 200,250

delta S 12 14 74 11,155

Inert

Process Theoretical Minimum Energy Requirements

Anode kWh/kg

Electrolytic Work Requirement (delta G) 8.16

Cell

O1.5

Thermal Energy for Temperature Maintenance (delta H - delta G) 0.48

Thermal Energy for Al at 960°C 0.39

Theoretical Minimum 9.03

Thermal Energy for O

at 960°C

0.27

Note: Thermodynamic values for G, H and S are from Table J.6. Heat capacity data are from Table J.9 and Appendix K

123

Appendix J: Theoretical Energy Data and Calculations

Table J.3: Theoretical Minimum Energy for Carbothermic Reduction

(products

–

Reactants Products Reaction Thermodynamics at 298

reactants)

Temp Temp

( Al

+ 3 C ) to ( 2 Al + 3 CO ) = Net

cal/gm mole Al

2 Al

delta G -378,179 0 0 -98,424 139,878

25°

C 960°C

delta H -400,500 0 0 -79,247 160,626

Carbo- delta S 12.2 4.1 13.5 141.9 21,347

Thermic

Process Theoretical Minimum Energy Requirements

Reactor kWh/kg

3 C

3 CO

Work Requirement (delta G) 6.03

25°C 25°C

Thermal Energy for Temperature Maintenance (delta H - delta G) 0.90

Thermal Energy for Al at 960 C 0.39

Theoretical Minimum 7.32

Thermal Energy for CO at 960°C 0.19

Note: Thermodynamic values for G, H and S are from Table J.6. Heat capacity data are from Table J.9 and Appendix K

Table J.4: Theoretical Minimum Energy for Reduction of Aluminum Chloride

(products

–

Reaction Thermodynamics at 298 K reactants)

Reactants Products

( 2 AlCl

) to ( 2 Al + 3 Cl

) = Net

cal/gm mole Al

Temp Temp

delta G -300,574 0 0 150,287

2 Al

delta H -336,616 0 0 168,308

960°C

delta S 48.1 13.5 159.9 18,664

Inert

2 AlCl

Anode Process Theoretical Minimum Energy Requirements

Cell kWh/kg

Electrolytic Work Requirement (delta G) 6.48

3 Cl

25°C

Thermal Energy for Temperature Maintenance (delta H – delta G) 0.78

25°C

Thermal Energy for Al at 960°C 0.39

Theoretical Minimum 7.65

Thermal Energy for Cl

at 960°C

0.19

Note: Thermodynamic values for G, H and S are from Table J.6. Heat capacity data are from Table J.9 and Appendix K

124

Appendix J: Theoretical Energy Data and Calculations

Table J.5: Theoretical Minimum Energy for Kaolinite / Aluminum Chloride Reduction

Appendix J: Theoretical Energy Data and Calculations

Net Aluminum Reaction:

(products

–

reactants)

CarboChlorination (900°C) cal/gm mole Al

( 3 Al

.2SiO

+ 14 C + 21 Cl

) to ( 6 AlCl

+ 6 SiCl

+ 7 CO + 7 CO

) = Net

Reactants Products

Temp Temp delta G -2,374,053 0 0 -901,721 -884,799 -229,655 -659,819 -50,323

delta H -2,531,520 0 0 -1,009,847 -942,161 -184,910 -658,363 -43,960

7 Al

.2SiO

14 Al

delta S 19 1,119 144 474 331 357

25qC

960°C

Kaolinite Chloride Reduction (700°C)

( 2 AlCl

) to ( 2 Al + 3 Cl

) = Net

to delta G -300,574 0 0 150,287

14 SiO

delta H -336,616 0 0 168,308

Aluminum

25°C

delta S 48 14 160 18,664

Chloride

Process Theoretical Minimum Energy Requirement

7 CO

to kWh/kg

25°C

Thermal Work for AlCl

Production

-1.90

Reduction Electrolytic Work Requirement (delta G) 6.48

14 C

Thermal Energy for Temperature Maintenance (delta H – delta G) 0.78

25°C

System

7 CO

Thermal Energy for Al at 960°C 0.39

25°C

Theoretical Minimum 5.76

Thermal Energy for CO

at 960°C

0.11

Thermal Energy for CO at 960°C 0.25

Note: Thermodynamic values for G, H and S are from Table J.6. Heat capacity data are from Table J.9 and Appendix K

125

Table J.6: Thermochemistry Data for Elements and Compounds Associated witb Aluminum Production

Molecular Chemical

CAS RN Weight Formula J/mol cal/mol

H (s)

J/mol cal/mol

G (s)

J/mol K cal/mol K

S (s)

J/mol K cal/mol K

Cp (s)

Aluminum 7429-90-5 26.98 Al

Aluminum Chloride 7446-70-0 133.34

AlCl

Corundum 1334-28-1 101.96

Gibbsite* 155.96

.3H

Kaolinite* 1332-58-7

.SiO

.2H

Kaolinite, meta 162.04

.SiO

Graphite 7440-44-0 12.01 C

Chlorine 70.91

(g)

Carbon Monoxide 630-08-0 28.01 CO (g)

Carbon Dioxide 124-38-9 44.01

(g)

Oxygen 7782-44-7 32.00

(g)

Water 18.00

O (g)

Silica Dioxide 14808-60-7 60.08

SiO

Silicon tetrachloride 10026-04-7 169.90

SiCl

(g)

0 0

-704,200 -168,308

-1,675,700 -400,500

-1,293,100 -309,058

-4,119,000 -984,465

-843,840

0 0

-110,523 -26,416

-393,513 -94,052

0 0

-241,826 -57,798

-910,700 -217,663

-657,000 -157,027

0 0

-628,800 -150,287

-1,582,300 -378,179

-1,154,900 -276,028

-3,793,900 -906,764

-791,351

0 0

-137,268 -32,808

-394,383 -94,260

0 0

-228,582 -54,632

-220,615

-617,000 -147,467

28.3 6.764

100.7 24.061

50.9 12.165

5.7 1.361

222.9 53.286

197.9 47.301

213.6 51.061

205.0 49.003

188.8 45.132

41.46 9.909

330.70 79.039

24.35 5.82

91.84 21.95

79 18.88

33.598

90.3 21.6

Source: Handbook of Chemistry and Physics, David R Lide, Editor, 80th Edition CRC

*Source: Advances in Physical Geochemistry, Edited by Saxena, S.K. - data courtesy of Mr. Bruce Hemingway at U.S.G.S.

Table J.7: Changes in Heat of Formation Values as a Function of Temperature

Temperature

Heats of Formation in calories per gram mole

C * Al *

CO *

25 298

727 1,000

827 1,100

927 1,200

960 1,233

1,027 1,300

1,127 1,400

-400,300 0 0 -94,050 -26,400 0

-404,400 2,310 18,710 -94,400 -26,750 9,249

-404,000 3,320 21,710 -94,250 -26,900 10,515

-403,600 3,850 24,740 -94,300 -27,000 11,843

-403,500 4,030 25,750 -94,300 -27,100 12,295

-403,200 4,390 27,790 -94,300 -27,300 13,233

-402,800 4,930 30,850 -94,300 -27,350 14,685

Source: *Values from "Technical Working Group on Inert Anode Technologies" Appendix A.9, page 11

126

Appendix J: Theoretical Energy Data and Calculations

Table J.8: Theoretical Minimum Energy for Gibbsite Dehydration

Reactants

Temp

.3H

25°C

Kiln

Products

Temp Gibbsite Dehydration Reaction

(products

–

reactants)

( Al

.3H

0 ) to ( Al

+ 3 H

0 ) = Net cal/gm mole Al

25°C

delta H -618,117 -400,500 -173,393 44,223

Process Theoretical Minimum Energy Requirements

kWh/kg alumina

3 H

0.50

25°C

Note: Thermodynamic values for G, H and S are from Table J.6. Heat capacity data are from Table J.9 and Appendix K

Table J.9: Heat of Capacity Equations for Gases Associated with Aluminum Production

Standard Molar Heat Capacity, C = a + bT + cT

(T in degrees K)

cal/ mol K a

b x 10

c x 10

6.148 3.102 -9.23

6.420 1.665 -1.96

6.214 10.396 -35.45

7.576 2.424 -0.65

Source: A Different Approach to Thermodynamics, W. F. Ludar, 1967

127

Appendix J: Theoretical Energy Data and Calculations

Appendix K: Alumnum Heat Capacity and Heat of Fusion Data

Heat Capacity (cal/mol.K) = A + B.t + C.t

+ D.t

+ ---

(B.t)

(C.t

) (D.t

)

S tandard Enthalpy (kcal/mol) = A.t + ------------ + --------------- + --------------- –

---

+ F – H

2 3 4

(C.t

) (D.t

)

Standard Entropy (cal/mol.K) = A. ln ()t + B.t +

---------------

–

--------------

+ G

2 3

(2.t

)

where, t = K/1000 and A, B, C, D, E, F, G and H are constants

Source: http://webbook.nist.gov (Standard Reference Data Program)

Formula Constants for Aluminum

Solid (298 to 933.45 K, 1 atm):

A B C D

6.71348 -1.29418 2.04599 0.819161

-0.066294

-2.18623

14.7968

Liquid (933.45 to 2790.812 K, 1atm):

A B C D

7.588681 9.40685E-09

4.26987E-09 6.43922E-09

1.30976E-09

-0.226024

17.5429

2.524381

128

278 5.67

328 5.92

378 6.10

428 6.24

478 6.36

528 6.48

578 6.61

628 6.74

678 6.89

728 7.05

778 7.22

828 7.41

878 7.62

928 7.85

Appendix K: Alumnum Heat Capacity and Heat of Fusion Data

Table K.1:Aluminum Energy Requirements for Heating and Melting

Energy Cumulative Energy Cumulative

Temperature

Heat Capacity at

for Step Energy to Raise for Step Energy to Raise

Temperature

Change from 25

C Change from 25

(

C) (K)

(cal/mol K)

(kWh/kg) (kWh/kg) (Btu/lb) (Btu/lb)

25 298 Solid 5.79 0.00 0.00 - -

660 933 7.88 0.19 0.19 286 286

660 Fusion 94.5 (cal/gm) 0.11 170

660 933 Liquid 7.59 0.29 456

775 1048 7.59 0.04 0.33 58 515

960 1233 7.59 0.06 0.39 94 608

2000 2273 7.59 0.34 0.73 526 1,134

Smelting

C to 960

Total 0.39 608

Furnace Melting

C to 775

Total 0.33 515

Note: Heat capacity for solid aluminum varies significantly with temperature, whereas for molten aluminum, it is nearly

constant.

Heat Capacity as a Function of Temperature

4.00

4.50

5.00

5.50

6.00

6.50

7.00

7.50

8.00

8.50

9.00

0 500 1000 1500

Temperature,

cal/mol K

Solid Liquid

129

Appendix L: Impact of Secondary Metal Production on Energy

Requirements for U.S. Aluminum Production

The nominal U.S. energy required to produce aluminum metal has been rapidly declining as secondary aluminum

production has grown. Secondary aluminum requires only 6% of the energy necessary to manufacture primary aluminum

(Appendix F, Table F.6). The total U.S. ore-to-metal primary energy values include: refining of half the alumina supply,

anode manufacture, electrolysis and ingot casting. Combining the energy requirements for U.S. production of primary and

secondary metals lowers the average energy associated with U.S. aluminum metal from over 40 kWh/kg for primary alone

to about 21 kWh/kg for the combined metals.

Table L.1:Impact of Secondary Metal Production on the Nominal Energy Requirements to Produce Aluminum

Year

U.S. Primary U.S. Secondary

Aluminum Smelting Aluminum

Production Energy Production

(thousand (kWh per (thousand

metric tonnes) kilogram) metric tonnes)

Market Percent

of Secondary

Metal

Total Energy

Ore-to-Metal to Produce

Combined Metals

(kWh/kg)

Effective Smelting

Energy to Produce

Combined Metals

(kWh/kg)

1960 1,828 23.1 401 18% 69.7 19.2

1961 1,727 22.9 445 67.2 18.5

1962 1,921 22.7 533 65.7 18.1

1963 2,098 22.6 601 64.7 17.9

1964 2,316 22.4 648 64.6 17.8

1965 2,499 22.3 774 62.7 17.3

1966 2,693 22.1 833 62.2 17.2

1967 2,966 21.9 821 63.3 17.5

1968 2,953 21.8 935 61.0 16.8

1969 3,441 21.6 1,067 60.8 16.8

1970 3,607 21.4 937 21% 62.7 17.3

1971 3,561 21.0 1,004 60.5 16.7

1972 3,740 20.6 1,022 59.8 16.5

1973 4,109 20.2 1,127 58.6 16.1

1974 4,448 19.9 1,163 58.0 16.0

1975 3,519 19.5 1,121 54.5 15.0

1976 3,857 19.1 1,334 52.3 14.5

1977 4,117 18.7 1,456 50.9 14.1

1978 4,358 18.3 1,518 50.1 13.8

1979 4,557 17.9 1,612 48.8 13.5

1980 4,653 17.5 1,577 25% 48.2 13.3

1981 4,489 17.4 1,790 45.8 12.7

1982 3,274 17.2 1,666 42.2 11.8

1983 3,353 17.1 1,773 41.3 11.5

1984 4,099 16.9 1,760 43.8 12.1

1985 3,500 16.8 1,762 41.3 11.5

1986 3,039 16.6 1,773 38.9 10.9

1987 3,347 16.5 1,986 38.4 10.7

1988 3,945 16.3 2,122 39.4 11.0

1989 4,030 16.2 2,054 39.7 11.1

1990 4,048 16.1 2,393 37%

37.4 10.5

1991 4,121 16.0 2,286 38.0 10.6

1992 4,042 15.9 2,756 35.0 9.8

1993 3,695 15.8 2,944 32.7 9.2

1994 3,299 15.7 3,086 30.2 8.6

1995 3,375 15.6 3,188 29.9 8.5

1996 3,577 15.5 3,307 30.0 8.5

1997 3,603 15.4 3,547 28.9 8.2

1998 3,713 15.3 3,442 29.6 8.4

1999 3,779 15.2 3,695 28.7 8.1

2000 3,668 15.1 3,450 48% 29.0 8.2

2001 2,637 15.0 2,970 26.4 7.5

2002 2,705 14.9 2,927 26.7 7.6

2003 2,704 14.8 2,820 51% 27.1 7.7

Forty-Three Year Total Change in U.S. Energy Intensity:

1960 69.7 (kWh/kg)

2003 27.1 (kWh/kg) 61.2%

130

Appendix M: Impact of Using Different Technologies on Energy

Requirements for Producing Aluminum

Wetted cathodes allow the anode-cathode-distance to be reduced. This results in a lowering of the voltage requirement for

amperage to pass through the cryolite bath. The following table shows the effect lowering the anode-cathode-distance has

on energy consumption.

Table M.1: Impact of Wetted Cathode Technology on Primary Metal Electrolysis

WETTED

CATHODE

TECHNOLOGY

Typical modern

"prebaked" Hall-

Héroult cell

operating at 95%

current efficiency

ACD = 4.5 cm

Typical modern

"prebaked" Hall-

Héroult cell operating

at 95% current

efficiency and

retrofitted with a

wetted cathode

surface and reduced

ACD

ACD = 3.5 cm

Typical modern

"prebaked" Hall-

Héroult cell operating

at 95% current

efficiency and

retrofitted with a

wetted cathode

surface, aluminum

sump and reduced

ACD

ACD = 2.5 cm

Typical modern

"prebaked" Hall-

Héroult cell operating

at 95% current

efficiency and

retrofitted with a

sloped and wetted

cathode surface,

aluminum sump, and

r e d u c e d A C D

ACD = 2.0 cm

Energy

Requirements

V (dc) kWh/kg Al V (dc) kWh/kg Al V (dc) kWh/kg Al V (dc) kWh/kg Al

Reaction 1.20 3.76 1.20 3.76 1.20 3.76 1.20 3.76

Additional Energy

Requirements:

External 0.15 0.47 0.15 0.47 0.15 0.47 0.15 0.47

Anode 0.30 0.94 0.30 0.94 0.30 0.94 0.30 0.94

Anode

Polarization

0.55 1.73 0.55 1.73 0.55 1.73 0.55 1.73

Cathode

Polarization

0.05 0.16 0.05 0.16 0.05 0.16 0.05 0.16

Cryolite Bath 1.75 5.49 1.36 4.27 0.97 3.05 0.78 2.44

Cathode 0.45 1.41 0.45 1.41 0.45 1.41 0.45 1.41

Other 0.15 0.47 0.15 0.47 0.15 0.47 0.15 0.47

ONSITE Energy Values

Cell Total

% Energy Savings

4.60 14.43 4.21 13.21

3.82 11.99

17%

3.63 11.38

21%

Anode Manufacturing

Total onsite cell and

anode

% Energy Savings

0.61

15.04

0.61

13.82

0.61

12.60

16%

0.61

11.99

20%

TACIT Energy Values

Cell Total

% Energy Savings

4.60 32.23 4.21 29.50

3.82 26.78

17%

3.63 25.41

21%

Anode Manufacturing

Total onsite cell and

anode

% Energy Savings

6.02

38.25

6.02

35.52

6.02

32.80

14%

6.02

31.44

18%

131

Appendix M: Impact of Using Different Technologies on Energy Requirements for Producing Aluminum

Table M.2: Impact of Inert Anode and Wetted Cathode Technology on Primary Metal Electrolysis

A B C D

INERT ANODE and

WETTED

CATHODE

TECHNOLOGY

Typical modern Hall-

Héroult cell

operating at 95%

current efficiency

and ACD of 4.5 cm

Inert Anode direct

substitution operating

at 95% current

efficiency

Inert Anode operating

at 95% current

efficiency with

polarization

differences reported

for oxygen generation

Inert Anode operating

at 95% current

efficiency, with

oxygen polarization

differences and an

ACD of 2cm using

Wetted Cathode

Energy

Requirements

V (dc) kWh/kg Al V (dc) kWh/kg Al V (dc) kWh/kg Al V (dc) kWh/kg Al

Reaction 1.20 3.76 2.20 6.90 2.20 6.90 2.20 6.90

Additional Energy

Requirements:

External 0.15 0.47 0.15 0.47 0.15 0.47 0.15 0.47

Anode 0.30 0.94 0.30 0.94 0.30 0.94 0.30 0.94

Anode

Polarization

0.55 1.73 0.55 1.73 0.10 0.31 0.10 0.31

Cathode

Polarization

0.05 0.16 0.05 0.16 0.05 0.16 0.05 0.16

Cryolite Bath 1.75 5.49 1.75 5.49 1.75 5.49 0.78 2.44

Cathode 0.45 1.41 0.45 1.41 0.45 1.41 0.45 1.41

Other 0.15 0.47 0.15 0.47 0.15 0.47 0.15 0.47

ONSITE Energy Values

Cell Total

% Energy Savings

4.60 14.43 5.60 17.57

-22%

5.15 16.15

-12%

4.18 13.11

Anode Manufacturing

Total onsite cell and

anode

% Energy Savings

0.61

15.04

0.77

18.33

-22%

0.77

16.92

-13%

0.77

13.87

TACIT Energy Values

Cell Total

% Energy Savings

4.60 32.23 5.60 39.23

-22%

5.15 36.08

-12%

4.18 29.27

Anode Manufacturing

Total onsite cell and

anode

% Energy Savings

6.02

38.25

0.77

40.00

-5%

0.77

36.84

0.77

30.03

21%

Inert Anodes by themselves require additional reaction voltage as can be seen by comparing columns A and B. However,

this additional energy requirement is offset by the elimination of carbon anode manufacturing, along with the elimination

of the feedstock energy associated with carbon. The estimate for the energy associated with the manufacture of inert

anode is shown in Table M.3. Column C shows the impact that results from the lower anode polarization and the ability to

design to release gas more effectively. Wetted cathodes combined with inert anodes can provide additional savings as

shown in column D.

132

Appendix M: Impact of Using Different Technologies on Energy Requirements for Producing Aluminum

Table M.3: Estimate of Energy Requirement for Manufacturing an Inert Anode

The energy requirements for manufacturing an inert anode are significantly less than the total manufacturing energy of

the many consumable carbon anodes that it replaces. Below is an estimate of the energy required to manufacture an

inert anode. Assuming an inert anode has a cell life of two years, the equivalent carbon energy requirements can be

calculated. Using 1 cm

as a basis, the following can be calculated:

0.85 A/cm

is the typical current density for a modern Hall-Héroult Cell current density.

2980 Ah/kg Al is the theoretical amperage required to produce aluminum.

95% is the typical efficiency of a modern Hall-Héroult carbon anode cell.

0.446 kilograms of carbon are required to produce one kilogram of aluminum.

From the above data, the amount of carbon consumed and aluminum produced per cm

over a two year period is

calculated to be:

14.82 kg

of carbon are consumed per cm

over a two year operating period

33.23 kg

of aluminum are produced per cm

over a two year operating period.

From Appendix I:

6.02 kWh/kg of Al is the tacit energy associated with carbon anode. The total energy associated with the two-year

produced

operation of the 1 cm

of carbon anode can be calculated.

200 kWh of tacit energy are consumed for anode manufacture and use.

2.27 gm/cm

is the density of a carbon anode. The height of the 1 cm

of carbon can be calculated as below.

6529 cm

is the height of carbon anode with a 1 cm

base or anode face for two years of operation

The materials under consideration for inert anodes have no inherent fuel value as does the carbon anode. The tacit

energy requirement associated with the manufacturing of an inert anode is related to the extraction of materials and the

inert anode manufacturing process. Assuming that the inert anode is 5 cm thick per cm

of anode face and that it

requires 5 times the total tacit energy of a carbon anode (which includes its fuel value) to manufacture, it can be

calculated that:

0.77 tacit kWh/kg of aluminum will be required to produce an inert anode.

133

Appendix M: Impact of Using Different Technologies on Energy Requirements for Producing Aluminum

Table M.4: Estimate of Energy Requirement to Manufacture Aluminum Using Different Technologies

ALUMINUM

PRODUCTION

TECHNOLOGIES

Typical modern Hall-

Heroult prebaked anode

cell operating at 95%

current efficiency & with

ACD of 4.5 cm

Typical modern Hall-

Heroult cell operating at

95% current efficiency,

retrofitted with a sloped &

wetted cathode surface,

aluminum sump, & a

reduced ACD of 2.0 cm

Inert Anode operating at

95% current efficiency

with oxygen polarization

differences & ACD of

2.0 cm using Wetted

Cathode technology

Carbothermic Reduction

with a reaction efficiency

of 95% and a furnace

efficiency of 80%

Chloride Reduction of

Kaolinite Clays: electrolysis

current efficiency 95%,

reaction efficiency 95% &

heating efficiency 85%

Raw Materials

Onsite Energy

Required

Tacit Energy Required

kWh/kg Al

alumina 7.59

alumina 8.24

kWh/kg Al

alumina 7.59

alumina 8.24

kWh/kg Al

alumina 7.59

alumina 8.24

kWh/kg Al

alumina 7.59

alumina 8.24

kWh/kg Al

see Note B

kaolinite 8.14

kaolinite 8.84

Onsite Reaction Energy Requirements

Reaction Thermal

Furnace Losses

Reaction Elerctrolysis

Cell Ohmic

TOTAL Reaction

Energy

Anode Manufacturing

Onsite Energy

Savings

V (dc)

1.20 3.76

3.40 10.67

4.60 14.43

0.61

V (dc)

1.20 3.76

2.43 7.62

3.63 11.38

0.61

20%

V (dc)

2.20 6.90

1.98 6.20

4.18 13.11

0.77

7.71

1.93

9.63

36%

-1.90

see Note C 0.40

V (dc)

2.07 6.48

0.93 2.93

3.00 7.91

0.77

42%

Total Onsite Energy Requirements Including Raw Materials

Raw Ore Materials

Reactions

Carbon / Electrodes

Total Energy Savings

7.59

14.43

0.61

7.59

11.38

0.61

13%

7.59

13.11

0.77

7.59

9.63

see note A 0

24%

8.14

7.91

see note D 0.77

26%

Total Tacit Energy Requirements Including Raw Materials

Raw Ore Materials

Reactions

Carbon / Electrodes

Total Energy Savings

8.24

32.23

6.02

8.24

25.41

6.02

15%

8.24

29.27

0.77

18%

8.24

21.52

see note A 8.43

18%

8.84

18.60

see note D 12.13

15%

134

Appendix M: Impact of Using Different Technologies on Energy Requirements for Producing Aluminum

Note A: The Carbothermic reaction requires twice the carbon as the Hall-Héroult reaction. The Hall-Héroult

reaction on a theoretical basis requires 0.33 kg C / kg Al. In this estimate, it is assumed that the

carbothermic reaction utilized carbon at a 95% efficiency. Unlike a carbon anode, the carbon used does

not require the energy associated with carbon anode manufacturing. Hence only the fuel or "secondary

energy" requirement of the carbon are used.

Note B: Kaolinite contains 27.2% aluminum by weight. Bauxite contains 45% aluminum by weight. Both

materials contain about the same percentage of impurities. For this estimate, it is assumed that the

processing and calcination energy of kaolinite is the same as bauxite. However, 66% more kaolinite must

be processed to produce a kilogram of aluminum than bauxite. Kaolinic clays contain valuable titanium,

in addition to their silicon content. These materials will likely be recovered in the processing plant and

account for approximately 35% of kaolinic clay content. This report allocates 65% of the total raw

material energy requirement to the kaolinite material used for aluminum manufacturing.

Note C: The Carbo-Chlorination reaction is exothermic (-1.90 kWh/kg Al). However, Toth Aluminum reports that

based on pilot plant experience and the challenges of capturing the off-gas energy, a small quantity of

energy (0.4 kWh/Kg Al) is required to maintain the reaction system temperature.

Note D: The Carbo-Chlorination reaction requires 0.89 kg of carbon per kg of aluminum produced. In this

estimate, it is assumed that the Carbo-Chlorination reaction utilized carbon at a 95% efficiency. It is also

assumed that the energy required and fuel value of the carbon is the same on a weight basis as the carbon

utilized for Carbothermic Reduction of Aluminum. (see note A)

135