Aluminiumworldjournal2014 141022112624 Conversion Gate02 PDF

Global Media Communication Ltd.
2014 Edition
Charles Martin Hall
Maximize your return on investment?

Absolutely.
ABBs history of powering primary aluminium plants started 45 years ago. Ever since,
we have been supplying complete electrification solutions and substations to more than
60 aluminium smelters worldwide. Demands for improved environmental performance
and increased energy efficiency, price fluctuations and intense competition are the
major challenges aluminium producers face today. ABB meets these challenges by
providing state-of-the-art electrification, automation and process optimization solutions
always with the objective to increase your productivity and maximize your return on
investment. For more information, visit us at www.abb.com/aluminium
Global Competence Center Aluminium

5405 Baden 5 Dttwil, Switzerland
aluminium@ch.abb.com
Aluminium World Journal 2014

Global Media Communication Ltd.
Managing Director
Christopher Fitcher-Harris,
Production Manager
Sofia Henriksson
Sales Manager
Peter Jones
Production Design:
row1graphics
Published by: Global Media
Communication Limited
Telephone: +44 208 579 0594
E-mail: gmcproduction@gmx.com
Website:
globalmediacommunication.com
The opinions and views expressed in
the editorial of content in this book are
those of the authors alone and do not
necessarily represent the views of any
organisation with which they may be
associated. Material in advertisements
and promotional features may be
considered to represent the views of
the advertisers and promoters. The
views and opinions expressed in this
book do not necessarily express the
views of the publisher. While every
care has been taken in the preparation
of the book, the publishers are not
responsible for such opinions and views
or for any inaccuracies in the articles
or advertisements.
2014 The entire contents of this
publication are protected by copyright.
Full details are available from the
publishers. All rights reserved. No
part of this publication may be
reproduced, stored in a retrieval
system or transmitted in any form or
by any means, electronic, mechanical,
photocopying, recording or otherwise
without the prior knowledge of the
copyright owner.
Cover illustration:
Alcoa
Foreword By Christopher Fitcher-Harris

Aluminium World Journal 2014 features editorials, case studies, company
profiles, and product reviews.
The publication is divided by industry sector sections to ensure ease of
navigation.
This edition contains special feature articles produced by TMEIC entitled
TMEIC Serving the Aluminium Industry, and Rio Tinto Alcan on the Start-up
of the Arvida Smelter, AP60 Technology Center. We are pleased to present
new independent authors for this edition: Dr. Ing. Joachim Heil from MetCons
with the paper Aluminium Reduction Cell Technology Providers a 2014
Review and Louis Dekker, Process Engineering Specialist from LeProCon,
with the concept article on An Intermediate Step in Cost Reduction for Inert
Anodes and would like to thank them for their contributions.
I take this opportunity to thank all the participating companies for providing
Aluminium World Journal 2014 with editorial, company profiles, advertisements
and corporate sponsorship.
Aluminium World Journal 2014 is available for you to read online, download,
and in print format. Visit us online at:
www.globalmediacommunication.com
If you should wish to discuss with me anything concerning the content of
this edition, do not hesitate to contact me.
Hope you enjoy the read!
Christopher Fitcher-Harris
Managing Director
Charles Martin Hall had a purpose to his life.

And it wasnt a small one, either.
Mr. Hall revealed that probably his

chief ambition in life was to make some
discovery which would be revolutionary
with regard to the present conception of the constitution of matter and
which would be of immense benefit
to mankind, wrote Arthur Vining Davis, former president and chairman of
the Aluminum Company of America
(Alcoa), which Hall helped found in
1888.
For Hall (1863-1914), the ticket to making his dream into a reality was his
love for science and interest in aluminum.
From the time he was a teenager, Hall
noted that although aluminum was
the Earths most abundant metal, the
process for extracting it from its ore
in a laboratory was so difficult it was
only made in small quantities. Supply
and demand made aluminum as expensive as silver. Hall vowed to find a
better way. During his years at Oberlin
College in Ohio, he tried and failed
repeatedly. Still, he stayed positive and
worked to discover an easier method
of extraction.
Day and night, consciously and subconsciously, he was still working on
the problem of producing cheap aluminum, wrote Julius Edwards in The
Immortal Woodshed: The Story of the
Inventor Who Brought Aluminum to
America. Hall was at heart . . . a tireless experimenter.
He approached science deliberately
and logically. He formed theories based
The first small, shining globules of aluminum reduced through the Hall Process. They
are referred to as Alcoas crown jewels. Shown here on a page of handwritten minutes
from a company meeting, circa 1890.
on his experiments, then asked others

to confirm his findings.
After graduating in 1885, Hall returned
to his familys home to continue his
experiments. He went over his records
to re-evaluate the problem, and then
embarked on a new strategy. He realized hed need more work space and
new equipment, so he moved his lab
out of the house and into the woodshed.
While his fellow graduates jumped
into the business world, Hall focused
on making his discovery so he could
make his mark in that world. He locked
himself in the woodshed, combining

countless substances in his quest. He
carefully logged each attempt and its
outcome. When he found a promising combination, he tried numerous
variations until he was sure it wouldnt
work.
Then, in February 1886, Hall made
his breakthrough: electrolyzing alumina dissolved in molten cryolite. Hed
discovered an inexpensive method
for isolating pure aluminum from its
compounds.
He wasnt alone, however: The potential
rewards for a cheaper aluminum isola-
Alcoas lightweight aluminum helped

revolutionize the automotive and aviation industries; aluminum foil eased
the lives of housewives everywhere.
Demand for Halls aluminum led to
production soaring from 10,000 pounds
in the companys first year to 15 million
by 1907. One plant grew to three.
In 1911, Hall was internationally recognized with the Perkin Medal for his
contributions to chemistry.
1886-1920. The Hall family home in Oberlin, Ohio. Hall discovered the aluminum process
in a summer kitchen attached to the back of the home.
tion process had scientists the world

over racing to find a workable method.
French chemist Paul L.T. Hroult was
one of them, and he developed the
same method at about the same time
as Hall. The process became known
as the Hall-Hroult process.
Quick Action
Aware of the other efforts, Hall moved
immediately to protect his method. He
wrote immediately to the U.S. Patent
Office, submitting his process.
Patent number 400,655, granted to
Hall in 1889, changed the aluminum
industry forever.
To make his efforts profitable, Hall

knew he had to make the process available for widespread use. So he worked
as relentlessly in finding backers and
raising capital as he did in the lab.
He made a list of industries that might
use aluminum. He prepared drawings
and charts to show how the process
could be applied. Then he made appointments with various wealthy individuals to show how theyd benefit if
they invested in his idea.
His presentation persuaded some investors to join him, and the Pittsburgh
Reduction Company was born. The firm
was re-named the Aluminum Company
of America (Alcoa) in 1907.
Halls process is a new discovery. It

is a decided step forward in the art of
making aluminum. Since it has been
put into practical use, the price of aluminum has been reduced from six
or eight dollars a pound to 65 cents.
This is a revolution in the art and has
had the effect of extending the uses
of aluminum in many directions not
possible when its price was high . . .
Hall was a pioneer and is entitled to the
advantages which that fact gives him
in the patent law, said Judge William
Howard Taft, later U.S. president, in a
1893 ruling in Halls favor regarding
a patent case.
By 1914, the cost of aluminum was
down to 18 cents a pound.
Halls parents gave him a solid educational foundation. His mother taught
him to read before he was 5. Books
were plentiful in the Hall household,
and young Charles pored through every one he could get his hands on. He
even delved into his fathers college
chemistry books: the heavy tomes
introduced him to, and sparked his
love of, science.
I have often seen him, after he had
read for a while, lying asleep with his
face on the book. . . . Someone would
pick him up, still sleeping, and put
him and his beloved book in a safe
place, Halls sister Julia recalled years
later.
Halls love of reading and education
stayed with him his entire life.
Drawing of the interior of the Smallman Street works of the Pittsburgh Reduction Company
depicting the reducing pots used in the companys process. (1888)
He used to read the Encyclopedia

Britannica night after night, year after
year, literally . . . He used to . . . open it
wherever it happened to open; then he
would spend the evening reading, and
he accumulated a big fund of information in that manner, Davis said.
Learning From The Best

Figuring he could learn from those
whod gone before him, Hall studied the lives of successful people,
especially inventors such as George
Westinghouse. From the Scientific
American, he learned about patent
law and practices, and keeping ideas
secret until theyre ready.
The New Kensington office building of Pittsburgh Reduction Company. (1891)
Even as his success and net worth increased, Halls work ethic remained
solid. He was not just satisfied with
having someone else promote his
process, Edwards wrote. Although a
director and vice president of his company, he worked long hours at the plant,
determined that the success of his process and (of the) company should far
exceed any of his original prophecies.
Science wasnt Halls only interest,
however. He had a lifelong love and
appreciation of nature, and music had
been a passion for him since childhood. Playing the piano was a source
of relaxation his entire life, and helped
him clarify scientific problems, Edwards wrote.
He also fed his soul. He attended
church regularly, and drew strength
from the stories of great men who
sacrificed for their convictions. The
creed which found most significant
expression in his works and deeds emphasized the importance and value
of good character, said his brother,
George Hall.
While Hall helped to change industry
and make many goods available to the
masses that would otherwise have
been unaffordable, he never forgot
what helped make him a success. Upon
his death, Hall bequeathed Oberlin
College more than $5 million.
First ingot being charged into remelting furnace at Alcoa Tennessee Plant. (1920)
Authors: Investors Business Daily

Photo Credit: Alcoa
INDEX
Special Feature
p. 9-12
Global Issues
p. 13-17
Primary Smelting and Processes
p. 19-59
Anode Plant Technology
p. 61-87
Materials Handling And Transportation
p. 89-95
Company Profiles
p. 97-103
Advertiser and Web Index
p. 104
TMEIC Serving the Aluminum Industry
15 ALUMINUM MILLS AUTOMATED

IN THE PAST 10 YEARS.
TMEIC delivers. In fact, weve been a leading force in the metals
industry for more than 50 years, and have been the preferred
partner for most of the recent aluminum mills in the world.
TMEICs Advanced Process Automation Control System features
fast and effective level 1 controls and integrated level 2 models
for aluminum mills.
tmeic.com/cranes
1-540-283-2250
TMEIC Corporation
1325 Electric Road
,>i]6{n
1-540-283-2000
SPECIAL FEATURE
U >V
Gage Control
U VViV
Compensation
U >i
U V>V}
Email: cranes@tmeic.com
TMEIC Japan
/->i>`i]L>
]/]>>{
+81-(0)3-3277-5511
tmeic.com | info@tmeic.com
SPECIAL FEATURE
TMEIC
Serving the Aluminium Industry
p. 10-12
AWJ 2014

Introduction
The Aluminum industry has been facing
continuing market challenges for the
last 50 years and the future looks to
be just as demanding and challenging
as the past. Currently the aluminum
market depends on the transportation
industry, the construction industry,
industrial applications and the UBC
market for utilization of aluminum flat
rolled products. With the UBC market
declining in traditional uses as well as
the construction industry rebounding
very slowly after the 2008 market
collapse in the developed portion of
the world, there is hope for continued
growth in the transportation sector and
in the emerging markets.
Traditionally, the commercial aircraft
segment of the transportation market
has grown with the ever increasing
number of wide-bodied aircraft being
bought worldwide. While the use of
composite material has replaced many
kilograms of aluminum utilized in the
two newest planes from Boeing and
Airbus, the rise in the total number
of single-aisle planes expected to be
ordered through to 2020, will keep
the total amount of aluminum being
delivered into this market segment
growing at a single digit rate.
The real opportunity for aluminum
growth is in the automotive segment
of the transportation market. As
automotive manufacturers are being
pressed to deliver higher fuel mileage
many strategies are being evaluated,
with weight reduction being primary.
Replacing low carbon sheet steel with
an alternative material that is lighter and
competitively priced, but still retains
the high strength required for structural
integrity, is in high demand.
Aluminum mills serving these markets
are challenging the traditional material
suppliers for market share. Buyers
are seeking tighter gauge tolerances,
tighter temperature control, more
product classifications, better shape and
flatness performance, better surface
10 SPECIAL FEATURE
quality and most of all, complete coil

documentation to be delivered to the
customer along with the coil. While
raw material costs or scrap prices are
controlled by upstream operations
or outside forces, the mill must
understand and control operational
costs such as energy usage, labor,
maintenance and upgrade costs, and
scrap losses. In these areas, mills in
Europe and North America may be at
a disadvantage against those more
recently built in the Pacific Rim. Most
flat mills in Europe and North America
have been in operation for at least 30
years while those in the Pacific Rim,
outside of Japan, have been built
within the last 10 years, with several
in the planning or construction phases.
This gives the operational advantage
to the newer mills with the latest in
technological improvements in mill
design, level 1 control and higher levels
of automation, while older mills have
the advantage of better operational
practices and an established customer
base. The latter is open for invasion
by new suppliers providing better
pricing, better customer service or
better quality, if available.
An existing mill must develop and
depend on its suppliers as a partner to
enable new ideas to be incorporated,
to help develop a strategy to upgrade
performance and to keep the mill from
becoming obsolete. These suppliers
can be a source of ideas on how to
reduce downtime, reduce scrap, reduce
energy consumption, or at least recover
lost energy, and possibly to increase
throughput beyond design capacity.
TMEIC the Company

TMEIC was formed in 2003 through
a powerful alignment of global leaders, Toshiba, Mitsubishi-Electric and
GE. TMEIC has earned a reputation by
supporting the legacy control systems
of its parents and providing reliable,
state-of-the-art industrial products
and system solutions for new mills.
Advanced technology, excellence in

engineering and years of accumulated
experience are brought to each system
to provide the customer with a solution
to match the project needs. TMEIC
serves a variety of industrial markets
including Metals, Material Handling,
Oil and Gas, Mining, and Cement, as
well as utility scale Solar Power.
In Metals, TMEIC applies its capabilities
built on 60 years of rolling mill
experience supplying comprehensive,
high-performance control solutions.
TMEIC is recognized as the leading
global supplier of level 2 and process
model automation. Our range of control
and automation includes the ability to
supply complete systems using Motors,
Drives, level 1 control consisting of
Programmable controllers, I/O, and
HMIs, Level 2 and networks, process
models and instrumentation. Projects
range from small upgrades to resolve
obsolescence issues, to complete major
upgrades of mill capabilities to meet
the current market needs. One recent
development, TMEICs uTool, provides
the ability to upload mill performance
data, such as production, coil data,
energy usage, or mill delays, through
the user companys intranet to any
mobile device or computer. This allows
maintenance, support personnel or
mill management to react and analyze
issues from anywhere accessible by
internet. Improved response that
shortens delays or minimizes scrap
losses translates directly to increased
productivity and to the customers
bottom line.
Recent Aluminum Projects

Of the 10 hot aluminum mills built in
China in the last 10 years, six chose
TMEIC as the system control and
automation supplier. These mills
include 1+1, 1+3, 1+4 and 1+5 mill
configurations. Including all of the
Pacific Rim, there are 2 additional
new mills that chose TMEIC. The
pictures below show the first coil put
through the mill. Success is measured

in meeting and exceeding customer
expectations.
TMEIC has also focused on revamp
projects. Control system revamps
require very close cooperation

between the customer, TMEIC, and
the mechanical supplier, if mechanical
modifications are necessary. TMEIC has
worked with more than 15 aluminum
mills worldwide in delivering upgrade
solutions. Detailed discussions are

required to clearly define the work
scope, the customers goals during
multiple shutdown periods, the list
of pre-shutdown tasks, and a detailed
schedule for the entire shutdown
period. This schedule must be reviewed
and agreed to by all stakeholders
involved, including management,
production, maintenance, major
vendors and engineering personnel.
Active participation from all parties
is required to allow for joint success
after the start-up.
Mill Control System
Aluminum Strip
First Coil from Mill
TMEICs AC main drive motors are

designed and built to meet or exceed
industry standards, and are known
for exceptional quality. Driven by
our customers continuous need for
sustained reliability and reduced lifecycle costs, TMEIC employs cuttingedge technology in design supported
by state-of-the-art manufacturing
capability to offer the worlds most
advanced motors. With more than
100 years of motor experience, TMEIC
consistently tackles tough applications
around the globe with designs
delivering quality, performance, and
efficient operation. TMEIC is among
the few large motor suppliers with the
capability to provide both Induction and
Synchronous motors for rolling mills, in
the range of 1,500 kW through 30,000
kW depending on the application.
With over 30 years of variable speed
drives experience, TMEIC has the
broadest offering of high performance
coordinated system drives, ranging
from low voltage drives to powerful
3,300 volt drives for large mill stands.
The TMdrive-70 medium voltage drive
has become the industry leading
drive with a reliability MTBF of over
30 years, utilizing the IEGT (Injection
Enhanced insulated Gate bipolar
Transistor). This drive can provide up
to 36,000 kVA power in its four-bank
configuration. Over 1,200 of these
AWJ 2014
11
TMEIC Main Rolling Mill Motor
TMdrive-70e2 Variable Frequency Drive
Customer Service
water cooled 3,300 volt drives have
been supplied worldwide to rolling
mill applications.
Process models are critical in the
aluminum industry to provide the
demanding product specifications.
TMEIC has worked with aluminum
companies to provide complete
control automation including level 2
and models or systems that allow for
the customers proprietary models.
TMEIC has a global network of offices

and engineers to support customers
around the world. This support includes
spare parts for drives and control
systems with immediate delivery
in Europe, India, the Americas and
Asia. Training classes are available for
projects as well as on-going training for
mill personnel. In addition to normal
maintenance support, the focus of our
training is to allow the customer to
analyze and determine any production
issue or adjust control systems for
Our modern control systems include:

Pass schedule calculation
Inner-stand tension control
Roughing mill setup
Inner-stand cooling
Finishing mill setup
Work roll coolant control
Finish temperature control
1 Gbps Ethernet communications
Coiler temperature control
Hot backup level 2 strategy
Roll thermal wear
On-line and Off-line model operation modes
Strip crown and flatness control
Remote diagnostics
Automatic gauge control
Graphical interface that allows operators to

visualize operation and performance
12 SPECIAL FEATURE
new products. TMEICs technical

advisory service provides a backup
for the customers personnel through
our 24-hour phone support, or onsite support as requested by the mill.
Long term partnering with TMEIC allows
aluminum companies to access our
engineering expertise to plan for future
capital modernizations as well as make
comparisons of existing operations
against design capabilities. This service
has been used by some customers
to plan upgrades that extend market
reach with new products.
Authors:
Paul Weary, Metals Sales
Manager, TMEIC
Phone: (+1) 540-283-2110
Jim Trexel, US Metals Sales
Manager, TMEIC
Phone: (+1) 540-283-2193
GLOBAL ISSUES
UC Rusal
New Horizons
p. 14-17
AWJ 2014 13
364"- New Horizons
Boguchansk HPP, 50% owned by RUSAL
China, the worlds largest aluminium

market, is showing a serious commitment to improve efficiency in
the countrys aluminium industry.
These changes could play a pivotal
role in the global aluminium market
development, and unlock potential
for a tighter cooperation between
China and Russia in the winged
metal production.
The big and the growing
China is the worlds fastest growing
economy. According to analysts estimates, China is on track to surpass
the US and become the largest world
economy by late 2020s. Over 46%
of Chinas soaring GDP comes from
the countrys rapid industrial growth
driven by the massive urbanization
which is increasing demand for aluminium and the raw materials used
in its production.
The winged metals consumption in
the country is supported by increasing car production and infrastructure
investments. During 2013, the Chinese
automotive industry was the top gainer,
14 GLOBAL ISSUES
surging 14.9% after record sales of

21.98 million vehicles according to
the China Association of Automobile
Manufacturing. The National Bureau
of Statistics data also showed that new
construction projects rose by 13.5% in
2013. China is forecast to post robust
growth in its auto market in the coming years, whereas the construction
sector is strongly expected to expand
further following the governments
latest urbanization initiatives.
According to the recently published
blueprint, authorities intend to raise
the proportion of urban residents to
60-65% of the total population by
2020, from the current 53.7%. By
2030, Chinas cities will have added
350 million more people and five million buildings will be built. The new
growth agenda will need the expansion of railways, roads, highways, and
airlines to facilitate labour flows.
Urbanization along with the urban income growth will drive Chinas transport
and construction sectors which jointly
account for over 50% of the countrys
total aluminium consumption, thus

propelling demand for aluminium. As
of today, the country accounts for 45%
of global aluminium consumption, but
is forecast to boost this share to 56% by
2025, extending its lead as the worlds
biggest aluminium consumer.
Focus on efficiency
In 2013, China produced over 25 million
tonnes of primary aluminium, almost
half of the global output. However,
further development of the Chinese
aluminium industry is subject to certain limitations in terms of power consumption and emissions by operating
smelters.
Efficient resources utilization is one of
the urgent issues now in China where
over 90% of primary aluminium smelters source energy from coal-fired power
plants that account for 75% of all CO2
emissions in aluminium production.
The government is also encouraging
reduction in consumption of power,
which accounts for about 40% of a
smelters operating costs.
364"-
VAP production at RUSALs Bratsk smelter
In particular, the National Development and Reform Commission (NDRC)

announced at the end of 2013 that
efficient aluminium producers will
continue to pay the same rates, but
less-efficient producers will have to pay
more. According to NDRC, producers
that require 13,700-13,800 kilowatts
to produce a tonne of aluminium will
be charged an additional 0.02 yuan
per kilowatt, while those who exceed
13,800 kilowatt per tonne must pay an
additional 0.08 yuan per kilowatt. The
surcharges would be effective increases
of 1.8% - 7.4% to produce the metal in
Henan province. The government is
hoping that the move will push producers who have kept older facilities
running in the hope of higher prices
to finally cut their losses.
The situation in the industry is nevertheless still characterized by a net
capacity increase. In 2013, despite depressed prices for aluminium, record
high capacities were commissioned
in China in 2013 (4.3 million tonnes)
resulting in a 2.2 million tonnes net
capacity increase.
In the first two months of 2014, the

trend continued as Chinese aluminium
industry experienced a net capacity rise
of 1.6 million tonnes. Shutdowns in the
central and southern parts of China
amounted to 700 thousand tonnes
in Jan-Feb 2014. Some aluminium
smelters in the Central parts of China
continue cutting output to reduce loss
due to the falling domestic aluminium
price.
Over 60% of Chinese aluminum production is underwater at the current
domestic SHFE aluminium price. As
expected, around 3 million tonnes of
Chinese aluminium production will
be cut in 2014 as a result of a low aluminium price. However, some amount
of new low-cost aluminium capacity
will still go into production in Xinjiang
and other North Western regions in
2014.
It should be noted here, that although
China still appears to be a self-sufficient
aluminium market, the countrys 12th
five-year national development plan
presumes transfer of some aluminium
production to the western parts of China

with abundant coal resources and lower
power costs as well as abroad.
Siberia next door

With that said, closer cooperation with
Russia which shares a border with China
could open up new opportunities for
the Chinese aluminium industry that
is taking important steps to improve its
environmental footprint by spearheading innovation and developing renewable energy and reducing its addiction
to coal the source of 70% of Chinas
electricity and a major contributor of
CO2 emissions.
Indeed, with a shared boundary of more
than 4,000 km in length, it is logical
that Russia and China are bound to
develop mutually beneficial cooperation. Russia is home to the worlds
second-largest hydro-energy resources
with 75% of hydro-energy capacities
located in Siberia. The greatest unrealized resources are in Eastern Siberia
and the Russian Far East, perfectly
located to meet growing demand from
China.
AWJ 2014 15
364"-
Pot Room at Rusals Khakas smelter
Cost-effective, renewable and environmentally friendly hydro-energy

constitutes as a major competitive
advantage of the region, home to six
HPPs and eight power plants with possible capacity expansions, Siberias
hydro potential utilization rate is only
20%.
Chinas proximity to Siberia, where
most of the countrys production capacities are based, is yet another factor that would enable China to reap
considerable benefits from expanding
cooperation with Russia. The countrys clear logistical advantage allows
delivering physical metal to Chinese
consumers at lower shipping costs
within 2 weeks, versus 3-4 weeks offered by other global suppliers. This
is particularly important, as Chinese
aluminium smelters are increasingly
being shifted to the Western provinces
which will result in additional transport
implications for downstream enterprises in the East of the country.
Another promising avenue of cooperation with Chinese companies could
16 GLOBAL ISSUES
be the development of downstream

clusters in Russia which have considerable growth potential on the home
market in the coming years. In the light
of expectations for the strong increase
in Russias per capita aluminium consumption and the downstream segments profitability, it is clear that any
capital injections into this area will
generate a healthy return. In terms
of returns potential, aluminium can
production, automotive components
and extrusion production are seen as
particularly promising.
RUSAL is currently working on conversion of its production facilities in the
Western part of the country to produce
aluminium- and aluminium alloysbased automotive components, rolled
and cable products. The potential is
huge. For instance in the automotive
industry, despite a slight drop in car
sales in 2013 due to the negative macroeconomic environment, the Russian automobile market remains the
second-largest in Europe and is poised
to overtake Germany to become Europes largest by 2016, and the worlds
fifth biggest, by 2020, according to

the latest forecasts. Presently there
are only 290 cars per 1,000 Russians,
versus the already saturated market in
Europe, where 560 of every 1,000 is a
car-owner. The first step in this direction
has been made recently, with RUSAL
teaming up with an Israeli company
Omen High Pressure Die Casting to
create a joint venture to produce automotive components at the site of the
Volkhov aluminium smelter.
The worlds biggest aluminium companies RUSAL, Chalco and Shandong
Xinfa Group are already discussing the
prospects for partnership including a
joint smelting project in Siberia, bauxite
exploration and technology exchange
in red mud processing. Moreover, RUSAL has prepared several road maps that
set up plans for investment projects
aimed at transforming its loss-making
aluminium smelters and the development of new hi-tech production, which
are open to foreign capital.
In view of the above, it is clear that the
potential for deepening Russia-China
364"-
UC RUSAL is a leading global aluminium producer
aluminium cooperation is as enormous

as the benefits that both countries
could reap through strengthening their
ties. Therefore, the aluminium sector could become yet another area of
intense bilateral cooperation, on top
of successful projects in oil and gas,
energy industries as well as various
high-tech sectors.
Company profile
UC RUSAL is the worlds largest aluminium producer, accounting in 2013
for approximately 8% and 7% of global
aluminium and alumina production
respectively. The Companys current
capacity allows it to produce 4.5 million
tonnes of aluminium and 11.9 million
tonnes of alumina per annum.
UC RUSAL is vertically integrated to a
high degree, having secured substantial supplies of bauxite and alumina
production capacity. RUSALs assets
include over 40 smelters and production facilities in 13 countries, across 5
continents. RUSAL employs 67,000
people.
The Companys core smelters, located in Siberia, benefit from access to

stranded hydro generated electricity,
with its principal Siberian facilities in
close proximity to important European
and Asian markets.
The Companys key sales markets are
Europe, Russia and the CIS countries,
North America, South-East Asia, Japan
and Korea. The major end users consist
of over 700 companies representing
transport, construction and packaging
industries.
Value added products account for over
40% of total metal produced.
RUSALs ordinary shares are listed on
The Stock Exchange of Hong Kong
Limited (Stock code: 486). Global
depositary shares representing UC
RUSALs ordinary shares are listed
on the professional board of NYSE
Euronext Paris (RUSAL/RUAL). Russian depositary receipts representing
RUSALs ordinary shares are listed
on the Moscow Exchange (RUALR/
RUALRS).
RUSAL owns a 27.8% stake in MMC

Norilsk Nickel, the worlds largest producer of nickel and palladium and one
of the worlds largest producers of
platinum and copper.
Together with the Kazakhstans National Welfare fund Samruk-Kazyna
RUSAL is developing the Ekibastuz
coalfield in Central Asia. The 50/50
LLP Bogatyr Komir coal joint venture
in Kazakhstan provides RUSAL with a
natural energy hedge.
RUSAL is currently focusing on
strengthening its competitive advantages, including its considerable raw
material base, access to renewable
energy sources, proprietary R&D capabilities and proximity to key markets.
UC RUSAL
Phone: +7 (495) 720-51-70
Email: Press-center@rusal.ru
Web: www.rusal.ru/en/
AWJ 2014
17
AP Technology
TM
Rio Tinto Alcans AP Technology solutions:

The worlds most productive smelter technology
2013: Start-up of AP60 pots at

the historical center of aluminium
development in Canada
A new milestone for reduction technology has been recently
reached with the successful startup of the Arvida AP60
Technological Center in Jonquire, Quebec, Canada.
With the demonstration of AP60 at Arvida Technology Center
and APXe at Laboratoire de Recherche des Fabrications (LRF)
inFrance, Rio Tinto Alcan makes available high productivity
andlow energy consumption technologies to its partners
andcustomers, and thereby offering the most productive,
costeffective and cleanest smelting technology in the world.
AP60/APXe: the reduction technology of choice for your project!
18
PRIMARY SMELTING AND PROCESSES
Technology sales department

725, rue Aristide Bergs - BP 7
38341 Voreppe Cedex
France
T +33 (0)4 76 57 85 00
For more information about Rio Tinto Alcan
and its AP Technology solutions, visit
www.riotintoalcan.com
ap-technology.com
PRIMARY SMELTING AND PROCESSES

ECL
ECL makes your operations easier.
p. 20-23
Aluminium Reduction Cell Technology Providers

A 2014 Review: Dr.-Ing. Joachim Heil
p. 24-45
Rio Tinto Alcan

Start-Up Of Arvida Smelter, AP60 Technological Center
p. 46-48
FLSmidth
MLLER Alumina Handling Systems
p. 49-51
Streamline.
Sensotech
ECL
makes your
Inline
concentration
monitoring
operations easier.
p. 52-55
Power Jacks
Precise Anode Beam Positioning from Power Jacks
p. 56-59
ecl.fr
AWJ 2014 19
Streamline.
ECL makes your
operations easier.
20 PRIMARY SMELTING AND PROCESSES
ecl.fr
ECL makes your operations easier.
Regulation system to improve quality of the metal sucked during tapping operation
One of the objectives you can target

from the whole process of primary
aluminium production is to deliver a
metal free from impurities. The tapping
operation consisting of sucking liquid
aluminium from the pot in a crucible
through a tapping tube remains an
operation requiring precautionary
measures. On one hand, the operator
has to correctly insert the tapping
tube into the electrolytic cell at the
lower part of the metal pad. And on
the other hand, the volumetric flow
during tapping is difficult to regulate.
If the flow is excessive, it can result
in the bath being sucked in with the
metal. Bath adjunctions have many
negative effects, both on electrolytic
cell operation and equipment soiling
but above all on metal casting.
To avoid such concerns, ECL works in
close collaboration with Rio Tinto Alcan
in order to develop and adapt a system
based on components of the shelves.
This system allows controlling and
regulating the flow rate of aluminium
sucked from the pot by means of loops
control and signal processing in the
PLC, which controls a valve on the
compressed air supply. 200 tapping

operations have been performed in
the Alma plant resulting in proving the
efficiency and reliability of the solution
and providing significant benefits, such
as less equipment cleaning cycles and
better metal casting.
Here is the reality whatever is the
production: produce more by combining
quality, rapidity, cost-savings and
safety. The Engineering Department
of equipment suppliers such as ECL
works hard to meet these expectations.
The aim is to provide the smelters
with solutions allowing them to both
save money, in particular by supplying
energy-saving equipment or solutions,
reducing equipment maintenance costs;
and produce high quality aluminium
in a safe environment.
The solution of the ECL regulation
system meets all these criteria. It took
as its starting point that a significant
amount of electrolytic bath (typically
15 kg of electrolytic bath per ton of
molten aluminium in most cases)
was sucked during tapping operation
due to a lack of flow rate control. This
undesirable bath intake has negative

effects notably for metal casting and
especially when it comes to producing
certain aluminium alloys requiring a low
sodium concentration. Consequently,
the bath removal from the electrolytic
cell impacts its operation negatively.
Tapping equipment is soiled faster
and metal treatments before casting
are more demanding with regard to
efforts and costs. It should also be
noted that the more the electrolytic
bath is sucked in with the metal, the
more the tapping tube and the crucible will be soiled, eroded and even
blocked. The required frequency of
the cleaning of the crucible therefore
becomes substantially elevated.
The regulation system is also
in correlation with the technical
developments of the electrolytic
process, particularly with the new
standard of low Anode-CathodeDistance (ACD) pots. Decreasing the
amount of ACDs lowers the voltage and
energy requirements of the cell (costsavings) but weakens the stability of
the process, especially during tapping
operation. That is why the ability to
AWJ 2014 21
A risk of sludge aspiration

In the second case, the bath will be
sucked by a vortex effect.
Once well positioned, a vacuum is
induced into the crucible, usually
using an air injector whereby the metal
is aspired through the tube. The air
flow through the air ejector can be
controlled manually using a valve on
the compressed air supply. To resume,
a good tapping operation depends on
the right immersion depth of the tube
(operations conducted carefully and
diligently) and the flow rate control
(good and stable target flow rate).
regulate and control the metal flow

rate to avoid bath fluctuations will
impact positively the stability of the
process.
The objective of the regulation system
is clear: limiting the siphoning of
electrolytic bath during the tapping
operation to minimize those negative
effects and help smelters in their daily
efforts to produce more, cheaper and
faster.
As a brief reminder of the aluminium
production process: many different
operations on the electrolysis cells
are essential in producing metal in the
pots. These operations can be grouped
into two categories: operations related
with anode changing and operations
related with tapping operation. A
tapping operation consists of drawing
liquid metal from an electrolytic cell
and filling a crucible with a predefined
mass of metal. The mass of metal to be
siphoned is predefined, in accordance
with standard operating procedures,
and will depend on the production
levels of the electrolytic cell and the

minimum metal levels required to
maintain a cell in operation. When
it comes to proceeding to tapping
operation, several aspects have to be
taken into account in order to limit
bath siphoning and reaching a good
quality level of molten aluminium. The
tapping operation from an operating
electrolytic cell is usually done with a
crucible embarked on the Pot Tending
Machine.
The first important step is to insert
the tapping tube into the electrolytic
cell at the right depth in the metal.
The insertion should be neither too
deep nor too high above the metal
where the bath is. In the first case, we
can observe:
An excessive speed due to the reduced liquid flow cross section and
consequently an erosion of the cathode. This excessive speed could also
lead to a powerful vortex resulting in
more bath entrainment
In practice, very light touch is required

so as not to overshoot the target metal
flow rate. Consequently a stable metal
flow is rarely, if ever, obtained and very
large fluctuations can be observed
during tapping of a bunch of cells.
Some of the numerous factors, which
can explain some of the variations in
flow rate are for example: the position
of the crucible relative to the metal/
bath interface, any obstructions limiting
free flow of metal into the tube; such
as surface variations on the cathode
surface of the electrolytic cell or
lumps of solidified bath, variations
in air temperature during tapping,
variations in how well the crucible is
sealed during tapping, variations in air
pressure supply, crusting of tube from
bath entrainment, movements of the
metal in the pot etc.
General principle of the tapping

regulation system
Given the difficulty to provide
manually the fine adjustment in
vacuum to maintain an ideal metal
flow rate which maximizes productivity
without compromising quality (bath
entrainment), ECL designed, set up
and tested in the Alma plant a system
based on the automatic control of the
flow rate to reach the target metal flow
rate. Basically the system comprises
among others a control unit and by
means of loops control and signal
processing in the PLC adjusts the
supply of compressed air in the air
ejector through a valve and therefore

the vacuum pressure depending on
the headspace in the crucible and
the weight of the crucible during
tapping.
This system, allowing for siphoning
the metal and for it to be transferred
at a pre-determined target flow rate
into the crucible, consists of:
An air ejector coupled to a source
of compressed air and in close communication with the headspace in the
crucible.
A vacuum transducer between the
source of compressed air and the air
ejector, connected to the control unit
delivering actual vacuum pressure
changes in the headspace.
A valve assembly operated by a valve
actuator responsive to an electric current-to-pressure converter which is
coupled to the control unit. The valve
actuator receives information from
the control unit to determine the flow
through the outlet of the valve assembly. It will open or close the compressed
air supply as needed.
A dedicated algorithm which filters
the weight signal and the vacuum level
to reach a stable target.
Control units connected with weighing means in order to receive weight
measurements and calculate on due
time the liquid flow rate of liquid being
drawn in the crucible. The control unit
then simultaneously adjusts the flow
rate of the compressed air flowing
through the regulating valve in order to
reach the target flow rate. The control
unit includes a programmable logic
controller (PLC). The PLC is directly
connected with a main compressed
air directional valve to open the valve
when a tapping operation begins and to
close it when the target mass of metal
has been siphoned into the crucible.
Advantages of the solution

More than 200 tapping operations have
Schema of the principle of the aluminium tapping operation

Load
cell
PLC
I/P
Master loop
PTA
compressed
air line
ejector
Short loop
P/I
vacuum pipe
been performed with the regulation

system at the Alma plant. The results
are clear. The maintenance of an ideal
metal flow rate maximizes productivity
without compromising quality. The
less the bath is siphoned, the less the
tapping tube is soiled or blocked or
requires changing. The less the bath
is siphoned; the easier the metal is
processed in the cast house. The
less the bath is siphoned, the less
the crucible is soiled and needs to be
cleaned. Consequently we can expect
a decreased frequency of the crucible
bricklaying. All those quantifiable
advantages will help smelters to save
money on maintenance costs and spare
parts costs while decreasing the cost of
the aluminium alloy treatment. Casting
operation will be made easier and the
quality of metal sucked will generate
less waste.
Crucible
integrated directly in the automatic

system of the Pot Tending Machine
or installed in the tapping beam of
the crucible.
Conclusion
The regulation system is a selfadaptive system. It requires no action
or adjustment from the operator. The
system provides transparency and
combines good process quality with
fast potline operation.
Author:
Anne-Galle Hequet
ECL Communication and Public
Relations Manager
The solution, whether we are talking

about a Greenfield project or a
Brownfield one, is adaptable to any
smelter configurations using the AP
Technology. The system can be
AWJ 2014 23
Aluminium Reduction Cell Technology Providers a 2014 Review

Introduction
This article is the second, updated
edition of a paper published in the
context of the European Metallurgical
Conference 2011 (EMC2011), organized
by GDMB of Germany. Special thanks
go to my former colleagues, Dr. R.
Minto and T. Heitling, who helped
establishing the first edition which
has been published in the EMC2011
conference proceedings [1].
Consultation of an article on the topic
published in 2000 gave rise to the
question who would be providing
aluminium reduction cell technology
today.
The referenced article elaborates on
cell technologies developed by wellknown companies which mostly have
been in business for a long time, some
since inception of the Hall-Hroult
process. Potline current values cited in
the article are in a range of 250 320
kA for the then latest technologies;
further tiers of reduction current are the
150 200 kA range and anything below
that down to 50 kA, the latter mostly for

an illustration of the historical evolution
of the electric current as a qualifier for
the advancement of reduction cell
technology.
Since 2000, the global primary
aluminium industry has grown at a
remarkable rate of 5,5 % year-onyear: production capacity rose from
23,7 million tpy (Mtpy) in 2000 to 39,8
Mtpy in 2008, with a recess to some
37,5 Mtpy in 2009 in the aftermath of
the financial crisis, just to rebound to
47,3 Mtpy in 2013. China has grown its
share in the primary aluminium market
from about 10 % in 2000 to some 21,5
Mtpy or 45 % of global supply in 2013,
which is equivalent to almost all of the
above increase in global production
tonnage.
The same period has seen an equally
unprecedented change amongst
the players in the primary business:
mergers and acquisition have led to
a concentration of the industry into
fewer but bigger players. This trend

along with management buy-outs,
bankruptcies and changes of business
strategy has led to the disappearance
of quite a few of the traditional primary
producers names including some of
the long-established cell technology
providers.
128 years after Hall and Hroult
independently applied for their patents
for the still unrivalled aluminium
electrowinning process, this paper
gives an updated review of who today
would be developing and providing
aluminium reduction cell technology to
primary smelters, be it new greenfield
or brownfield expansion projects.
1 Summary of Reduction Cell

Technology as at Year 2000
In early 2000, Tabereaux published a
global review on prebake cell technology [2] in which he elaborated the then
prevailing situation with regard to cell
technology developers and operators.
The article also included an overview
Table 1: Most Advanced Reduction Cell Technologies as at the Year 2000, excerpts from [2]
UPBN: Universal Prebake Cell Nomenclature, proposed by Tabereaux
Company
Cell Type
UPBN
I / kA
Alcan
A-275
AC-28
A-310
AC-31
P-225
A-817
Alusuisse
EPT-18
Comalco-Dubal
CD-200
Hydro
HAL-230
HAL-250
Kaiser
P-80
KA-18
190
Pechiney
AP-30
AP-30
300 - 325
Reynolds
P-20S
RY-17
170
P-23S
RY-18
180
CA-180
VAW-18
180
CA-240
VAW-24
240
CA-300
VAW-30
V-350
VN-35
C-255
C-300
Alcoa
VAW
Venalum
Russia/VAMI
China
Pots installed
Year
Remarks
280
1981/92
310
n. inv.
n. inv.
AA-23
225
n. inv.
n. inv.
Massena, Tennessee
AA-30
300
n. inv.
n. inv.
Portland
AS-18
180
n. inv.
n. inv.
Rheinfelden, closed 1991
CD-20
200
1990
Test cells at Dubal
HAL-23
230
n. inv.
n. inv.
Hoyanger, Venalum PL5 (1988), Slovalco (1995)
HAL-25
250
n. inv.
Test cells in Ardal
1981
Test cells in Tacoma, shut down
2040 + 720*
n. inv.
Various smelters, global spread
n. inv.
n. inv.
Alcasa, Alscon
n. inv.
n. inv.
Test cells at Alcasa
115
1980
Upgraded Tging version now in Nordural, 120 pots
1980/93
Test cells in Tging, shut down for CA-300 prototype
300
1992/93
Test cells in Sayanogorsk, shut down**
320
n. inv.
Test cells
RU-26
255
n. inv.
n. inv.
Tajik, Sayansk, Volgograd
RU-30
300
1992/93
P-280
CH-28
280
n. inv.
n. inv.
Qingyang
P-320
CH-32
320
30
n. inv.
Test cells, Pingguo
Test cells in Jonquire, shut down
Test cells in Sayanogorsk, shut down**
*: 720 cells under construction at that time

**: VAW and Sayanogorsk jointly built and operated a test facility in Russia, each partner contributing 3 pots
n. inv.: not investigated
Dr.-Ing. Joachim Heil MetCons Metallurgical Project Consultancy

of the developmental steps taken by
individual companies. This historic
part of Tabereauxs review will not be
repeated here and interested readers
are referred to the original source. For
this update, only the 2000 spearhead
cell technologies, in terms of highest
amperage will be quoted as reference
points. A condensed summary of those
reduction cell technologies is presented in Table 1.
Table 1 shows more than one entry for
some companies and countries. The
intention is to highlight the development potential that can be seen in
operational test cells.
The original table further included
companies Montecatini, Elkem, Sumitomo and Egyptalum as cell technology
holders. These have been omitted here
as their cell technologies are considered outdated at publication in 2000,
no progress is recognized since, or due
to their solely local significance, all in
the context of this article.
The 10 companies in possession of
aluminium cell technology mentioned
in Table 1 comprise the big traditional
industrial names, four of which can
even be traced back to the inventors
of the Hall- Hroult (HH) process: Alcoa and Alcan are the direct result
of Charles Halls entrepreneurial activities in North America while Alusuisse and Pechiney are the European
offspring of Paul Hroult. VAW can
be considered a late arrival, founded
1917 to support German armament
during WW I but without direct ties to
the founding fathers of the industry.
Kaiser and Reynolds can be regarded
as the generation of heirs as they came
into the primary aluminium business
in close timely connection to WW II,
e. g. by snapping up from Alcoa, in
an US government initiated auction,
what was considered overcapacity
after the war. Norsk Hydro, founded
1905 as a hydro-power company with
associated power-consuming assets
(fertilizers, explosives), entered the
aluminium business even later, after
VAW-led initiatives to build a German

production basis in Norway during WW
II had not been finished before the
end of the war. Although Hydro itself
had contemplated aluminium production repeatedly since 1907 (including
failed own process inventions outside
Hall-Hroult) only in 1963 did Hydro
diversify into the aluminium business
by building its first smelter in Karmy;
later in the last century Hydro started
buying history through acquisition of
older Norwegian smelters [3]. That
leaves Comalco-Dubal and Venalum
as representatives of an upcoming
new generation of more recent birth
and, due to the still prevailing lack of
detail insight (in 2000), the Russian
and Chinese aluminium industries
pooled by Tabereaux just under the
country names.
Concluding from the above summary,
it can be seen that by the year 2000,
know-how that had been deployed
internationally appears to be almost
exclusively held by big western enterprises with a long history in the industry
to the extent that the original inventors
can still be traced. The Russian and
Chinese industries had been contained
within their respective borders and, due
to their lack of involvement outside of
their territories, had remained opaque
until way into the 1990s. However, internal cell development had reached a
similar amperage level as the western
technologies.
The reduction cell development had
obviously reached peak line amperages
of 300 325 kA while a lot of cell technologies still hovered at between 180
and 280 kA. Tabereaux in his outlook
mentions, without being specific, that
further testing into the 400 kA region
was underway and that this amperage
was expected to establish the next
reduction cell generation.
While in principle aluminium can be
produced in cells with either Sderberg
(S) or prebaked (PB) anodes, all of
the modern high-amperage cells are
based on prebaked anodes. Another

distinguishing element of reduction
cell construction and operation is the
concept of supplying the alumina feed
to the electrolyte. Historically, PB cell
feeding has been developed from side
work (SW) to center work (CW), and
finally to point feeding (PF) systems.
While SW pots were fed (several) hundred kilograms of alumina at a time
in intervals of 1 several hours, CW
pot feeding occurred in doses of tens
of kilograms several times per hour
and PF feeding involves quantities of
1 1.5 kg/shot some 2 3 times per
minute. All modern high-amperage
cells exclusively utilize point feeders
and can thus be characterized as pointfed pre-bake or PFPB cell types.
Finally, at the bottom line of Tabereauxs
article, his minibio significantly refers to
Dr. Tabereaux as working for Reynolds
Metals. However, the article was printed
just a few weeks before Alcoa finally
finished its takeover of Reynolds Metals
in May 2000. This leads to the indicated
sub-topic of dramatic changes in the
primary aluminium industry since publication of Tabereauxs article which
will also be highlighted below.
2 State of the Primary Aluminium

Industry at the Turn of the
Millennium
The 1990s had started off with one
of the worst economic periods in the
primary aluminium industry: as a consequence of the fall of the iron curtain,
aluminium that would have otherwise
been used by the former Soviet Union
and its allies was sent into the western markets and particularly into LME
warehouses. At that time, the traditional
correlation between metal inventory/
consumption and price was still intact,
so the influx of excess metal sent the
LME prices, coming from above 2.000
USD/t (incl. a peak of above 3.500
USD/t) into steep decline down to the
1.100 USD/t range at which level almost
all smelters would face losses. It took
the industry huge joint efforts in terms
of mutually agreed curtailments for
the price to escape the 1.100 1.300
AWJ 2014 25

4.000
Monthly Average Primary Aluminium Price (LME spot) in USD/t
3.500
Al Price LME spot
Mean
3.000
Median
Mode
2.500
2.000
1.666
1.552
1.500
1.164
1.000
500
Figure 1: Monthly Average Primary Aluminium Price, 01/1981 04/2014 [4]
USD/t range which only succeeded in

about mid 1994. Players and individual
smelters with less solid balance sheets
were forced into shutdowns or became
prey for takeovers.
In addition to considerable pressure
from marginal product proceeds at
low LME prices, cost pressures were
also on the rise, particularly from the
energy cost end. Smelters faced expiry
of their long-term power contracts and
more often than not the new contracts
included hefty increases of electric
power prices. In this context, the Bonneville Power Administration (BPA),
a US governmental (not-for-profit)
power agency, achieved some doubtful fame as a result of pressurizing
their US aluminium clients for many
years, in some instances to the brink
of bankruptcy.
Only during the second half of the
1990s did the primary aluminium industry regain enough stability to be
able to entertain new developments.
In reflection of the tough times, the
aluminium industry started forging
stronger entities through mergers and
by acquiring weaker players.
3 Aluminium Industry Consolidation

at Corporate Level from
2000 2014
3.1 Western Primary Aluminium
Industry
The new millennium started off with
two major reorganizations among the
big western players. In May 2000, Alcoa
finalized the acquisition of Reynolds
Metals in a 4,5 blnUSD deal, almost
one year after the offer had been submitted [5]. The merger combined the
two biggest aluminium producers of
the US, or numbers one and three on
a worldwide scale, making Alcoa by
far the biggest aluminium producer
globally.
Soon after, in October 2000, Alcan
(of Canada) finalized its merger with
Alusuisse (of Switzerland) [6]. This
merger was what remained of an initially contemplated three-way merger
that would have included Pechiney
(of France) as well. However, the idea
of including Pechiney was mutually
abandoned as the project faced stiff
opposition from regulatory authorities
over market dominance in the flat-
rolled products business resulting from

Alcans 50 % ownership in the giant
Alunorf rolling mill in Germany. Alcan
now was number two on the global list
of primary aluminium producers.
In February 2002, Kaiser Aluminium,
then the third largest aluminium producer in the US, filed for bankruptcy
protection under Chapter 11 following
a failed debt repayment of some 25
MUSD and facing another upcoming
debt repayment of 174 MUSD. The
Kaiser bankruptcy was mainly attributed to a failed diversification into
the chemical business [7]. However,
the weak aluminium business during
the 1990s will have contributed its
share. Additionally, Kaiser had been
hampered by an explosion, in July
1999, at its Gramercy alumina refinery,
which took its 1 Mtpy production off
the market for 1,5 years [8]. Kaiser was
also one of the victims of BPAs new
increased power tariffs which, among
others, forced them in late 2000 to
contemplate shutting down its Mead
smelter and selling the freed power
back to BPA at the higher price. Ironically, this idea was opposed by BPA

(and thus the US government) as they
did not entertain a private company
making a windfall profit to the tune of
300 MUSD out of a public utility [8]. It
actually took Kaiser until 2006 to reemerge from Chapter 11 protection.
Still an aluminium company today, Kaiser has, however, divested all alumina
and primary aluminium assets. Under
the new business model, Kaiser is now
a producer of engineered aluminium
components with an emphasis on the
aerospace market [9].
In March 2002, Hydro Aluminium (of
Norway) took over VAW aluminium AG
(of Germany) from E.ON AG, a German holding company formed in 2000
through the amalgamation of VIAG AG
and VEBA AG, in a 3,1 bln deal [10].
VIAG had been the holding owner of
VAW since its inception. VIAGs portfolio included basically power producing and power consuming industries
whereas VEBA held a portfolio of power
producers and chemical plants. The
new E.ON strategy was to concentrate
on power generation so all industrial
holdings, including VAW, were divested
as a consequence.
The takeover of VAW promoted Hydro
Aluminium to position four, behind
Alcoa, Alcan and RusAl, in the global
primary aluminium producer ranking.
Meanwhile, the new Alcan had obviously not entirely given up the idea of
integrating Pechiney since in September 2003 they gained clearance from
the European Commission, though
there was an obligation to divest major
parts of the downstream business including the flat-rolled production [11].
The latter was finally spun-off in 2005
as Novelis which now, since 2007, is
wholly owned by Hindalco. The incorporation of Pechiney boosted Alcans
primary aluminium output close to that
of Alcoa, however Alcan remained in
second place.
After almost 2 years of a long unsuccessful courting period, Alcoa then
made an unsolicited takeover bid to
Alcan early May 2007 [12] which was
immediately rejected as it supposedly

did not properly reflect the true value
of the new Alcan [13]. Alcoa bid 33
blnUSD for Alcan, however, after Alcan managements rejection of Alcoa,
Rio Tinto offered 38 blnUSD. When
Vale (CVRD at the time) also entered
the takeover-war, Rio Tinto and Alcan
settled the deal at 38,7 blnUSD, one of
the biggest takeovers ever. In October
2007, the aluminium activities of Rio
Tinto, i.e. the Comalco business, were
combined with Alcan and are known
today as Rio Tinto Alcan or RTA. The
combined primary production has put
RTA in second place, closely behind
the new RusAl.
In May 2010, Hydro Aluminium signed
an agreement with Vale to take over
Vales aluminium business (primary
smelters, alumina and bauxite activities) for 4,9 blnUSD [14]. After approval
from regulatory authorities, the deal
was finalized early 2011 [15], giving
Hydro upstream access to bauxite and
making Hydro a long alumina producer.
To summarize, the last decade has
shrunk the number of potential western
reduction cell technology providers
from 10 (or rather 8 + 2, the 2 being
Comalco-Dubal and Venalum) to 3 +
2: Alcoa, Hydro Aluminium and Rio
TintoAlcan + Dubal and Venalum, see
graphic representation in Figure 2.
Dubal appears to have discontinued the
joint technology development agreement it had with Comalco before 2005
and now has developed its own DX
series of high amperage cells. While
Dubal is continuing with reduction cell
development no similar information
is available from Comalco since 2006
- when Comalco reported about five
modified CD26 test cells operating at
the Boyne smelter, which were being
considered for the intended potline 1
and 2 modernization. The so-called
B32 (RTC-28) cell was operating at
270 and 280 kA between 2002 and
2005 [16]. Interestingly enough, for
Boynes potline 3 construction between
1995 and 1997, Rio Tinto Comalco had

already opted for AP-30 technology
over the in-house CD technology. Developments of Comalco cell technology have probably been discouraged
after the Rio Tinto Alcan merger in
2007 since this has given Rio Tinto/
Comalco direct access to the more
advanced Pechiney technology.
3.2 Eastern Primary Aluminium

Industry
Russia started primary aluminium production on an industrial scale in 1929.
All Soviet smelter technology R&D
was concentrated in the All-Union Aluminium Magnesium Institute (VAMI)
founded in 1931 (and re-named AllRussian Aluminium Magnesium Institute VAMI in 1993) [17]. Historically,
Sderberg technology had long been
dominant, and still continues to be
largely present, in Russian smelters.
The dissolution of the communist bloc
after the fall of the iron curtain brought
about unprecedented upheavals in
the formerly planned and centralized
economies, specifically in the Former
Soviet Union (FSU). Both, aviation and
armament industries, the biggest consumers of aluminium in the FSU, had
broken away almost entirely, and domestic consumption dropped from
17 kg/capita in 1990 to a mere 2 kg/
capita in 1994. Before production outputs could be adjusted, an overhang of
aluminium had been produced which
was subsequently shipped westward
deluging the global markets. FSU
smelters found themselves disconnected from their alumina supplies
which were now situated in foreign
countries (i.e. in the now independent
previous Soviet republics) and started
operating on a tolling basis. In an almost lawless, mafia-like environment,
proverbial aluminium and alumina
wars took place with huge profits to
be made but also leaving casualties
at the wayside. Since the state-owned
smelters were effectively ownerless,
a major privatisation took place from
1993 onwards.
AWJ 2014 27

In this environment, a few individuals
started building ownership in individual
smelters, then progressing into grouping individual plants together to form
strong groups almost mimicking the
earlier communist structures, but now
under private ownership. So-called
oligarchs concentrated aluminium assets under the names Sibirsky Alumini
(1997, Oleg Deripaska), Sibneft (1999,
Roman Abramovich) and SibirskoUralskaya Aluminievaya Kompania
(SUAL, 1996, Viktor Vekselberg).
Also in the eastern hemisphere, the
new Millennium started with yet another major concentration of market
share. In 2000, Sibirsky Alumini and
Sibneft merged to form Russian Aluminium (RusAl) with a production
capacity of more than 2 million tpy of
aluminium representing almost 10 %
of global output [18].
During the following years RusAl and
SUAL grew independently through
further acquisitions of international
scope and in 2003, RusAl acquired the
All-Russian Aluminium Magnesium
Institute VAMI [19].
In 2007, with the merger of RusAl,
SUAL and the alumina business of
Swiss trading house Glencore, a new
industrial giant was born. The new
United Company (UC) RusAl was then
worth some 30 blnUSD and controlled
4,4 million tpy of primary aluminium
output - placing the new RusAl on top
of the producers ranking and overtaking Alcoa [20].
In summary, the Russian primary aluminium industry is now controlled by
UC RusAl. RusAl, after a total disintegration, in the 1990s, of the stateowned assets, has almost rebuilt the
Soviet-era industry including control
of the VAMI R&D facilities, though now
under private shareholding ownership and with a global reach, through
acquisitions.
The early days of the Chinese primary
aluminium industry remain obscure
due to a combination of long-lasting

shielding of the country and the existence of a multitude of small smelters
(down to the 5 ktpy level) which went
unrecognized globally or remained
unknown due to non-reporting. According to Zhongxiu, in 2002 there were
still 128 operating Chinese smelters
with only 17 smelters having more than
50 ktpy capacity [21]. Taking the IAIpublished Chinese production figure
of 4,321 Mtpy for 2002 into consideration [22], the average output from a
Chinese smelter was a mere 33,7 ktpy.
By 2013, China had increased primary
output to 21,936 Mtpy [23] equivalent
to an average of 175 ktpy from each of
its 125 operating smelters.
The ownership of Chinese smelters
appears to be scattered between the
government, semi-public entities and
partially or wholly private ownership.
The largest single Chinese entity in
this context is the Aluminium Corporation of China Ltd. (Chalco), which was
formed in September 2001 to oversee
the aluminium and alumina business
of state-owned Aluminium Corporation
of China (Chinalco). Chalco was partly
floated on the New York and Hong
Kong stock exchanges in December
2001 which reduced Chinalcos majority ownership to some 44 % while
Alcoa picked up an 8 % share of Chalco
[21]. Chalco has continued to expand
by acquisitions (of other Chinese
smelters) and by building new smelting capacity at rapid pace. Despite a
production increase from 690 ktpy in
2000 to >4,2 Mtpy in 2012, Chalcos
share of the total Chinese primary aluminium output has, however, fallen
from 25 % to some 21 % [23], [24].
Concluding from company information collated by Pawlek [26], Chinese
aluminium production appears to have
started in the 1930s, based on VAMI
Sderberg pots, but later Elkem and
Japanese technology providers have
also been sporadically mentioned. In
the 1980s, obsolete Japanese smelter
equipment was imported into China
(as a consequence of Japan exiting the
primary business after the oil crisis)

and the VAW CA 115 from Tging (as a
consequence of the smelter shutdown
in 1994 after Russian metal flooded
the market) had been bought secondhand.
However, the overwhelming majority of
Chinese smelters apply home-grown
which has historically been developed
by two institutes: Shenyang Aluminium
& Magnesium Engineering & Research
Institute (SAMI, founded in 1951) and
Guiyang Aluminium Magnesium Design
& Research Institute (GAMI). Both are
now managed by the China Aluminium
International Engineering Corporation
(Chalieco), which is a wholly owned
subsidiary of Chinalco. These two institutes, SAMI and GAMI, have recently
been developing high-amperage cell
technologies separately and they are
competitors, even though both have
the same parent company. SAMI and
GAMI designed potlines constitute
the bulk of Chinas current primary
aluminium industry.
Established in 1981 and restructured
in 2003, the Northeastern University
Engineering & Research Institute
(NEUI) has followed a similar technology development path as SAMI,
and within a few recent years, NEUI
has developed and put into operation
a series of high-amperage reduction
cell technologies in China.
The historic development of western
and eastern reduction cell technology
providing companies is graphically
summarized in Figure 2.
4 Aluminium Reduction Cell

Technology Providers at the
Turn of 2013/2014
4.1 Alcoa
Alcoa has not reported any progress
on their 300 kA cell technology since
more than a decade as far as the TMSs
annual Light Metals proceedings are
concerned. Actually, it appears that the
only industrial application of Alcoas
Primary Aluminium Ancestry
Figure 2: Historic Timeline of Reduction Cell Technology Providers
own most advanced reduction cell is

at Portland Aluminium in Australia.
The acquisition of Reynolds by Alcoa
in 2000, including their cell technology
R&D department, did not bring about
any obvious revival of cell technology
development activities at Alcoa.
Alcoas North American operations,
which utilize Alcoas own cell technology, are applying line currents of
between 120 kA and 245 kA, according
to information available from Pawleks
PASaPoW [26]. Among these there are
3 smelters that exceed 210 kA, namely
Mt. Holly (215 kA), Tennessee (245 kA)
and Massena (230 kA) while the latter
also houses an unspecified number of
A-716 type test pots operating at 280
kA and 450 kA (?).
Much of Alcoas global assets today
have been acquired, i.e. these have
an inherent lower probability of using
Alcoa cell technology, and actually Alcoa inherited a wide variety of different
technologies from the original owners.
However, there are again 3 smelters
outside of the USA using Alcoa cell
technology beyond 210 kA: Point Henry
(215 kA, P-155 cells), Alumar (228 kA,

A-697 cells) and Portland (320 kA,
A-817 cells). Concluding from PASaPoW [26], Portland appears to be the
only smelter in the Alcoa organization
that has been built using Alcoas most
advanced technology. Portland was
commissioned in 1986 with an initial
line current of 275 kA, which has obviously been crept to 320 kA.
Since Portland was started up in 1986,
Alcoa appears to have reduced activities in terms of building its own new
smelter capacity. Only in the second
half of the first decade of the new millennium, did Alcoa resort to expand
through building new smelters: Alumar
underwent 2-step brownfield expansions which were commissioned by
September 2005 and from November
2005, respectively. Alcoa A-697 cell
technology (developed as AA-18, after
boosting now operating as AA-23) has
been used for the new potline 3 at
Alumar. In April 2007, Alcoa started
commissioning its new Fjaral smelter
in Iceland - which presently operates
at 380 kA. Interestingly, Alcoa did not
implement its own cell technology but
built a one-potline smelter based on

Alcan (i.e. Pechiney) AP38 cell technology. Also in Alcoas most recent
participation in the Maaden smelter
project in Saudi Arabia Rio Tinto Alcan
AP37/39 technology has been implemented [27].
The European Economic Commission (EEC) in 2003, on the occasion
of the Alcan/Pechiney merger, issued
a merger procedure that assessed the
concentration of market shares for the
new entity. Amongst other items, the
market shares of a combined Alcan/
Pechiney in the aluminium reduction
cell development and licensing business were investigated in relation to
their competitors. One of the competitors mentioned by Alcan/Pechiney was
Alcoa. However, the EEC assessment
found that Alcoa in fact had ceased licensing cell technology to third parties
in the 1980s. Consequentially, Alcoa
was regarded by the EEC as a hypothetical competitor only [11].
As a conclusion of the above, it seems
that Alcoa not only has largely discontinued implementation of its own
AWJ 2014 29

Table 2: Hydro Aluminium Cell Performance Data at Slovalco as per [26]
reduction cell technology in smelters

they own but has also discontinued
licensing to third parties. The latest Alcoa greenfield projects are based upon
reduction cell technology licensed
from RioTintoAlcan. This together
with the total absence of publication
of cell technology advances could
be interpreted that Alcoa has abandoned primary aluminium reduction
cell development altogether in favour
of external licensing.
4.2 Hydro Aluminium (incl. VAW)

When Hydro Aluminium acquired VAW
in 2002, the VAW cell technology R&D
department was also included in the
deal. VAW had operated five CA 240
(VAW-24, in Tging) and three 300 kA
test cells in Sayanogorsk, the latter
project having been hampered by the
Russian conditions in the years just
after 1990. This experience lead to a
VAW decision to replace the VAW-24
cells in Tging with CA 300 (VAW30) test cells. However, this project
was stopped in 1994, shortly after orders had been placed and construction work had begun. The so-called
Tging potline 2, which was to receive
the test cells, was decommissioned
(as a result of Russian metal flooding
the market depressing the LME ingot
price), dismantled and finally rebuilt in
Iceland (Centurys Norurl smelter).
The former VAWs cell technology R&D
group (aka VAW-ATG) continued to work
on cells, mostly on smelter upgrades,
retrofits and the like but the VAW-30
remained shelved. However, Hydro
acquired the residual know-how and
also the manpower and modeling and
engineering tools developed by VAW.
Today, the ex-VAW R&D know-how is
a vital part of the Hydro Aluminium
cell technology development as can
be concluded from ongoing Hydro
publications including former VAW
staff.
Hydro Aluminium had licensed its
HAL-23 cell technology to Venalum
(potline 5, commissioned 1988) and
also to the Slovalco smelter where
the HAL technology replaced three
Cell Technology (UPBN)

Parameter
HAL230
(HAL-23)
HAL250
(HAL-25)
Unit
Amperage (design)
230
250
kA
Amperage (operation)
230,3
258
kA
Number of Pots / Potlines
172 / 1
54 / extension
Current Efficiency (CE)
96
94
Anode Effect Frequency (AEF)
0,044
n.a.
AE/(day pot)
Specific Energy Consumption
13,5
13,2
kWh/kg aluminium
1950s Sderberg potlines. Slovalco

commissioned the HAL pots from June
1995 and achieved operational results
as presented in Table 2.
Slovalco was expanded by adding 54
pots of HAL250 technology which was
commissioned from July 2003. At the
same time the line amperage for the
existing potline had been increased
to match the HAL250 technology of
the new pots. Today, Slovalco operates
at 258 kA.
In December 2002, Hydro started commissioning 11/2 potlines comprising
340 pots in its Sunndalsra smelter
(the so-called Sunndal 4 or SU4 project), also replacing older Sderberg
potlines, implementing their HAL250
cell technology. Even during commissioning the amperage was raised
to 275 kA - the reported value when
the last pot was energized in August
2004. This cell technology is dubbed
the HAL275 (HAL-28) and the Sunndal
smelter is the biggest European single
site smelter [28], [29]. It has been reported that the HAL275 pots at SU4
have been crept to 290 kA (HAL-29)
as of April 2007 [26].
It appears that both the Slovalco and
the Sunndal SU4 potlines might go
down in history as the last newly-built
smelters in (Central) Europe, or at least
the last for quite some time to come,
unless the European energy prices allow for new smelter projects to proceed
again in the future.
The HAL275 cell technology was also
licensed to the new greenfield smelter
Qatalum, in Qatar, which was started
up in December 2009. According to
information available on the Qatalum

website, the operation was supposed
to start at 300 kA which would allow
a production of 585 ktpy of potroom
metal from their 704 pots [30]. This
would require a current efficiency of
94,5 %. Output in 2012 reached 628
ktpy [31] which would have required
an amperage creep to some 320 kA at
95 % CE, so the Qatalum pots should
now be categorized HAL-32. The rectifier-transformers (RTs) installed at Qatalum (5 x 85 kA) would even have enough
rated capacity for future line amperage
creep to 340 kA without compromising on the N+1 RT configuration [32].
In its latest development, in June 2008,
Hydro Aluminium has commissioned
six HAL420 or HAL4e (HAL-42) cells in
its rdal research facility, operating at
420 kA and designed to operate at up
to 450 kA. The first commercial implementation of the HAL4e technology
was foreseen to begin after 2014 [33].
In 2013, a 70 ktpy pilot smelter applying
HAL next generation technology to
be sited at Karmoy was under study
[34]. The pilot HAL-42 cells achieved
specific energy consumption of 12,5
kWh/kg in 2012, with a 2014 target of
12,3 kWh/kg and a mid-term target
of <11,8 kWh/kg for an extra energysaving variant called HAL4e ultra [35].
A full set of performance data from
the first months of operation of the
HAL-42 test cells had been published
in 2009, and the results achieved are
shown in Table 3.
One distinguishing unique HAL technology feature common to all of the
above mentioned variants (except perhaps at Venalum) is that a HAL potline
is housed under one common roof.

2 rows of pots.
This HAL specific potline configuration
is very advantageous in terms of land
usage, i.e. the annual output per m2
of built-up area is comparatively high.
The HAL potline concept also achieves
lower potroom construction investment
and operating costs.
A satellite image comparison of a traditional vs. a HAL potline arrangement
is shown in Figure 4, whereas the yellow lines are 1000 m and 250 m long,
respectively.
Figure 3: Typical HAL-32 Potline, Photo: copyright Qatalum
This is called by Hydro Aluminium

the double potroom concept, or alternatively the half-potroom concept.
Usually, modern PFPB side-by-side
potlines consist of two rows of pots.
These are traditionally housed in two
distinct buildings (potrooms) which are
spaced apart by an open courtyard of
typically some 60 m open width to keep
the reciprocal magnetic disturbance of
the two rows at a minimum.
Due to the courtyard, the center-tocenter spacing of pots between the
two rows would be of the order-ofmagnitude of 80 90 m and maybe

more for the very high amperage cell
technologies. Hydro Aluminium places
the two rows of a potline in two halfpotrooms which share a common yet
unclad central building wall instead
of an open courtyard. The center-tocenter spacing of HAL pots between
the two rows is then only about 30
m [37]. This configuration somewhat
resembles the traditional end-to-end
potline arrangement where there are
2 potrooms but each of them housing
Table 3: Hydro HAL420/HAL4e (HAL-42) Cell Performance Data as per [36]

Parameter
Value
Unit
Amperage
420
kA
Number of Pots
Test cells
95
% (assumed)
Pot Voltage
4,1
< 0,03
AE/(day pot)
12,83 12,93
kWh/kg aluminium
1 potline, 360 pots (AP36). 360 ktpy - Aluminium
2 potlines, 2 x 352 pots (HAL275), 585 ktpy - Qatalum
Hydro Aluminium also reports that

its development will consider potshells with forced cooling (with an
undisclosed cooling medium) on the
sidewalls and usage of the resulting
off-heat for power generation. Heat
extraction from the pot off-gas in the
GTC area for district heating purposes
is already a feature of some Norwegian smelters. Another topic of Hydro
technology development is dealing
with concentrating the CO2 content
in the pot off-gas (from <1 % to > 4
%) which would reduce the size of
gas handling and treatment equipment and eventually facilitate future
uses, e.g. in CCS (carbon capture and
sequestration) [37].
2010 Google
2011 LeaDog Consulting
2011 GeoEye
The HAL-32 technology based Qatalum smelter cost was 9.000 USD/ktpy
installed capacity [38].
4.3 RioTintoAlcan (including

Comalco, Alusuisse & Pechiney)
As already discussed, RTA is now
pooling the previous R&D activities
of Comalco, Alcan, Alusuisse and
Pechiney. The current RTA reduction
cell technology is equivalent to the
former Pechiney technology (RTA
technology is still marketed under the
APXX denomination). In the context
of this review, it is assumed that RTA
reduction cell technology today is
equivalent to Pechiney AP technology
and the other technology developments have been discontinued or, if
not, at least their contribution remains
Figure 4: Land Usage of 1 AP Potline vs. 2 HAL Potlines (yellow lines: 1000 / 250 m long)
AWJ 2014 31

Table 4: Overview of Smelters based on RTA AP Cell Technology as per [44], [45]
invisible to the public (this contrasts

with RTA alumina handling and storage technology which is still marketed
by RTA under the previous Alusuisse
brand Alesa). Pechiney has a longstanding and well documented track
record of reduction cell technology
development. Their AP18 (180 kA)
technology was commercialized in
1979 and almost 10 years later, the
AP30 was first commissioned on an
industrial scale in 1986. The first higher
amperage applications were both built
inside the Pechiney smelter facilities
at Saint-Jean-de-Maurienne, France.
Extrapolating from this historical path,
it was justifiable for Tabereaux to expect
the launch of the next generation AP
reduction cell about the time he wrote
his review in 1999. The next generation
was expected to be of 400 kA while
he also expected that this required
the solution of some technical problems, e.g. wear of cathode lining, heat
balance, emissions, cell instabilities,
higher magnetic fields and metal loss
due to increased cell turn-around time
for relining [2].
Tabereaux was not mistaken, since in
July 2000, Pechiney indeed presented its new cell generation. Pechiney,
however, had skipped the 400 kA and
immediately went to the AP50 technology - to be operated at 500 kA [39].
Within about a year, a first project site
was identified at Coega/RSA to host
a 460 ktpy greenfield smelter, which
was to be the first commercial implementation of the AP50 technology on
a large industrial scale. Agreements
for power supply with Eskom were
made and environmental clearance
was achieved by early 2003, however
Pechiney looked for investment partners as they only wanted to retain about
40 % ownership in the project. After
Alcan had gained control over Pechiney
in late 2003, including the Coega project, the project was delayed triggering investigation of several alternative
scenarios. The whole process was further protracted due to Rio Tinto then
taking over Alcan which, in mid 2007,
resulted in a downscaling of the project

Parameter
AP3X
(AP-30/39)
Unit
Total Potlines (PLs)
19 + 3 *
PLs
Total Pots
5274 + 810 *
Pots
Average Pots
280 (excl. u/c pots)
Pots/PL
Total Installed Capacity
5,25 (excl. u/c pots)
Mtpy
Average Output
ktpy/PL
Avg. Potline Voltage **
V/PL
*: 3 PLs with 810 pots under construction in Iceland and India; pots not included in below
calculations
**: assuming 4,2 V/pot
to 360 ktpy combined with a decision to

implement the project with AP36 cell
technology. In the winter of 2007/08,
Eskoms severe shortfall of maintaining power generation and distribution
systems came to the surface - leading
to country-wide blackouts in RSA. This
was probably only the last in a string of
events that caused RioTintoAlcan to
abort the Coega AP50 project finally
in October 2009 [40].
was achieved in December 2013 [42].

Jonquire could later be expanded to
460 ktpy using the second generation
AP60 cells which would be operated
at 600 kA [43].
RTA still markets its AP30 technology successfully which has been further developed stepwise. Due to the
creeping amperage this technology is
now called AP3X and can be operated
Table 5: RTA AP3X and AP50 Cell Performance Data as per [45], [46], [47]
Parameter
Value
Unit
Amperage
300 500
kA
94,1 96 ,0
Pot Voltage
4,2
0,23 < 0,03
AE/(day pot)
13,01 13,41
kWh/kg aluminium
Obviously frustrated by the inability to

launch the AP50 at Coega, Alcan had
started building a semi-industrial short
potline of 44 AP50 pots within its own
organization, at the Jonquire smelter
in Canada. Commissioning of this 60
ktpy potline was envisioned for mid
2008. However, the financial stress
caused by Rio Tintos 38 blnUSD outlay
for Alcan still persisted when the global
financial crisis started to hit in 2008.
This did not favour the Canadian AP50
project which was then slowed down.
During the slowdown, the project was
re-engineered and RTA announced
that the Jonquire short potline will
now receive the latest development,
AP60, instead of the AP50 previously
announced [41]. In keeping with the
60 ktpy production capacity target,
the pilot potline now consists of 38
pots of first generation AP60 cells
operating at 570 kA after full capacity
at up to 390 kA. RTAs AP3X range of

reduction cells has so far dominated
the reduction cell technology licensing
business outside of Russia and China.
The AP technology market share of
the worlds modern smelters outside
of Russia and China is estimated to be
at least 80 %. The global application
basis of AP3X is summarized in Table
4. Besides that, there is one 405 kA
potline under construction at Kitimat.
The latest AP performance data can
be characterized as follows (see Table
5), summarizing from various publications in TMS Light Metals and RTA
company brochures. This appears to
be supported by the RTA confirmations
that the AP3X and the AP50 test pots
have maintained their performance
data level throughout the entire amperage range.

The higher amperage range of the
AP3X reduction cells is understood
to be applied to pots with unchanged
outer dimensions with moderate adjustments to anode size and potlining.
This means that at the high amperage
end, current density and energy input
to the AP3X cells is higher compared
to the basic AP30 cell. It is also understood that this will require forced
sidewall cooling, which consists of
low pressure air blown through channels attached to the sidewalls of the
potshells. The resulting heated air is
released to atmosphere.
The AP-36 technology based Sohar
smelter was built at 6.670 USD/ktpy
installed capacity [48], while the AP60 pilot potline has cost a staggering
18.330 USD/ktpy [42], and it remains
to be seen how much this cost can be
lowered for a full commercial smelter
project.
4.4 United Company RusAl

(including VAMI)
Most of UC RusAls aluminium smelters were built between 40 and 60
years ago, and the majority of these
smelters are still based on Sderberg
technology [49]. According to RusAl,
more than 80 % of Russian primary
aluminium originates from Sderberg
cells [50] while the international share
of Sderberg smelters was only 18 %
in 2005 [51]. Modernizing their Sderberg aluminium production sites has
an ongoing high priority for RusAl (dry
anode technology, hooding, gas treatment, anode gas incineration, alumina
Figure 5: Typical RA-30 Potline, Photo: copyright Rusal
feeding etc.). Prebake smelters have

been built in the FSU from around
1975 [26]. An overview of RusAl high
amperage reduction cell performance
is presented in Table 6.
A year into its existence RusAl started
development of a high amperage PFPB
reduction cell (in 2001) and five pilot cells were commissioned at their
Sayanogorsk smelter (SAZ) at the end
of 2003. The so-called RA-300 (RA30) reduction cells have been used for
the construction of the Khakas smelter
(KhAZ) which was started-up in 2006
and operates 341 (336 + 5?) pots at 320
kA. In 2005, a newly developed RA-400
(RA-40) prototype was commissioned
at SAZ, and by 2010, sixteen RA-400
cells were in operation at 435 kA.
As example for a typical Rusal PFPB

potroom see a photo from the Khakas
smelter in Figure 5.
The RA-400 is to be installed at RusAls
new Taishet smelter; construction commenced in 2007 but was suspended by
the end of 2008. The Taishet smelter
will comprise 672 pots with production
capacity of 750 ktpy [57].
BEMO (Boguchanskoye Energy and
Metals Complex) is a combined hydropower plant (HPP) and aluminium
smelter project under construction.
The 3 GW HPP project originally
started 1979 but was stopped from
19942005.
Meanwhile 6 out of 9 generators are
Table 6: RusAl Cell Performance Data as per [52], [53], [54]

Parameter
OA-300M1
(SU/RA-30)
RA-300
(RA-30)
RA-400
(RA-40)
RA-500
(RA-50)
Unit
Smelter Site
IrkAZ
KhAZ/ *
SAZ/ **
SAZ
Amperage (design)
300
300
400
500
kA
330
320
415 435
520
kA
Number of Pots
200
336 + 672*
16 + 672**
94
95
> 93,5***
> 93,5***
Pot Voltage
4,33
n.a.
4,3 - 4,4***
4,3 - 4,4***
Anode Effect Freq. (AEF)
0,13
0,15
< 0,05***
< 0,05***
AE/(d pot)
Specific Energy Cons.
13,73
n.a.
< 13,8***
< 13,8***
kWh/kg Al
*: under construction (BEMO project, 588 ktpy)

**: under construction (Taishet project, 750 ktpy)
***: target values
AWJ 2014 33

operating, and smelter construction
would see first hot metal later in 2014.
The smelter comprises 672 pots of
RA-300 technology for a total output
of 588 ktpy [58].
is expected to be < 12 kWh/kg. In the

absence of information to the contrary,
it is assumed that a cryolite-based
electrolyte would be used as opposed
to the chloride-based trials that Alcoa
conducted in the late 1970s using a
similar cell but with multiple horizontal
bipolar electrodes [50], [53].
Before their merger with RusAl, SUAL

reported that they were operating six
OA300M1 type 300 kA test cells (SU30) at its Ural smelter (UAZ), designed
by SibVAMI. Commissioned in 2005,
the amperage of the test cells was later
increased to 330 kA. In early 2010, a
full 170 ktpy potline (potline 5) at Irkutsk (IrkAZ) was commissioned with
plans to increase the amperage to 330
kA. The IrkAZ potline 5 comprises
200 OA300M1-based pots which are
now (after the merger with RusAl) also
dubbed RA-300 [55].
RusAl claims that they can build a

smelter in Russia at a cost of 2.300
2.800 USD/tpy installed capacity [56].
The Khakas smelter is said to have been
built in less than 24 months.
4.5 Dubal
Dubal started operations in 1979 with
3 potlines implementing National
Southwire technology (an improved
version of Kaiser P69 (KA-15)) [59].

The reduction cells were modified and
retrofitted over the first decade of operation by Kaiser and Norsk Hydro [26].
When potline 4 was commissioned in
1990, the first five CD-type test pots,
jointly developed with Comalco, were
also started at 190 200 kA. Potlines
5 (commissioned from 1996) and 6
(1999) both implemented the so-called
CD20 cells on an industrial scale. In
the Comalco-Dubal nomenclature
the number actually represents the
number of anodes and only roughly
coincides with the amperage level.
So, in UPBN terminology, this was
a CD-21 (210 kA) cell. In 1997, again
five test cells of further advanced amperage were commissioned, called
During 2007/2008, RusAl further

advanced development of a 500 kA
reduction cell. However, it remains
unclear if a prototype has already been
built or if this is yet to happen. There
are plans to build an experimental
RA-500 potline between 2011 and
2014 [54].
RusAl further reports that it is experimenting with inert anode technologies
in two ways: firstly, as a replacement
for prebake carbon anodes in standard
Hall-Hroult cells and secondly, in trial
cells that implement multiple vertical inert anodes and cathodes. The
latter trial cells would have a much
higher time-volume-related output
as compared to standard Hall-Hroult
cells. Specific energy consumption
Figure 6: Dubal DX Pilot Potline, Photo: copyright Dubal
Table 7: Dubal Cell Performance Data as per [67], [69], [70]

Parameter
DX
(DU-35)
DX
(DU-38)
DX+
(DU-44)
Smelter Site
Emal 1 *
Dubal
Dubal, Emal 2
Amperage (design)
340
340
440
kA
380
380
440
kA
Number of Pots
756
40
5 + 444
444 DX+ under commissioning

at Emal 2
95,8
95,5
95
Pot Voltage
4,2 4,22
n.a.
4,24
0,1
< 0,02
< 0,05
AE/(day pot)
13,12
13,04
< 13,4
kWh/kg aluminium
*: Emal 1 values during commissioning phase
Unit

CD26 (CD-28). Further expansions
into potlines 7 (2003) and 9 (2005),
however, deployed the cell type D20
(DU-23) which may be taken as an early
indicator that the Comalco co-operation
had been put on the backburner. In
2006, five independently developed
test cells of DX (DU-34) type have been
commissioned and, in 2008, the technology was semicommercialized in an
in-house 40 pot short potline (potline
8) at Dubal [60], [61]. A photo of this
pilot potline is shown in Figure 6.
Between 1997 and 2004, Dubals interest to expand outside of its UAE Jebel
Ali smelter site was also based on commercialization of the Comalco-Dubal
CD reduction cell technology. Projects
for a proposed 530 ktpy smelter in the
Gulf, using CD20 technology [62], a
500 ktpy smelter in the Bintulu region
of Malaysia [26], and a 520 ktpy smelter
in Qatar based on CD26 technology
[63] did not progress, however, and
Dubal withdrew from the latter project in January 2004 for undisclosed
reasons [64].
The impending split between Comalco
and Dubal became evident in mid 2006.
On 26th June 2006, Comalco and General Holding Corporation (GHC) of Abu
Dhabi signed a heads-of agreement
for a feasibility study of a 550 700
ktpy greenfield smelter at Ruwais/Abu
Dhabi whereas a joint venture between
Dubal and Mubadala Development
Company of Abu Dhabi had, on 28th
June 2006, awarded the feasibility
study for a 700 ktpy greenfield smelter
to be located at Taweelah/Abu Dhabi
[65]. While the ComalcoGHC project
did not materialize, the DubalMubadala project, called Emirates Aluminium
(Emal), has since been constructed
and is fully commissioned since 31st
December 2010 [66].
For the Emal project, Dubal has licensed its own DX (DU-35) technology.
The 2 potline smelter comprises 756
pots and has been started up at 350
kA with key performance indicators
keeping up with those of the experi-
mental pots at Dubal (see Table 7). At

this amperage, Emal will achieve an
output of 740 ktpy.
Meanwhile, Dubal has further challenged its DX cells in potline 8 at the
Jebel Ali smelter, Dubai. Originally
designed for 320 kA, the DX pots have,
since October 2010, been boosted to
380 kA while sustaining the target key
performance indicators (see Table 7).
Dubal anticipate that, with a modified
potshell, the DX cell technologys
operating envelope can be pushed
even further to 400 kA before reaching physical limitations [67]. This improvement has already been transferred to the Emal smelter which
operates, after a rectifier upgrade,
at 380 kA since 2012 boosting the
Phase 1 output to 800 ktpy [71].
In an attempt to avoid such physical
constraints, Dubal has begun developing a new generation of reduction cell
technology called DX+ (DU-44). Five
DX+ test cells, built at Jebel Ali, are
already being tested at 420 kA since
August 2010 and reached 440 kA in
February 2012 [68]; the performance
data achieved so far are included in
Table 7. Dubal is predicting that this new
generation DX+ technology can even
operate at above 440 kA with good key
performance indicators [69].
Emal has meanwhile built an additional
potline implementing DX+ technology.
FHM was achieved on 15.09.2013, and
it is said that the potline current is already set at 440 kA. This single potline
expansion comprises 444 pots, and
as such is the longest globally, with
capacity of 545 ktpy. DX+ technology
may be further deployed if and when
the Alba Line 6 expansion gets the goahead; for now, DX+ technology has
been the basis of a recent feasibility
study evaluating this expansion.
The DX (DU-35) technology based
Emal Phase 1 smelter was built at
8.240 USD/ktpy installed capacity
[71], while the DX+ (DU-44) technology for Emal Phase 2 has cost 7.500
USD/ktpy [72].
4.6 Venalum
In 2000, Corporacin Aluminios de
Venezuela S.A. (CAVSA) presented
a technology website describing its
so-called V-350 aluminium reduction
cell developed by the CVG Venalum
R&D department. The V-350 (VN-35)
cell was designed for operation at 320
350 kA and test pots were said to
be operating at 322 325 kA [73].
Further performance data mentioned
are presented in Table 8.
In 2004, Berrueta presented a paper
at TMS on the planned expansion of
the Venalum smelter by two potlines
with 240 reduction cells each of V-350
(VN-35) type. Groundbreaking was
scheduled for March 2004 and the
first of the two potlines (potline 6) was
planned to be commissioned from
the end of 2006. The additional electric power was to be provided by CVG
Edelca, a state-owned power provider
within the same CVG group, which was
said to have included some generating units in the Caruachi dam project
on the Caroni River to provide for the
Venalum demand. This hydroelectric
power project was supposed to generate energy by 2005.
Berrueta also presented an estimate
of the capital expenditure (CAPEX)
necessary for one potline, including
casthouse, carbon and dock facilities.
The estimated CAPEX of 652 mUSD,
for a production capacity of 220 230
ktpy, would translate into a specific
investment of around 2900 USD per
tonne installed capacity [74].
However, according to Pawleks PASaPoW [26], no further potline construction activities have been recorded at
Venalum, and the V-350 test pots are
supposed to have been taken out of
service and dismantled.
The author therefore believes that Venalum cell technology development
was ceased altogether.
4.7 Chalieco-SAMI
SAMI had initially developed Sderberg
cells (until the mid 1970s) and by the
AWJ 2014 35

Table 8: Venalum V-350 (VN-35) Cell Performance Data as per [73]
Parameter
Value
Unit
Amperage (design)
320 350
kA
322 325
kA
95 96
Pot Voltage
4,1 4,2
0,1 0,2
AE/(day pot)
13,0 13,3
kWh/kg aluminium
late 1970s designed a first prebake

anode cell for 135 kA (SY-14). Development was now focused on prebake
cells and continued stepwise (SY-16,
SY-19/20, SY-23/24, SY-28) until the
turn of the millennium. All these cell
types have been applied commercially
in the Chinese primary aluminium industry. In June 2002, the first full 300
kA potline in China using SY300 (SY30) cell technology was commissioned
after 12 months of potline construction;
the 200 ktpy potline is quoted to have
been built at 1.707 USD per t installed
capacity [75]. By 2007, 13 potlines using SY-300 technology had already
been commissioned in China, including
2nd generation pots operating at 317
kA and 3rd generation pots applying
335 kA - with plans to further increase
potline current to 350 kA. The investment cost was quoted as being 1.200
1.500 USD per t installed capacity
for a standard SY300 potline of 200
ktpy capacity [76]. The 350 kA target
was obviously achieved by 2010 and
SAMI even reported implementation
of 350 400 kA cells and a first 500
kA potline being under construction.
Summarizing from a presentation given
by SAMI in November 2010 [77], SAMI
high amperage ( 300kA) reduction
cell technology had reached more
than 5,5 Mtpy capacity in operation
and over 3 Mtpy under construction,

exclusively located in China. Furthermore, 23 additional operating potlines
utilizing SY190 SY240 reduction cells
were mentioned [78]. A fresh comparison with Pawleks PASaPoW [26]
reveals a lot of uncertainty regarding
the aforesaid as the smelters directly
attributed to using SAMI technology
do not match. The author, therefore,
lumped together explicit SAMI technology smelters with those mentioned in
Pawleks PASaPoW as using Chinese
Technology, clustering the result into
3 amperage categories. The result is
presented in Table 9.
So far, only one SAMI cell technology
application is known outside China
which is the Iralco expansion in Arak/
Iran, consisting of 1 potline with 120
ktpy capacity implementing SY200
cell technology which was commissioned from mid 2007 - 2009. There
is some confusion because Pawlek
stipulates GAMI as technology provider
[26] whereas SAMI itself mentions it
in one of their publications [78]. The
following Table 10 summarizes the
performance data of SAMI reduction
cells as published.
A typical SAMI potline is shown in
Figure 7.
4.8 Chalieco-GAMI:
GAMI seems to have followed a similar
development to SAMI, although there
is very little publication. The GAMI
development was initially based on a
160 kA cell - first commercialized in
the Guangxi smelter in 1994. It appears,
from an evaluation of information
provided by Pawlek in his PASaPoW
directory [26], that GAMI has further developed this technology into the 200
280 kA range. A number of potlines are
reported to have been commissioned
using GAMI reduction cell technology
in this amperage range, though there
is no primary source that confirms this
[26]. In 1998, the GP320 reduction
technology was jointly developed by
GAMI and the Pingguo Aluminium
Company (PGAC). The first GP320 pots
came on stream in October 1999, in a
30 cell trial potline. This trial potline
achieved a current efficiency of 94,4 %
with specific energy consumption of
13,323 kWh/kg Al, operating at up to
325 kA. The cell voltage was reported
as 4,18 V with anode effect frequency
of 0,3 0,4 AE/(d pot) [79]. Further
performance data for GAMI technology
appear to be publicly unavailable.
As there is little GAMI publication,
Pawleks PASaPoW seems to be the
only accessible source of information
about GAMI. However, due to limited
information about Chinese aluminium
smelters in general, this directory also
remains vague on China and some of
the information is contradictory (e.g.
there sometimes appears to be a mismatch between amperage, number
of pots and output) which cannot be
finally resolved by the author. That said,
evaluation of PASaPoW for explicit
Table 9: Overview of Chinese Smelters based on SAMI Cell Technology 300 kA as per [26], SAMI and Chinese Technology lumped
together for better match with [77]
Parameter
SY300
(SY-30)
SY400
(SY-40)
SY500
(SY-50)
SY300
(SY-30)
SY400
(SY-40)
SY500
(SY-50)
Unit
Smelter Status
operatg.
operatg.
operatg.
u/c
u/c
u/c
27
20
PLs
Total Pots
6310
5902
1236
2256
2584
Pots
Average Pots
240*
290
310
275
325
Pots/PL
4,17
4,84
1,71
2,45
3,51
Mtpy
1010
1220
1300
1160
1365
V/PL
*: excluding 1 short potline of 86 pots/70 ktpy considered an outlier


related industrial applications remains
largely unknown. The first potline using
NEUI400 (NE-40) cell technology has
been energized in 2008 at the Zhongfu
smelter. The potline comprises 216
pots with capacity of 240 ktpy, now
operating at 406 kA with pot voltage
of < 3,9 V. Further operating results
have been reported as follows: current
efficiency of 94 %, specific energy
consumption of 12,5 kWh/kg Al and
anode effect frequency of < 0,01 AE/
(d pot) [80], [81].
Figure 7: Typical SAMI SY400 Potline, Photo: copyright SAMI
GAMI high amperage ( 300 kA) reduction cell technology (with one attempt
to mend an obvious mismatch and
back-calculating some missing data
in the directory) has resulted in the
following tentative overview of GAMI
technology proliferation (to be read
with due caution), see Table 11. In addition, > 2,8 Mtpy capacity appear to
be based on GAMI 200+ kA cells which
seem to have been fairly frequently implemented during the past 12 years besides the high amperage technologies.
Outside of China, GAMI has licensed its
technology to some smelter projects.
The GP320 technology has been the
basis of an expansion of the BALCO
smelter, located in the Korba district, India. The project comprises one potline
of 288 GP320 pots with a capacity of
250 ktpy; the potline commissioning

was started in February 2005. GAMI has
also licensed its GP215 technology for
the Press Metal smelter project in the
Sarawak region, Indonesia. The project
consists of one potline with 204 pots
and has a capacity of 120 ktpy; commissioning began in mid 2009. The Det.
Al smelter in Ganja, Azerbaidjan, uses
GP240 cells for its 164 pots potline
with capacity of 100 ktpy.
4.9 NEUI
As a fairly young entity, NEUI also has
little published track record. However in
2010 they reported their latest achievements with the development of highamperage reduction cells. The NEUI
technology portfolio is reported to
cover the ranges 200/240 kA, 300/330
kA and 400 kA, while the extent of its
The NEUI 400 kA technology is currently applied in four potlines, built

between 2008 and 2010 running at
415 460 kA. Six new potlines of 400
kA are said to be under construction
without giving any details. The four
operating smelters are reported to
run with current efficiency around 94
% and specific energy consumption
of less than 12,5 kWh/kg Al [82]. Not
entirely surprising, also in case of NEUI
technology, information provided by
Pawlek in his PASaPoW directory [26]
does not fully match the aforesaid,
see Table 12.
Meanwhile, NEUI has taken their reduction cell development to the 500
kA level. However it is understood that
to date this only covers simulation
and modeling, based on the modeling
tools used for the development of the
NEUI-400 reduction cells [83].
NEUI further reported that they are continuously testing (since March 2008)
pots with so-called novel structural
cathodes (NSC) which consist of cathode blocks with a baffled surface.
Table 10: Summary of SAMI Cell Technology Performance Data ( 300 kA) as per [77], [78]
Parameter
SY300
(SY-30)
SY350
(SY-35)
SY400
(SY-40)
SY500 (u/c)
(SY-50)
Unit
300 335
348 378
400
500
kA
Number of Pots
3074
444
n.a.
288
93,8 95,7
94,15 - 94,5
94,16
94
Pot Voltage
n.a.
n.a.
4,19
< 3,94
Anode Effect Freq. (AEF)
0,1
< 0,3
0,08
0,05
AE/(d pot)
Specific Energy Consumpt.
12,9 0,06
12,8 13,5
13,26
< 12,5
kWh/kg Al
AWJ 2014 37

Table 11: Tentative Overview of Chinese Smelters based on GAMI Cell Technology 300 kA as per [26]
Parameter
GP300
(e.g. GP-30)
GP400
(e.g. GP-40)
GP500
(e.g. GP-50)
GP300
(e.g. GP-30)
GP400
(e.g. GP-40)
GP500
(e.g. GP-50)
Unit
Smelter Status
operatg.
operatg.
operatg.
u/c
u/c
u/c
26
PLs
Total Pots
6322
510
288
840
Pots
Average Pots
280*
260
290
210
Pots/PL
5,75
0,55
0,39
0,76
Mtpy
1175
1100
1220
880
V/PL
*: excluding some obvious test sections and 2 short potlines of 84 pots/138 ktpy
There are different shapes and patterns of protrusions under test, all with
the aim of reducing the metal rotating
velocity and ultimately, although not
explicitly mentioned, lessening the
While for decades Alcoa and the US

had been the unchallenged leaders
in primary aluminium production, their
share of global primary aluminium
production today is 7,9 % and 4 %,
Table 12: Tentative Overview of Chinese Smelters based on NEUI Cell Technology
300 kA as per [26]
Parameter
NEUI300
(NE-30)
NEUI400
(NE-40)
NEUI500
(NE-50)
Unit
Smelter Status
operatg.
operatg.
operatg.
PLs
Total Pots
180
1006
Pots
Average Pots
180
250
Pots/PL
0,15
1,19
Mtpy
Avg. Potline Voltage *
760
1050
V/PL
*: assuming 4,2 V/pot
5 Summary and Conclusion

The author has highlighted the developments since the turn of the new
millennium that have led to a considerable concentration of primary
aluminium production capacity by
way of mergers and acquisitions while
some traditional producers have exited the primary aluminium business.
This trend is equally observed in the
western world as well as in Russia and
China. Along with this development,
the number of companies developing
and licensing reduction cell technology has shrunk in number, specifically
in the western world.
This trend replicates the equally tremendous shift of the primary aluminium production base from west
to east.
respectively, whereas Chalco (8,9 %)

and RusAl (8,8 %) are now the biggest players with almost equal market
share and China is the biggest producing region, representing a massive
45 % of global supply. As an update
to the reduction cell technology providers presented in Table 1, the following Table 13 gives an overview of
the remaining and new players at the
turn of 2013/2014, together with their
It can be concluded that two of the

remaining technology providers,
namely Alcoa and Venalum, appear
to be inactive in the field of reduction
cell technology development and their
technologies would probably not be
available for licensing today. This leaves
reduction cell technology know-how
with only two western companies (RTA
and Hydro Aluminium), one each from
the Middle East and Russia, and three
Chinese research and development
institutions.
From the western companies, RTA has
7.000
6.000
Reduction Cells in Operation
ACD. The test pots are reported to operate at around 3,7 V resulting in 12,0
kWh/kg specific energy consumption
and achieving 93 % CE [84].
achievements in terms of cell amperage and the underlying application

basis, expressed as the number of
operating pots and installed capacity.
The reference to new reduction technology players actually only applies
to Dubal because the other Russian
and Chinese new players in fact have
quite a history though this went largely
unrecognized due to their long-lasting
geopolitical isolation. A graphical representation of pots in operation by
technology provider is given in Figure 8.
5.000
300 - 399 kA
400 - 499 kA
> 500 kA
4.000
3.000
2.000
1.000
Alcoa
Hydro
RTA
Dubal
Venalum
Rusal/VAMI
SAMI*
Figure 8: Reduction Cell Technology Proliferation (operating pots)
GAMI*
NEUI*
*: estimated

Table 13: Reduction Cell Technology Providers as at the Year 2013
Company
Alcoa
Hydro
RTA
Dubal
Venalum
Chalco / Chalieco
Rusal/VAMI
NEUI
SAMI
Cell Type
I / kA
Pots
Capacity / Mtpy
installed
u/c
installed
u/c
none
Remarks
A-817
AA-32
320
448
none
0,388
HAL-230
HAL-26
240 - 258
ca. 410
none
0,285 e
none
Hoyanger, Venalum PL 5, Slovalco
HAL-275
HAL-30
290 - 320
1044
none
0,90
none
SU4, Qatalum
Portland
420
none
negl.
none
test cells in Ardal
300 - 400
5274
1194
5,25
1.22
various smelters, global spread
HAL4e
HAL-42
AP30/40
AP-30/40
AP50
AP-57
570
41
none
0,065
none
St. Jean-de-Maurienne, Jonquire
AP60
AP-60
600
none
none
none
none
460 kpty planned for Jonquire
DX
DU-38
380
796
none
0,85
none
Dubal PL 8, Emal 1
DX+
DU-42
440
449
none
0,55
none
Emal 2, Dubal test cells
V-350
VN-35
325
none
none
none
test cells shut down and dismantled (?)
RA-300
RA-33
330
536
672
0,47
0,59
Tajik, Sayansk, Volgograd
RA-400
RA-44
435
16
672
0,02
0,75
test cells in Sayanogorsk
RA-500
RA-52
520
none
none
test cells (to be) in Sayanogorsk (?)
SY300
SY-30-38
300 - 375
4.752
4,17
SY400
SY-40-45
400 - 450
4.402
2.256
4,84
2,45
SY500
SY-50-52
500 - 520
1.236
2.584
1,71
3,51
CT300/350
CT-30-35
300 - 350
1.558
1,32
400
1.500
1,63
300 - 365
6322
840
5,75
0,76
China
unspec CT400
GAMI
UPBN
CT-40
PASaPoW 05-2013, plus private

communications R. Pawlek
unspecified Chinese Technology per PASaPoW
05-2013 & private communications R. Pawlek
GP300/370
GP-30/37
GP400
GP-40
400
510
0,55
GP500
GP-50
500
288
0,39
NEUI300
NEU-30
306
180
0,15
NEUI400
NEU-46
415 - 460
1006
1,19
6 PLs u/c: ??
NEUI500
NEU-50
500
modeling completed ?
the most widespread global proliferation of their Pechiney-based reduction

cells. RTA also licenses its technology
independently of project ownership.
Their application base of AP3X pots is
impressive and is still growing, however
this is almost dwarfed by the spread
of China-developed reduction cells
which, at comparable amperage, all
occurred during little more than the
last 12 years.
While in the past Hydro Aluminium
and RTA have licensed their technologies on a global scale, the application
of Dubal, RusAl and the Chinese cell
technologies remains, by and large,
confined to their respective homelands, with only a few exceptions.
This is, among other reasons, probably due to some persisting questions
over the long-term performance of
the eastern technology cells. Looking
at the performance presented in the
tables of the previous chapters, this
appears not to be justifiable since the
gap between published eastern and

western performance data seems to
have narrowed. Other issues are the
reported low investment costs and
short construction durations in their
home countries, which raise the question of transferability into an international framework where the specific
Russian or Chinese local conditions
do not apply. Also, the project scope
of facilities and materials/equipment
supply sources behind such figures
often remain unknown - which makes
benchmarking extremely difficult and
potentially misleading. Finally, international aluminium smelter projects
are often based on project finance with
money to be raised on the international
financial markets. International financing institutions would normally apply
a rigid scrutinization of the project,
technically and financially, including
compliance checks with international
standards such as the World Bank Standards or the Equator Principles. These
issues will be much less prominent if
PASaPoW 05-2013, plus private

communications
R. Pawlek, see Chapter 4.8
a project is financed out of equity or

by a government, as may be the case
in the eastern hemisphere.
Coming back to Tabereauxs expectations that future high-amperage
reduction cells would require the solution of technical problems related
to, e.g., wear of cathode lining, heat
balance, emissions, cell instabilities,
stronger magnetic fields and greater
metal losses due to increased cell
turn-around time for relining [2], the
following general trends can be noticed.
The reduction cell technologies have,
over the years, undergone a few creeping changes that are worth mentioning.
Reduction cells are usually developed
for a certain amperage. Cells developed
up to the 1980s were still very delicate
when operating parameters, specifically
the cell amperage, were changed and
small amperage increases frequently
met with major disturbances. It appears
AWJ 2014 39

that cells developed in the 1990s and
later have a much higher tolerance. For
example, this can be concluded from
the AP30 creep to AP39/40. This creep
has happened within a given potshell
with only minor modifications such as
the addition of cooling fins or forced
(air) cooling systems, adjustments
to the anode size and changes in the
potlining. While increasing the amperage (and thus the aluminium output)
by about 30 %, the performance figures could be maintained despite the
stronger magnetic fields with their
potentially detrimental effect on cell
performance. This can be taken as evidence that the modeling tools applied
during reduction cell development
have reached a level of sophistication
that the behaviour of reduction cells
under modified operating parameters
can be simulated quite precisely. The
same conclusion is applicable for the
development of new reduction cells which seems to happen in ever shorter
periods of time with less time required
for pilot cell trials.
The creep previously mentioned comes
along with an increased energy input
to the cell, which up to now is just
dissipated as heat into the environment. The crucial issue of the presence (or absence) of a frozen layer of
bath (ledge) to protect the sidewall
lining is largely left to the success of
ventilation enforced by natural draft.
Thus, the aluminium reduction cell
can be regarded as rare sample of
a metallurgical furnace (in German:
Elektrolyse-Ofen) operating at just
short of 1.000 C that is left to natural
forces for the successful furnace cooling as opposed to forced cooling (using media other than air) which would
even allow the partial recovery of the
dissipated energy.
An achievement of modern PFPB cells
that usually goes unnoticed outside the
industry is the lowering of the anode
effect frequency (AEF) and duration
which is due to ever smarter, more reliable alumina feeding and pot control
systems. Obviously, this lowers the
amount of energy wasted during anode
effects and disruption of the electrolytic

aluminium deposition as well as the
thermal cell disturbance related to
the additional heat input. Generation
of the powerful greenhouse gases CF4
and C2F6 (PFCs) is directly related to
the occurrence and duration of anode
effects and the industry as a whole
has achieved remarkable reductions
in the co-production of PFCs. Modern
PFPB cells can almost entirely suppress anode effects which in the old
days (which started ending maybe
only 20 years ago) were considered
an indispensable means for good cell
operation. Older technologies (SWPB,
CWPB, Sderberg) have inherent difficulties in supporting such achievements because the feeding systems
do not allow the elimination of anode
effects, which means that smelters
operating such reduction cells will
be under increasing environmental
pressure in the future.
Modern high-amperage reduction
cells would normally use highly or
fully graphitized cathode blocks which
are softer than blocks made from anthracite with low graphite content. As
a consequence, one would expect
shorter cathode life for cells with
graphitized cathodes. Nevertheless
it appears that the cathode life is not
becoming shorter but indeed longer
with a tendency to achieve 2000 days
or more on average. Connected to
this there should also be no negative
effect on productivity (or rather cell
availability) from a relining point of
view. If cells are not lined in-situ but
in a dedicated pot delining/relining
facility, turnaround time - and hence
loss of metal production - can be kept
low and, as a positive side effect, the
working environment for the delining/relining activities can be better
controlled.
As a summary to the above, todays
computer-based modeling tools (for
modeling electro-thermal, electromechanical, magnetic, and magnetohydrodynamic effects) used during
reduction cell design are very capable
of predicting the behaviour of such
cells. Modern cells designed with such

tools have high probability of achieving low energy consumption and, at
the same time, high current efficiency
because detrimental effects of strong
currents and the associated magnetic
field can be compensated during design. This results in the required robust
cell design necessary for stable cell
operation. Sophisticated cell control
algorithms (including fuzzy logic) and
development of more robust sensors
support operation to achieve favourable
operating parameters including low
AEF. These core reduction cell issues
seem to be fairly well controllable
even when current intensities well in
excess of 300 kA are applied.
From a potline construction point of
view, high-amperage cells then require
more focus on some rather profane peripheral issues which have been much
less important on lower amperage
levels. Aluminium reduction cells have,
despite some higher anodic/cathodic
current densities, almost exclusively
grown in one direction with increasing
amperage which is the length of the
potshell; potshell width and depth
have almost remained unchanged over
the past 2 3 decades. Construction
elements affected by a longer potshell are the potroom building width;
wider crane span combined with heavier
lifting loads; consequentially heavier
foundations and structural steel elements; wider superstructures that
need to support more own weight plus
more weight from additional anodes
without too much sagging, just to name
a few.
As has been indicated in the previous
chapters on individual reduction cell
designs, potlines have grown in number
of pots installed per line. Specifically
western technologies are now implementing potlines with 350 400 pots
per line whereas eastern technologies
seem to only follow this trend more
hesitantly (usual pots per potline are
still around 200 300). While the trend
towards longer potlines saves on investment cost for common equipment
like the rectifier-transformers (RTs) for

each potline it is also more demanding
in terms of voltage level for such units.
Recently built long potlines based on
Dubal, Hydro and RTA technology now
require RTs of 1.700 50 V while this
level for eastern technologies is still in
the 1200 V range (and below). The RT
manufacturers are offering 2.000 V RTs
by now, and the Emal 2 potline with 444
pots is the first where this RT technology is practically applied. The 2 kV RT,
however, has two major implications.
Firstly, potlines would be able to grow
into the 450 pots per potline size which
would support 500 ktpy capacity out of
just one potline (assumed @ 400 kA).
In case of a power failure (generation,
transmission, rectification), the entirety
of this production capacity would be
at risk and it remains to be seen how
venturesome the aluminium industry
and its financiers will be. Secondly,
the 2 kV RT creates higher demands
for the electrical insulation of potline
buildings and the insulation strength
of all affected electrical equipment
inside those buildings (motors, cables,
PLCs etc.).
Higher metal output per cell also
requires more metal handling and
more anode carbon to be replaced. An
increase of pot interventions (for tapping or anode changing) or an increase
of potroom and plant traffic would
not be welcome by operators so new
concepts might have to be devised,
e.g. for tapping, there are limitations
to increasing the height of the crucible
because of the metallostatic pressure
that needs to be overcome.
6 Alternatives and Outlook

The Hall-Hroult (HH) cell is a comparatively inefficient metallurgical
reactor, even at the 600 kA level: one
600 kA cell produces only 4,5 t per
day or 1.670 tpy, but occupies an area
of some 4 m x 18 m = 72 m2. For a
commercially viable plant, between
200 and 450 of these cells have to
be installed which causes quite a big
land usage. In comparison: one single
modern iron blast furnace (BF) unit of
10 m hearth diameter, or a footprint
of 78 m2, produces 7.260 t per day
or 2.650.000 tpy (2,65 Mtpy!), and

there are bigger ones also. HH cells
operate at 950 C or ca. 300 C above
the melting point of pure aluminium
and usually yield aluminium of 99,7 %
purity or better, whereas the iron BF
operates at a tapping temperature of
ca. 1.450 C which is actually some
90 C below the melting point of pure
iron (this is possible because the iron
BF does not produce pure iron but an
iron-carbon alloy with about 4 % C plus
some other metallic impurities which
together lower the melting point of
the impure, so-called pig iron, which
needs to be purified in another step).
So who is to blame for this disparity? In brief, aluminium production
is hampered by its very own natural
properties, first and foremost its very
strong chemical bond to oxygen,
but also its trivalence, its low atomic
weight and density, plus the fact that
common reducing agents like carbon
do not work quite well in aluminium
metallurgy. These facts required that
the harshest of all possible reducing
agents had to be deployed to break
the aluminium-oxygen bond: pure
electrons in the form of electric current in an electrolytic process, and
its industrial manifestation is the HH
cell.
Up until now, all investigated or discussed alternatives struggle with
some form of fundamental problem
related to the strong oxygen-affinity
of aluminium and its position in the
electrochemical series. Lets look at
some of the potential alternatives:
Electrolysis
Near ambient temperature and/or water based electrolytic process (maybe
similar to copper): aluminium hardly
dissolves in aqueous solutions, and
if, its position in the electrochemical
series favours reduction (= production)
of other components of such systems
instead of aluminium.
Electrolytic process closer to
aluminiums melting point of 660 C,
say at around 700 C (instead of 950
C): no suitable electrolyte has been

found that would combine alumina
solubility, electric conductivity etc.
like fluoride based cryolite. There are
chloride based alternatives which have
their own disadvantages, see below.
Inert anodes: Materials investigated
so far are not exactly inert yet, causing
co-deposition of more noble metals
affecting impurity levels. Widespread
industrial application of inert anode
materials always seems - at any given
time - to be 15 20 years away. In principle, such anodes would just release
oxygen, no carbon would be consumed
and no CO/CO2 generated, no pot
intervention for anode change, and no
anode plant would be required. However, the change of the sum reaction
would require some 0,5 1 V higher
pot voltage compared to HH cells.
Wettable, drained cathode: would
reduce magneto-hydrodynamic (MHD)
effects and would help avoid the deep
pool of liquid aluminium in each cell
(15 25 cm of dead inventory) required
to control the MHD effects; industrial
implementation, similar to inert anodes, always seems far away. A new
development from China comprises
the mentioned NSC baffled cathode
blocks which might help in this context
though these NSC cathodes are made
of conventional carbon material and
therefore are not to be confused with
wettable, drained cathodes.
Novel Processes
Carbothermal Reduction (preferably
similar to the iron blast furnace
process): the quick answer is that
thermodynamically, carbon can only
reduce alumina above 2000 C, and
that the product is an unwanted and
unstable Al4C3 instead of aluminium.
However, the issue is much more
complicated and can be summarized as
follows: up to 2.160 C, only an aluminacarbide slag (AlO, AlO2, AlC) but no
liquid aluminium metal is produced;
above 2.160 C there is also a second
phase: something like liquid metal
but in fact it is an aluminium-carbon
alloy with some 10 % C; at the required
AWJ 2014 41

process temperature of > 2.200 C,
additionally there will be considerable
evaporation of aluminium in the form of
aluminium and aluminium sub-oxide
(Al2O) vapors which can amount to up
to 25 % of the produced metal and may
require a vacuum process stage [85].
Carbothermal reduction is said to be
further investigated by an Alcoa-Elkem
JV which designed process reactors
called Advanced Reactor Process
(ARP) and Vapor Recovery Reactor
(VRR). While Alcoa-Elkem estimate
that Capex and Opex of a commercial
ARP + VRR plant would be 30 % less
than an equal-sized HH smelter, the
authors of [85] also point out that major
issues of the carbothermal process
still remain unsolved.
In the personal opinion of the author
of this review, it appears to be very unlikely that the 50 Mtpy primary smelter
community can be convinced to switch
from the established HH process (operating at a manageable 950 C, directly
yielding a single, high-quality product
without fritting) to a novel carbothermal
process that requires extremely high
temperatures (> 2.200 C), and scatters
the desired aluminium across 3 product
streams, whereby those products are
far from being pure aluminium hence
requiring some sort of post-treatment
for metal purification.
Alcoa Smelting Process (ASP): it may
be argued that this is not a novel process as it has been developed around
1970 already, that the pilot facility was
shut down in 1985, and that it should
have been mentioned under Electrolysis above. Nevertheless, the author of
this review wants to point out some
of the ASPs conceptual advantages
which may justify its categorization as
novel process. Published data on the
ASP are scarce, so the below rests on
summary information published in [86]
and [87], from which also the ASP cell
(or rather electrolyser) sketch is taken,
see Figure 9.
Cl3, dissolved in a chloride electrolyte

(e.g. NaCl + LiCl), operating at 700
30 C. No PFCs would be generated/
emitted.
HH cells consist of one single, horizontal anode-cathode pair (also PB
cells where the anode may consist of
some 40 individual anode blocks). In
contrast, the ASP electrolyser comprises 12 horizontal bipolar anodecathode pairs stacked above each other,
i.e. the underside of each electrode
acts as anode whereas the upper side
acts as cathode (see Fig.9). In this way,
one ASP electrolyser replaces 12 HH
cells.
The 1980s ASP electrolyser is quoted
to have produced in excess of 13 tpd;
crunching the few available numbers,
back-calculation reveals that the amperage would have been 140 kA (which
fits very well with the rectifier capabilities of the early 1970s) resulting in
13,2 tpd or 4.820 tpy per single ASP
electrolyser (at an assumed CE of
98 %), which is 3 times the output of
one AP60 cell.
No dimensions for an ASP electrolyser
are available but, based on a number for current density, the author of
this review dares an educated guess:
the size of one ASP electrolyser was
probably about 3 m x 3 m (W x L) with

height between 2 and 3 m. This means,
one single ASP electrolyser produces
1,47 t/(m2 day) which is 2350 % of
one AP60 cell, despite the ASP only
requiring 140 kA or 23,3 % of the AP60
current intensity.
The bipolar electrodes are inert, i.e.
no electrode changes would occur
(probably, after some time the ASP
electrodes would also deteriorate to
some extent and would have to be
replaced about once every 3 years,
which is more similar to cathode relining in HH cells). An ASP smelter would
therefore not require any anode plant
and no CO/CO2 would be generated/
emitted.
The interpolar distance (ACD) is quoted
as 1 2 cm (HH: ca. 5 cm), and the
cathode surface only carries a thin film
of aluminium which is swept off by the
anodic chlorine gas-induced electrolyte circulation. As a consequence,
MHD disturbances in ASP electrolysers
would be minimal which would most
certainly allow placing electrolysers
much closer to each other in an industrial ASP smelter (recap: HH pot rows
are spaced some 30 90 m apart). A
higher CE of (assumed) 98 % seems
also justifiable for ASP electrolysers
due to lower MHD impact.
Feed port
Upcomer
Terminal
anode
Downcomer
Bipolar
Terminal
cathode
Anode - cathode
space
Aluminium
sump
The ASP consists of electrolysing AlFigure 9: Sketch of Alcoa Smelting Process Electrolyser [87]

Due to use of a chloride electrolyte,
single-cell voltage (i.e. one anodecathode pair) is 2,7 V, so one electrolyser would require some 32 V, plus
external busbar losses. ASPs specific
power consumption is quoted as < 9,5
kWh/kg.
Despite the clear advantages that such
an ASP based smelter would present,
the author does not want to hide that
there are obviously major challenges
that lead Alcoa to stop this development 30 years ago. One area of concern
is the preparation of the intermediate AlCl3 which is a nasty chemical
compound as it is highly corrosive,
volatile, hygroscopic (attracts water)
and hydrolyses (e.g. with humid air)
forming aluminium oxi-chlorides. Both
water and oxygen impair the electrolytic
process and must be kept at minimum
levels. There would also be formation
of chlorinated biphenyl which needs
to be removed in order to avoid pollution issues.
However, the author of this review believes that it may be worthwhile revisiting the underlying ASP concept
again, given some 30 years of scientific
progress in material science, process
engineering, and computational simulation tools.
In order to demonstrate the potential
behind the ASP concept lets make the
following thought experiment: Assume
it is possible to stack 20 bipolar anodecathode pairs (instead of 12) in one
ASP electrolyser. Additionally you put
2 of these stacks into one electrolyser
so one electrolyser would now have
dimensions of 3 x 6 x 3,5 m3 (W x L x
H), and the increased electrode area
would permit 300 kA at only incrementally higher current density. Imagine a
smelter with 100 ASP electrolyser units,
and the capacity would be a whopping
1.720.000 tpy!
Looking forward, it looks as though
the Hall-Hroult process at its 128th
anniversary is here to stay for quite
some time. Alternative processes

are not close to any kind of industrial
implementation.
Concluding from this it means that
the primary aluminium industry would
continue to improve the Hall-Hroult
process gradually. How far the increase
in potline amperage, one-dimensional
growth of cell length and ever longer potlines can be sustained before
some kind of optimum configuration
is reached remains to be seen.
Aluminium demand growth seems to
continue in line with global urbanization and population growth, and recycling efforts should be reinforced as
much as possible. However, recycled
material will only continue to complement primary aluminium production
which will likewise (have to) continue
to be the major source of any future
aluminium supplies.
7 Acknowledgements
Special thanks go to my former colleagues, Dr. Robert Minto and Thiago
Heitling, who helped establishing the
first edition. For this second edition,
I would like to thank Anne Tappen
for her friendly support and Rudolf
Pawlek for his private communications regarding latest intelligence on
Chinese smelters.
[5] MATTHEWS, R., Alcoa Completes

Acquisition of Reynolds Metals, The Wall
Street Journal, New York 04.05.2000
[6] New Alcan hungry for more acquisitions:
Earnings surge, National Post, Montreal
17.10.2000
[7] Metals Giant files for bankruptcy, BBC
News, 12.02.2002, http://news.bbc.co.uk/2/
hi/business/1816870.stmv
[8] Kaiser restarts Gramercy alumina refinery,
Metal Bulletin, 18.12.2000, 5
[9] Can Jack Hockema make Kaiser fly?,
American Metal Market, 13.03.2006
[10] Norsk Hydro acquires VAW, Metal Bulletin
07.01.2002
[11] Regulation (EEC) No. 4064/89,
Merger Procedure, ALCAN/PECHINEY (II)
EEComm. 53, 29.09.2003, http://www.
worldlii.org/eu/cases/ECComm/2003/53.
html#Heading5#Heading5
[12] Alcoa to offer to acquire Alcan for
US$73.25 per share in cash and stock,
Business Wire 07.05.2007
[13] Recommendation to Reject the Alcoa
Offer, Excerpts from May 22, 2007 Directors
Circular, Alcan Presentation, downloaded from
Alcan website (website no longer exists)
[14] Hydro will pay $4.9 billion for
Vales aluminium assets, Metal Bulletin
10.05.2010,5
[1] HEIL, J., Minto, R., Heitling, T., Aluminium

Reduction Cell Technology Providers
an Updated Review, Proceedings of the
European Metallurgical Conference Emc
2011, Volume 3, pp 793 824, Dsseldorf,
Germany, 2011
[15] Vale concludes sale of Al, alumina and

bauxite assets to Hydro, Metal Bulletin
28.02.2011
[16] BILLINGHURST, D. et al.,Development of
B32 Cell Technology, in: Light Metals 2006,
Proceedings of the Annual TMS Aluminium
Committee Meeting, GALLOWAY, T. J. (Editor),
A Publication of TMS, Warrendale 2006,
255 257
[2] TABEREAUX, A. (2000): Prebake Cell

Technology: A Global Review, Journal of
Metals, February 2000, 23 29
[17] DROZDOV, A., Aluminium: The Thirteenth

Element Encyclopedia, The RUSAL Library,
Moscow 2007, 81
[3]SAGAFO,O,Eine norwegische Erfolgsgeschichte Hydro 1905 2005, Pax Forlag,

Oslo 2005
[18] History of the Russian Aluminium

Industry, http://rusal.ru/en/aluminium/
history.aspx
[4] Data for LME price chart extracted

from:
http://www.indexmundi.com/commodities/
?commodity=aluminum&months=360
[19] RusAl Company History, http://rusal.

ru/en/about/history.aspx
References
AWJ 2014 43

[20] Russia sees Birth of Aluminum Giant,
BusinessWeek online, 03.10.2006, http://
www.businessweek.com/globalbiz/
content/oct2006/gb20061003_404723.
htm
[21] ZHONGXIU, L., China races up to keep
up with demand, Metal Bulletin Monthly,
March 2003, 22 - 25
[22] Chinas primary aluminium production
1999 to present, International Aluminium
Institute, downloaded on 18.10.2010 from
https://stats.worldaluminium.org/iai/stats_
new/historical.asp?alt CurrentYear=2010&alt
Form=11&altDataTyp
[23] Chinas primary aluminium production
2006 2013, International Aluminium
Institute, downloaded on 12.03.2014 from
http://www.world-aluminium.org/statistics/
primaryaluminium production/#data
[24] Aluminium Producer Ranking, http://
www.rusal.ru/en/aluminium/manufacturers.
aspx
[25] LILI, S., China faces decisions over
surplus capacity, Aluminium International
Today, July/ August 2010, 15
[26] PAWLEK, R. (Editor), Primary Aluminium
Smelters and Producers of the World
(PASaPoW), Beuth Verlag Gmbh, Berlin,
Update 05/2013
[27] Saudi Arabian Mining Company pursues
integration vision, Middle East Economic
Digest, 09.09.2009, http://www.meed.
com/supplements/2009/gcc-aluminium/
saudi-arabian-mining-company- pursuesintegration-vision/3000571.article
[28] Capacity Increase Official Opening
of New Sunndal, Hydro Aluminium Press
Release 03.12.2002, downloaded 27.09.2007,
http://www.hydro.com/en/press_room/
press_releases/archive/2002_12/1799/1799_
en.html
[29] New Sunndal Plant Officially Completed,
Hydro Aluminium Press Release 04.11.2004,
downloaded 27.09.2007, http://www.hydro.
com/en/press_room/news/archive/2004_11/
sunndal_opening_en.html
[30] Qatalum Primary Aluminium Plant,
Qatalum publication, http://www.qatalum.
com/en/Technology-and-solutions/Primaryaluminium-plant/
[32] Qatalum Rectifiers and Transformers,

Qatalum publication, http://www.qatalum.
com/en/Project-development/Procurement/
Contractors/Rectifiers-and-Transformers/
[33] Record amperage at rdal, Hydro
Aluminium Press Release 11.06.2008, http://
www.hydro.com templates/NewsArticlema
spx?id=20934&epslanguage=EN
[34] Pilot Aluminium Technology Plant at
Karmy under Study, Hydro Press Release
07.03.2013, http://www.hydro.com/en/Pressroom/News/Archive/2013/Pilot-aluminiumtechnology-plant-at-Karmoy-under-study/
[35] BRANDZG, S., Capital Markets Day
2013, Hydro Aluminium Presentation,
05.12.2013,
http://www.hydro.com/en/Investor-relations/
Presentations/Capital-Markets-Days/
[36] BARDAL, A. et al., HAL4e Hydros
New Generation Cell Technology, in: Light
Metals 2009, Proceedings of the Annual TMS
Aluminium Committee Meeting, BEARNE,
G. (Editor), A Publication of TMS, Warrendale
2009, 371 375
[37] SEGATZ, M., Designing the Next Smelter
Generation, Presentation at 24th Metal
Bulletin International Aluminium Conference
2009, Dsseldorf, 14. 16.09.2009
[38] Qatalum to lose $585m if smelter mains
out of action, MEED 30.08.2010
[39] Pechiney unveils its new AP50 aluminium smelting technology, Metal Bulletin
13.07.2000
[44] MARTIN, O. et al., AP30 toward 400 kA,

in: Light Metals 2009, Proceedings of the
Annual TMS Aluminium Committee Meeting,
BEARNE, G. (Editor), A Publication of TMS,
Warrendale 2009, 445 450
[45] AP30/AP40 The worlds benchmark
reduction technologies, RioTintoAlcan
Brochure, distributed during TMS 2011
conference
[46] MARTIN, O. et al., The Next Step to the
AP3X-HALE Technology: Higher Amperage,
Lower Energy and Economical Performances,
GALLOWAY, T. (Editor), A Publication of TMS,
[47] BENKAHLA, B. et al., Last Development in
AP50 Cell, in: Light Metals 2008, Proceedings
of the Annual TMS Aluminium Committee
Meeting, DEYOUNG, D. (Editor), A Publication
of TMS, Warrendale 2008, 275 280
[48] Sohar Aluminium starts production,
Gulf News 21.09.2011
[49] Rusal completes smelter upgrades,
Aluminium International Today January/
February 2011, 25
[50] MANN, V., Aluminium Technology
& Production in Russia, Presentation at
the Annual TMS Aluminium Committee
Meeting (?), 09.02.2006
[51] Innovative Aluminium, Metal Bulletin
Monthly August 2007, 11 - 14
[40] Rio said to cancel $2.6 Billion Africa

Aluminium Plant, Bloomberg 15.10.2009,
download from http://www.bloomberg.
com/apps/news?pid=newsarchive&sid=a
8D7Lhpxr97c
[52] VESELKOV, V. et al., Intensification of

the Process in 300 kA Pre-Baked Anode
Cells, in: Light Metals 2007, Proceedings
Meeting, SRLIE, M. (Editor), A Publication
[41] Alcan pledges $1B in 2011 to revisit

long-standing expansion projects, Metal
Bulletin
15.12.2010
[53] MANN, V., Aluminium Industry: Paving

The Way to a Safer World, Presentation at the
10.03.2008
[42] RioTintoAlcan inaugurates its leadingedge AP60 aluminium smelter in Canada,

RTA Press Release 16.01.2014, http://
www.riotinto.com/media/media-releases237_9743.aspx
[54] Development of Cutting-Edge Reduction

Processes, RusAl information downloaded
from http://rusal.ru/en/own_technologies.
aspx, 22.02.2011
[43] AP60 A breakthrough in unsurpassed

productivity, RioTintoAlcan Brochure,
distributed during TMS 2011 conference
[31] AlTogether November 2013, Qatalum

magazine, https://www.qatalum.com/Media/
Publications/Pages/Magazines.aspx
[55] SKOMYAKOV, V. et al., SUAL 300 kA

Pre-Baked Cells, in: Light Metals 2006,
Committee Meeting, GALLOWAY, T. (Editor),
A Publication of TMS, Warrendale 2006,
307 311

[56] UC RusAl 2009 Annual Results, RusAl
Presentation 12.04.2010, downloaded from
http://rusal.ru/en/iv_presentations.aspx ,
22.02.2011
[57] Taishet Smelter Project Fact Sheet,
http://rusal.ru/en/about/invest/taishet.
aspx
[58] BEMO Project Fact Sheet, http://rusal.
ru/en/about/invest/bemo_factory.aspx
[59] AL FARSI, Y., CD20 Reduction Cell
Upgrade for Dubals Expansion Project,
in: Light Metal 2005, Proceedings of the
KVANDE, H. (Editor), A Publication of TMS,
[60] 30 Years of Excellence Dubal 1979
2009, Dubal Anniversary Brochure, Dubai
2009, download from http://www.dubal.ae/
our-innovations/development-and-transfer.
aspx
[61] Dubal Development and Transfer of
Technology, download from http://www.
dubal.ae/ourinnovations/development-andtransfer.aspx
[62] Dubal seeks partner for new Gulf aluminum smelter, Platts Metals Week Vol. 68,
No. 46, 17.11.1997
[63] Qatar: Joint venture construction plans for
proposed $2,000,000,000 build-own-operate
(BOO) primary aluminium smelter, Dubai
Aluminium Co. Ltd (Dubal) [UAE] & United
Development Co. (UDC) [Qatar], Worldwide
Projects Report on Engineering Construction
& Operations in the Developing World, Vol.
12, Issue 06, 01.06.2003
[64] Dubal exits Ras Laffan smelter, MEED
23.01.2004
[65] GHC, Comalco tie up for smelter, MEED
30.06.2006
[66] Emal hits the smelters full production
on first workday of 2011, Emal Press release
14.01.2011, download from http://www.emal.
ae/en/Default.aspx?view=NewsDetail&id=5
9&itemID=21
[67] ZAROUNI, A., DX Reduction Cell
Technology, Presentation at 25th Metal
Bulletin International Aluminium Conference
2010, Bahrain, 20. 22.09.2010
[69] Dubal DX Technology improvements,

Aluminium International Today January/
February 2011, 52
[70] REVERDY, M., Recent development of
Dubal aluminium reduction cell technologies,
Aluminium 1-2/2014, 18 - 22
[71] EMALs official inauguration 18 April
2011, Metal Bulletin Special
[72] EAG to expand globally, Khaleej Times
16.09.2013
[73] The V-350 Cell, downloaded on
05.09.2000 from Corporacin Aluminios
de Venezuela S.A. (CAVSA) http://www.
aluminio.com.ve/english/av350_en.htm
(page unavailable today)
[74] BERRUETA, L., CVG Venalum Lines VI
and VII, in: Light Metals 2004, Proceedings of
the Annual TMS Aluminium Committee Meeting, TABEREAUX, A. (Editor), A Publication
[75] HAIBO, S. et al., Henan HongKong
Longquan Aluminium Co. Ltd., China Growing Up, in: Light Metals 2004, Proceedings
Meeting, TABEREAUX, A. (Editor), A Publication of TMS, Warrendale 2004, 233 236
[76] JIAMING, Z. et al., The Continuous Development of SAMIs SY300 Technology,
Annual TMS Aluminium Committee Meeting, DEYOUNG, D (Editor), A Publication of
TMS, Warrendale 2008, 275 280
[77] DONGFANG, Z., Aluminium Reduction
Cell: New energy saving technology Developments & Applications, Presentation held
at Metal Events 7th International Aluminium
Conference, Abu Dhabi, 30.11. 01.12.2010
[78] XIANDONG, Y. et al., The Pot Technology Development in China, in: Light Metals
2010, Proceedings of the Annual TMS Aluminium Committee Meeting, JOHNSON, J.
(Editor), A Publication of TMS, Warrendale
2010, 349 354
[79] ENSHENG, Y. et al., Developing the
GP-320 Cell Technology in China, in: Light
Aluminium Committee Meeting, ANJIER, J.
2001, 213 218
[80] XIQUAN, Q. et al., Successful commercial operations of NEUI400 Potline, in: Light
Aluminium Committee Meeting, JOHNSON,
J. (Editor), A Publication of TMS, Warrendale
2010, 359 363
[81] BAN, Y. et al., Baking Start-up and Operation Practices of 400 kA Prebaked Anode
Pots, in: Light Metals 2010, Proceedings of the
JOHNSON, J. (Editor), A Publication of TMS,
[82] DINGXIONG, L. et al., New Progress
on Application of NEUI400kA Family High
Energy Efficiency Aluminum Reduction Pot
(HEEP) Technology, in: Light Metals 2011,
Committee Meeting, LINDSAY, S. (Editor), A
Publication of TMS, Warrendale 2011, 443
448
[83] DINGXIONG, L. et al., Development of
NEUI500kA Family High Energy Efficiency
Aluminum Reduction Pot (HEEP )
Technology, in: Light Metals 2011, Proceedings
Meeting, LINDSAY, S. (Editor), A Publication
[84] JIANPING, P. et al., Development and
Application of an Energy Saving Technology
for Aluminium Reduction Cells, in: Light
Aluminium Committee Meeting, LINDSAY, S.
2011, 1023 1027
[85] BALOMENOS, E. et al., Theoretical
Investigation of the Volatilization Phenomena
Occurring in the Carbothermic Reduction of
Alumina, Erzmetall 64 (2011), 6, 312 320
[86] Grjotheim, K. et al., Aluminium
Electrolysis, 2nd Edition, Aluminium-Verlag,
Dsseldorf 1982, 17 21
[87] Thonstad, J. et al., Aluminium Electrolysis,
3rd Edition, Aluminium-Verlag, Dsseldorf
2001, 340 341
Author: Dr.-Ing. Joachim Heil
Contact Information:
jogaheil@t-online.de
[68] Advancements in reduction technology

improve specific energy consumption and
reduce greenhouse gas emissions at Dubal,
Aluminium 3/2013, 34 - 36
AWJ 2014 45
START-UP OF ARVIDA SMELTER, AP60 TECHNOLOGICAL CENTER.

well as other physical constraints, the
target intensity of the prototype was set
at 550 kA. Table 1 shows performance
results achieved with the cell over a 6
month period.
Key indicator
Results
Intensity (kA)
550
Current efficiency (%)
94.2
SEC (kWh/t)
13260
Table 1. Performances of AP60 prototype

cell at LRF over 6 months
The Arvida Smelter AP60 Technological Center
Fig 1: Arvida Smelter, AP60 Technological Center.
Introduction
A new milestone for reduction technology has been recently reached with
the successful start-up of the Arvida
Smelter, AP60 Technological Center
in Jonquire, Quebec, Canada. Since
December 2013, the 38 first generation AP60 cells deliver an annual production capacity of 60,000 tons of
aluminium.
Starting from the development of
the first prototype AP60 cells, this
article presents the start-up of the new
smelter in 2013 and gives an overview
of the future Rio Tinto Alcan reduction
technology developments to come.
Development of AP60 at the LRF

The development of AP60 technology
started 4 years ago at the Rio Tinto
Alcan Laboratoire de Recherche
des Fabrications (LRF) installation
in Saint-Jean de Maurienne, France.
Reducing Full Economic Cost (FEC)
using new knowledge developed in
Rio Tinto Alcan R&D centers was the
key driver that pulled up the development of AP60 over formally proposed
AP50 technology.
Key Development steps of AP60

technology at prototype level
AP60 was designed using in-house
Rio Tinto Alcan specialized thermoelectrical, MHD and thermo mechanical
models. After the design phase, it was
decided to build one cell at the St-Jean
de Maurienne LRF facility in France in
order to test the performance of the
technology at a prototype level while
at the same time starting the construction of the new Arvida Smelter, AP60
Technological Center in Saguenay.
The AP60 prototype cell was started in
December 2011. Taking advantage of
the numerous AP30 reduction potlines
and cells, a large number of the technology bricks used in the cell design
had already been optimized and tested
on AP30 technology. For example, the
design of the low resistance cathode
has been tested on hundreds of AP30
cells. This capability to test technology
bricks on several cell technologies
(AP18, AP30, and P155) has sped up
and strengthened the AP60 cell development.
Due to substation limitations at LRF as
While the prototype cell at the LRF

did operate at 550 kA, the objective
of the Arvida smelter AP60 Technological Center is to operate the AP60
technology up to 600 kA. Some of the
differences between the prototype cell
and the cells at the Arvida Smelter
AP60 Technological Center include;
anode height, cover material composition, magnetic fields and exhaust gas
flow rate. All these differences were
analyzed and taken into account in the
industrial design of the cell, and in the
projected plant performance.
The notice to proceed for construction
of the 38 cell demonstration plant
was awarded in December 2010. The
substation was built in order to sustain a future expansion to a full line
smelter.
Technology Transfer Process

The rigorous Rio Tinto Alcan process
of technology package transfer was a
key element of the strategy to build the
Arvida Smelter AP60, Technological
Center (Figure 1). This methodology
ensures that all engineering phases are
done with carefully controlled information. Emphasis was put on quality
control by technology experts for the
most important aspects of the new
technology. The plant was designed,
engineered and built without having
to go back to the drawing board and
without any constructional issues being
encountered.
specialized thermocouples on the

cathode surface and other measurement techniques.
Out of the 38 cells, 11 were started using a dry start-up methodology. This
way of starting a cell is important as it
provides a method for starting a greenfield smelter without having liquid
bath at hand.
For all 38 cells, an extensive measurement campaign was performed to
provide data on mechanical behavior
of the shell, superstructure and busbar
at start-up and during the stabilization
phase. Among the measurements done
were the dynamic deformation of the
shell and superstructure.
Fig 2: AP60
Given that the new plant will be the

platform for future developments of
the AP60 technology, strategic R&D
equipment was integrated into the
package at the construction phase. A
booster section, complete with exhaustive instrumentation for cell monitoring, will be used by R&D teams with
expertise in modeling, measurements,
cell design, and operation to develop
the full potential of the technology.
The commissioning of the
technology
At the end of the construction phase,
the plant was turned over to the operations team who began a series of tests
to commission the new equipment.
Among these tests were those more
specifically related to the ability of
busbars to handle 600 kA of current
intensity. Measurements of temperature, displacement of the busbars and
short circuit equipment testing were
done at different current intensities
allowing development of an in-depth
knowledge of this critical subject.
The short circuit tests were followed
by magnetic field testing of potroom
equipment. The level of amperage
of AP60 technology leads to the
development of very high magnetic
fields (greater than 600 gauss). It was
identified early on as a technology
risk that the magnetic field may cause
malfunctions of potroom equipment.

Well before the start-up phase, many
components were tested in high magnetic fields and mitigation plans were
put in place. Even with all this early
testing, it was still found that some
equipment had to be modified. However, no significant outstanding issues
were found at this late stage confirming the efficiency of the rigorous gate
process used along the implementation of the project.
During this time, more general potroom equipment testing was also being performed. This phase allowed
operators to be trained and get used
to the new equipment and operating
procedures. This also allowed operating people to interact with internal
technology experts dispatched on the
site to increase local knowledge of the
new AP60 technology.
Cells start-up
The 38 cells of the smelter were started
in Q4 2013. The first two cells were
started at 550 kA while the remaining
36 were started at 560 kA. Each AP60
cell required the addition of up to 14 t
of liquid bath.
Before adding the liquid bath to the
cell, the preheating phase was critical
to limit the risks of infiltration. This
phase was carefully controlled using
The measured values were compared

with the modeled ones and the deviations were analyzed. The results will
be used for further development of
the technology.
The start-up of the line was accomplished according to the schedule and
without significant start-up and early
life issues, a tribute to the expertise
of all the teams involved.
Early results at Jonquire.

The MHD modeling software developed
by the RTA R&D teams has evaluated
the bath metal interface of the LRF cell
and the Arvida smelter cells. It shows
that the Arvida smelter cells will have
a flatter interface than the LRF cell,
opening the way for very promising
results.
This prediction has been confirmed by
the early operation of the cells at the
Arvida smelter. After a few months of
operation at 565 kA, the cell stability
was very good. The cells instability
(WRMI) was less than 40 Nano-Ohms.
This unique cell stability will be used
for further development of the AP60
technology.
After the initial start-up phase, current
intensity has been ramped-up from
560 kA to 565 kA in Q1 2014.
AWJ 2014 47
A specific and comprehensive performance test will be done to assess the

industrial performance of the technology. At the time this article was written,
the performance test is planned to be
organized in Q2 2014.
The next steps of the AP60 technology development.
As table 3 shows, the two technologies

have been developed and tested in
parallel, using the same optimizedframework (busbars, shell and superstructure) and equipment to operate
the cells. Specific elements such as
cathodes, anodes, and shell ventilation differentiate the two cell designs
in order to operate at a high amperage
(AP60) or low energy (APXe).
The Arvida Smelter, AP60 Technological Center will be the platform for
future development of the AP60 technology. Table 2 shows the targets for
the development of this technology.
13,5
kWh/t
Intensity
(kA)
570
600
12,0
SEC
(kWh/kg)
13.3
13.0
11,5
Production
(t/day)
4.31
4.51
AP60 and APXe

In response to market demands, Rio
Tinto Alcan has developed a strategy
based on a common platform able to
deliver high performance cells: a high
amperage cell with AP60; and a low
energy cell with APXe.
AP50
13,0
AP60
Second
Generation
The center will also be used to develop

new environmental technology as well
as operational automation development such as a fully automated anode
change.
Conclusion
The 38 AP60 cells have been successfully started in the new Arvida smelter
AP60 Technological Center. In delivering the project, Rio Tinto Alcan teams
14,0
AP60
First
Generation
Table 2. Futures targets of AP60

Technology
ment), APXe will benefit from the industrial validation of AP60 at Arvida
Smelter.
12,5
450
Hi Productivity
AP60
LE
APXe
500
550
Amperage (kA)
600
650
Figure 3: Operating regions of new AP cell technologies
AP60 will operate at 600 kA with energy

consumption in the 13-13.3 kWh/kg
range. APXe will operate around 500
kA with an energy consumption target
of 12 kWh/kg (see Figure 3).
In addition to energy efficiency and
cost-effectiveness, AP60 and APXe
comply with Rio Tinto Alcans demanding HSE standards.
Since APXe (low energy) and AP60
(high productivity) share the same
optimized framework (busbars, shell,
superstructure and operating equipAP60
APXe
Busbar
Common
Shell
Common
Superstructure
Common
Alumina Feeding device
Common
Anode assembly
Common
Cathode and Lining
High productivity
Low energy
Shell ventilation
High productivity
Low energy
Gas flow
High productivity
Low energy
Pot Control System (Alpsys)
Common
Equipments (Pot Tending Assemblies,

Vehicles, ladles,)
Common
Buildings
Common
Table 3. AP60 and APXe configuration
have shown complete mastery of all

aspects of this very challenging high
amperage pot technology, from design
to construction, to start-up and finally
operation. This represents an important
milestone for the aluminum industry
as AP60 dramatically increases cell
productivity over previous technologies.
The new industrial platform is already
equipped with a booster and will be
the basis for future development of
the technology.
With the validation of AP60 at the Arvida Smelter, AP60 Technology Center
and the APXe cell at the LRF, these two
technologies enable Rio Tinto Alcan
to stay in the vanguard of reduction
technology for the benefit of its own
pipeline of internal growth projects,
and of the projects of its partners and
customers.
The authors would like to thank all of
the pioneers involved in the development, construction and demonstration
of the AP60 technology from both
France and Canada.
Norsk Hydro
Dubal
Expertise in
Alumina handling
FLSmidth is your expert in handling of fresh alumina, reacted alumina, crushed bath
and aluminiumfluoride
s
Large capacity storage silo including anti-segregation filling and discharge
s
MLLER airlift conveying
s
Pressure vessel dense phase conveying either with MLLER Turbuflow or

standard conveying pipe
s
Truck/wagon loading and unloading stations
s
Dosage systems
s
MLLER Fluidflow pipe air slide conveying systems
s
MLLER direct pot feeding systems
s
PTM filling stations
s
Modular designed systems - plug and play -
FLSmidth Hamburg GmbH

Tel: +49 4101 788-0 s hamburg@flsmidth.com
www.flsmidth.com
AWJ 2014 49
MLLER Alumina Handling Systems

High performance, high efficiency
About FLSmidth
FLSmidth is a market-leading supplier of equipment and services to the
global minerals and cement industries.
FLSmidth supplies everything from
single machine units to complete minerals and cement flow sheets including
associated services.
With more than 15,000 employees,
FLSmidth is a global company with
headquarters in Denmark and local
presence in more than 50 countries
including project and technology
centres in Denmark, India, USA and
Germany.
FLSmidth has over the past 131 years
developed a business culture based on
three fundamental values: competence,
responsibility and cooperation.
It is FLSmidths vision to be the customers preferred full-service provider
of sustainable minerals and cement
technologies. This is reflected in focused research and development efforts aimed at fulfilling customers
future needs in terms of innovative
technical solutions, high reliability and
availability, minimum environmental
impact and the lowest possible product
lifecycle costs.
FLSmidths in-house resources are

primarily engineers who develop, plan,
design, install and service equipment,
with most manufacturing being outsourced to a global network of subcontractors. This has proven to be both a
robust and sustainable business model.
FLSmidth therefore has a flexible cost
structure, which makes it possible to
plan and adjust resources to prevailing
market conditions.
FLSmidth is a learning organization,
and our people are our most valuable resource. FLSmidths strategy
entails strong emphasis on selecting, attracting and retaining the right
people who can support value creation
in FLSmidth.
FLSmidth in the alumina

business
FLSmidth first entered the alumina industry more than 100 years ago. Today
FLSmidth has an experienced team
of engineers and support staff with
extensive alumina experience located
in offices around the world and offers
the latest equipment for most areas
of an alumina plant.
Red side, white side and alumina

handling
Based on the Bayer process, invented
by the Austrian chemist Josef Bayer,
the alumina production process can
be split into a red side and a white
side.
Red side solutions
FLSmidth offers equipment for the
complete bauxite handling, storage,
crushing and grinding flowsheet, complementing the digestion or dissolution
of bauxite in hot caustic liquor. This
is followed by the complete SettlerWasher train flowsheet for Red Mud
using the leading technology acquired
from Dorr-Oliver Eimco.
White side solutions

FLSmidth offers white side equipment, covering the complete flowsheet after the hydrate precipitation
process, including MLLER equipment
technology for alumina handling and
load-out.
Overall, FLSmidth equipment covers
more than 50 percent of the equipment needs of a complete alumina
plant, from the bauxite mine to the
above refinery equipment. In addition,
FLSmidth also offers all equipment for

alumina handling in the smelters.
MLLER Technology
Through its MLLER technology,
FLSmidth specializes in design, engineering, procurement, erection and
commissioning of pneumatic material
handling systems for turnkey projects
and components for the alumina industry. Our capabilities of handling fresh
alumina, reacted alumina, crushed bath
and aluminium fluoride comprise:
Large capacity storage silos (up to
85.000 t realized) including antisegregation filling and discharge
MLLER airlift conveying systems
(up to 6oo t/h realized)
Pressure vessel dense phase conveying either with MLLER Turbuflow or our standard conveying
pipe
MLLER screw pump conveying
systems
Truck/wagon loading and unloading stations
Dosage systems
MLLER Fluidflow pipe air slide
and rectangular air slide conveying
systems
MLLER direct pot feeding systems either with 100% MLLER
Fluidflow pipe air slide conveyiing technology or as a hybrid of
MLLER Turbuflow conveying
pipe and MLLER Fluidflow pipe
air slide
PTM filling stations
Modular designed systems plug
and play For more than 75 years the MLLER
brand has stood for high quality standard systems with more than 5.000
references worldwide.
Presently FLSmidths Hamburg office
is executing three main orders in the
aluminum smelter industry.
plying two MLLER direct pot feeding

systems, reacted alumina silos, PTM
filling stations, fresh alumina truck
unloading stations and pressure vessel
dense phase conveying systems for a
mix of crushed bath/alumina oxide.
2. UC RUSAL/RusHydros Boguchansky Aluminium Smelter - A MLLER
direct pot feeding system is under
installation.
3. CVG ALCASA, Venezuela - Supply
of a MLLER pot feed system for two
production lines consisting of a total of
400 electrolytic cells. The installation
and commissioning has to be executed
during operation of the smelter and
shows the high adaptability of this
proven technology to serve brownfield aluminium smelters.
FLSmidth Hamburg GmbH
Haderslebener Strasse 7
25421 Pinneberg, Germany
hamburg@flsmidth.com
1. Emirates Aluminium Smelter Phase

2 with a length of 1.640 m the longest
pot rooms ever build - FLSmidth is sup-
AWJ 2014 51
Inline concentration monitoring in cleaning, pickling

and etching baths
Figure 1: Directly in the bath the LiquiSonic analyzer of SensoTech measures quickly and accurately the cleaner concentration and
degree of contamination.
Efficient cleaning of industrial parts

and treatment of surfaces by inline
analytical technology
Concentration fluctuations in cleaning and treatment baths affect the
bath quality and resource efficiency
if meeting desired nominal values is
not monitored continuously or only
imprecisely. Since the initial cleaner
concentration decreases during the
process, without proper bath monitoring the replenishment will be either
too much or too little. Underdosing
of the cleaner results in not meeting
the cleanliness requirements; overdosing results in wasting valuable
resources.
Further, the degree of bath contamination is subject to fluctuations that influence the cleaning performance and
resource efficiency. During the cleaning
process, the degree of contamination
increases steadily. Having the information about reaching the contamination
limit, bath changes can be done efficiently. Consequently, costs caused
by frequent bath changes, which are
reflected in the energy and raw material
consumption, can be reduced.
Cleaners are aqueous, anhydrous, basic

or acidic agents, for example. To measure the concentration of the cleaner
and the degree of contamination in
the bath continuously, accurately and
quickly, the LiquiSonic analyzer of
SensoTech will be directly integrated
into the process. Figure 1 shows an immersion sensor and the controller of
the analyzer. The technology measures
inline and properly in various bath liquids, because the measuring method
is independent of the turbidity, colour
and conductivity of the liquid and has
a high tolerance of soiling as well. The
automatic real-time measurements
every second show immediately how
much cleaner must be replenished to
keep the cleaning result constant. This
eliminates time-consuming manual
measurements, which provide delayed
laboratory results.
For example, the analyzer is used in
cleaning and degreasing baths, in rinsing baths, pickling and etching baths
and coating and plating baths as well.
Numerous industries like metal, automotive, semiconductor or plastics
industries, include such surface and
cleaning processes. In metal production, the LiquiSonic technology is

applied particularly in pickling baths. In
this application, the analyzer monitors
the concentration of acid or alkali, so
that the replenishment can be done
immediately and automatically. Moreover, in pickling baths LiquiSonic also
analyzes the content of the iron salt,
which arises as a by-product in the
bath due to the consumption of acid or
alkali. In addition to monitoring pickling
baths, the inline analytical technology
optimises further process stages in
metal production. For example, these
include acid regeneration processes,
rolling emulsion controls or electrolyte
analyses in continuous galvanizing and
roller chrome plating.
The pickling process

Pickling baths are used downstream
of the hot rolling process to clean,
modify or passivate metal surfaces and
to remove tinder or rust for the further
treatment in the following production
steps. Finally, the metal can be reformed in the cold rolling process, for
example, or electrolytically galvanized
within the surface refinement.
Figure 2: The pickling process can include several LiquiSonic measuring points for optimal process monitoring.
Pickling baths primarily contain solutions or mixtures of different mordants.

There are either diluted acids or bases
used. For example, typical chemicals
are sulfuric, hydrochloric, phosphoric
or nitric acid as well as caustic soda.
During the process, the concentration
of the mordant decreases, whereas the
portion of interfering components such
as metal salts increases at the same
time. In order to keep the pickling bath
quality in an optimal range, it is necessary to redose with fresh acid or lye in a
targeted manner. Moreover, the metal
salts do not only arise by etching of
contaminations, but also the mordant
dissolves the metal surface.
In order to produce with an awareness
of quality, economic factors and the
environment, checking the concentration of the bath ingredients precisely and continuously is required.
The permanent bath analysis reduces
the frequency of bath changes and
avoids overdosages. Thus, it leads to
resource-efficient process control. In
addition using online monitoring, an
optimum pickling result is ensured.
For this purpose the LiquiSonic

analyzer of SensoTech is integrated
directly into the pickling plant. Figure
2 represents a pickling process including different LiquiSonic measuring
points for optimum monitoring of each
process step. The analyzer consists of
one or more sensors and one controller. Installing the sensors into the bath
pipeline is easy and suitable for every
cleaning or surface processing plant.
For an effective measurement neither
a bypass nor smooth flow pipe sections are necessary. Due to the robust
construction and corrosion-resistant
material, the sensors are maintenancefree with a long lifetime. SensoTech
offers immersion sensors with variable
lengths or flange sensors.
Due to the option of a separated electronic housing, the sensors can be even
installed in mini-plants with limited
space. In case of low liquid flows, flow
meter adapters having a minimum
dead volume are used.
The measuring results are updated
every second and are available immediately and at any time. The controller
visualizes and manages the data that

can be transmitted to control systems
via fieldbus, analog outputs, serial port
or Ethernet. Thus an automatic, targeted cleaner dosage is guaranteed. For
process transparency and traceability,
the measuring data are permanently
recorded and stored in the controller.
Clear trend charts and custom data
protocols provide a comprehensive
overview of the process flow. In consequence of the reproducible process
management, it is possible to always
operate the pickling plant in optimum
condition.
Sonic velocity combined with conductivity

The bath liquid consists of the following components: water, acid or lye
and metal salts. The metal salts accumulate through the reaction of the
mordant with the metal in the bath.
The concentration monitoring of the
pickling solution and of the salts is essential for optimal bath results. In such
a 3-component mixture two physical
values are necessary to determine the
single concentrations of these both
substances. The measuring principle is
AWJ 2014 53
Inline concentration monitoring in cleaning, pickling

and etching baths
Figure 3: For concentration measurement in a 3-component mixture, a LiquiSonic measuring point consists of a sonic velocity flange
or immersion sensor and a conductivity probe.
based on the different effects exerted

by concentration changes of a liquids
single components on physical values
such as sonic velocity, conductivity or
density. This characteristic is stored as
a calculation model in the LiquiSonic
controller, so that the measuring results
of the physical values can be converted
into concentrations. So measuring
two physical values concurrently, the
simultaneous determination of two
concentrations can be realized.
According to studies, sonic velocity
combined with conductivity has turned
out as the best measuring method.
Therefore, the LiquiSonic analyzer
consists of a sonic velocity sensor and
a conductivity probe. The sonic velocity
sensor measures the concentration
of the mordant and is made of stainless steel or Hastelloy. In aggressive
chemicals the stainless steel sensor is
coated with a special material such as
Halar or PFA. The conductivity probe
is made of PEEK or coated with PFA.
Thus, both type of sensors are corrosion-resistant in almost all chemicals
and can be used at temperatures of
up to 180 C.
Furthermore, the LiquiSonic sensors
feature a highly efficient ultrasonic

ceramic that ensures a proper measurement even in the event of a high
proportion of gas bubbles in the liquid.
The conductivity probe measures the
salt concentration and is connected,
together with the sonic velocity sensor, to the controller. Figure 3 shows a
LiquiSonic measuring point, where a
flange sensor and a conductivity probe
is installed into a pipeline.
Use at AMAG
The company AMAG Austria Metall AG
is the nations leading manufacturer
of aluminium semis and casthouse
products for the processing industry.
The group purchased the LiquiSonic
analytical technology of SensoTech that
is used successfully to optimize the
processes in the aluminium pickling
plant. The bath operates with caustic
soda at a temperature between 50
C and 70 C. Therefore, it consists of
water, sodium hydroxide and sodium
aluminate. Sodium aluminate results
from the reaction of sodium hydroxide with aluminum. In the past AMAG
worked in a batch process and used a
titration analysis to control the pickling
bath. However, this measuring method
presents problems because of manual
sampling and time delays of the measuring results. Consequently, the goal
was to control inline and in real time
the quality of the pickling bath.
By the installation of the LiquiSonic
analyzer, this challenge has been satisfactorily resolved. The combination
of sonic velocity and conductivity measurement makes it possible to analyze
the pickling process continuously. The
concentration of caustic soda is determined by an immersion sensor made
of stainless steel and the concentration
of sodium aluminate by a conductivity
probe made of PEEK. Both sensors are
mounted with flange fittings in a DN
50 pipeline close to each other. The
rugged sensor design, non-corrosive
materials and the resistant sensor and
controller housing prepare the analyzer
for best usage under tough production
conditions. Figure 4 shows the installed
LiquiSonic immersion sensor in the
pickling plant of AMAG.
Via Profibus the measuring data are
passed to the process control system. The results are reproducible and
logged in the LiquiSonic controller.
In case of exceeding or falling below
a predefined threshold, or if process
Figure 4: In the pipe of the aluminium pickling plant of AMAG Austria Metall AG the LiquiSonic
immersion sensor is installed to monitor inline the concentration of caustic soda.
problems occur, the analyzer immediately signals an alarm. Therefore, it

is possible to react quickly on deviations and take countermeasures. In
our case, changing to the continuous
process was very important, because
in the past we had to stop the pickling
process once a week to exchange the
bath. Using the inline analyzer of SensoTech, we are now able to control our
pickling process continuously and in a
targeted way. So we have gained in yield
significantly and can save production
costs, explains Christian Pointner,
Process Engineer at AMAG.
Conclusion
The inline analytical technology LiquiSonic measures precisely, continuously and directly in the process the
cleaner concentration as well as the
degree of contamination in various
bath liquids. This enables an accurate and automatic replenishment
of cleaner and an efficient control of
bath changes. An application focus is
pickling baths in the metal industry,
for example. In these kinds of baths,
the analyzer monitors the concentration of the pickling solution and of the
metal salt. Via common interfaces,
the analyzer can be integrated into
existing control systems.
The online monitoring leads to an

efficient, safe and eco-friendly run
of the pickling process. On the one
hand significant results are quality
assurance and increasing the yield.
On the other hand costs are reduced
by saving energy and material.
Author:
Ms. Rebecca Dettloff
Marketing Manager
SensoTech GmbH
Steinfeldstr. 1
D-39179 Magdeburg-Barleben
GERMANY
Contact:
Headquarters
SensoTech GmbH
Steinfeldstr. 1
D-39179 MagdeburgBarleben
Germany
U.S. Office
SensoTech, Inc.
1341 Hamburg Tpk., Suite 2-3
Wayne, NJ 07470
USA
SensoTech:
For over 20 years SensoTech has been
focused on the development, manufacturing and sales of inline analysis systems for process liquids. With
worldwide installed, highly precise
and innovative measuring systems for
monitoring of concentrations, compositions and changes of chemicals
as well as properties directly in the
process, SensoTech has significantly
contributed to the enhancement of the
state-of-the-art. In addition to the measurement of concentration and density,
the phase interface detection as well as
the monitoring of chemical reactions
like polymerization and crystallization
are typical applications. SensoTech
inline analyzers set standards in the
technological and qualitative valence,
user friendliness and reproducibility
of process values. Special calculation
methods and sophisticated sensor
technologies enable reliable and precise measuring results even under the
most difficult process conditions.
The knowledge and the experiences
of the highly motivated and committed SensoTech staff are the result of
many different applications supported
by well-known customers from the
chemical and pharmaceutical industry,
food technology, semiconductor technology, automobile and metal industry
as well as many other industries. In
addition, these experiences also open
up unimagined solution possibilities
for new measuring challenges.
T + 49 39203 514100
F + 49 39203 514109
info@sensotech.com
www.sensotech.com
T +1 973 832 4575
F +1 973 832 4576
sales-usa@sensotech.com
www.sensotech.com
AWJ 2014 55
Precise Anode Beam Positioning from Power Jacks
UK Manufacturer Power Jacks

has been a supplier to the Primary
Aluminium Industry for Anode Beam
Positioning Systems ever since the
very first commercial smelters were
established.
Based near Aberdeen in Scotland, at
the heart of the North Sea Oil & Gas
industry, the company manufactures
for a diverse range of industries where
lifting and position solutions are
required and exports its products to
more than 70 countries worldwide.
The products are supplied to end users

and OEMs in various sectors including
Energy, Industrial Automation,
Defence, Medical, Transport and Civil
Engineering.
The Company was presented with
a Queens award for Enterprise in
International Trade in 2011.
The companys product portfolio
includes screw jacks, actuators
and lead screws that offer an
electromechanical alternative to
hydraulic actuation for positioning
applications with loads from a few
kilograms to hundreds of tonnes.
Electromechanical solutions can
provide significant advantages over
hydraulics, not least removing the
need for expensive hydraulic pumps
and pipework, but also where there is
a demand for increased safety (in the
event of power loss, screw jacks can be
self-locking), the need for machinery
to operate with better energy
efficiency, machinery that operates
with greater levels of precision, less
maintenance or manual intervention,
Hydro
and also providing cleaner and quieter

solutions.
While users normally associate the
term jack with lifting, the company is
keen to emphasise that its products are
used for many different applications
in industry. As well as lifting and
lowering, the products are equally
used for tilting and pivoting, rolling,
locking/unlocking and tensioning of
loads in any direction.
The energy sector is a key market for
the company with many applications in
the Oil and Gas industry, both topside
and subsea with bespoke pressure
compensated products operating
successfully 3000m below the
surface on the seabed. Applications in
renewable energy include wind, wave
and increasingly solar energy where
they are used to position mirrors (or
heliostats) in large solar arrays.
For the Primary Aluminium industry,
in particular aluminium smelter pot
rooms, screw jack systems are the tried
and proven method for the lowering
and precise positioning of the anodes

and anode beams and also for highjacking of frozen pots. The same
method is used in both Soderburg
and Prebake although jack styles and
configurations vary depending on the
technology employed.
The screw jacks can be individually
motorised or mechanically linked
through a combination of transmission
elements and the anodes are
positioned via an electronic control
system.
The company says it has more than
15,000 units deployed in smelters
around the world. Lifetime spares
generally represent less than 1% of
the total capital investment over the
lifetime of the products being least
35 years.
This demonstrates the quality and
reliability of the products in what is
an integral function of the potroom
operation.
For positioning the anode beam on a
reduction cell, the company designs
products to cope with the harsh

operating environment encountered
in an aluminium smelter.
The use of special alloys and heavy
duty sealing arrangements protect
the product from abrasive alumina
dust to ensure durability.
Products are designed that allow
specifically for side-loading stresses
caused by thermal expansions on the
superstructures with either tilting or
swivel mechanisms, used extensively
in pre-bakes where the beam is not
self-supported, allowing for pivoting
movement in the lifting screw from
the vertical in any direction.
Power Jacks has developed products
to suit reduction cells derived from all
the major PB technologies including
Alcoa, Alusuisse, Dubal, Hydro, Kaiser,
Pechiney, Reynolds, VAW and others.
In tailoring a product to fit within the
OEMs superstructure, Power Jacks
aim is to minimise potline cost and
support the client to maximise
yields.
The company says that while reduction

cell amperages have increased
significantly as has the carbon area
with the use of larger anodes, there is
scope to optimise the anode-cathode
interface.
Power Jacks believes that as many
smelters constructed 40 50 years
ago will ultimately need to replace
their anode beam positioning systems,
new technology could offer the
industry greater positional accuracy
that could lead to significantly reduced
running costs.
In pre-bakes, the anodecathode
interface should ideally remain
constant at a level where the voltage
maintained is as close as possible to
the theoretical tension imposed by
the electrochemical reaction and the
Faraday efficiency is highest.
If the interface is too high, parasite
Joule effect causes heat losses in the
bath. If too short, the same occurs in
the anode and stem.
Dubal
AWJ 2014 57
Dubal
Alusuisse
Pechiney
Traditional jacking systems adjust

periodically the interface by lowering
the anode system. This frequency in
a pre-bake is generally much lower
than the alumina feed frequency
which itself provokes a change
in bath resistance and justifies an
adjustment.
Without going into print on details,
the company claims that a new design
of anode jacking system could offer
both greater positional accuracy,
continuous duty cycles and load
feedback information enabling the
anode to be adjusted in conjunction
with the alumina feed.
Power Jacks is interested in talking to

aluminium smelters willing to act as
partners for the new technology.
In addition to new installations, the
company offers of all existing anode
beam positioning systems in the field
whatever the origin. It recognises the
need to eliminate downtime and
will work with smelters to offer fast
repairs and to establish annual spares
requirements.
Author: Bruce Bultitude,

Managing Director, Power Jacks
Ltd.
Contact:
bruce.bultitude@powerjacks.com
www.powerjacks.com
www.powerjacks.com
AWJ 2014 59
ANODE PLANT TECHNOLOGY

Fives
Performance of the Eolios Pitch Fume Treatment System
p. 62-65
Hydro
historical evolution of closed type anode baking furnace technology
p. 66-71
An Immediate Step In Cost Reduction For Inert Anodes

Louis Dekker
p. 72-73
Innovatherm
Operational and environmental benefits of the new baking furnace
at Boyne Smelters by use of an advanced firing technology
p. 74-81
Upgrade of an existing Fume Treatment Plant at ALUAR to cope with

higher production in the new open type anode baking furnaces
p. 82-87
AWJ 2014 61
Performance of the Eolios Pitch Fume Treatment System

Coal Tar Pitch: The Source of PAH
Emissions
Nature of PAHs
PAHs are the main components of
coal tar pitch, used as a binder for
the fabrication of anodes. To be used
in the process fabrication of anodes,
it is heated to become liquid and is
mixed with pre-heated aggregates
(Calcined Petroleum Coke) to obtain
a homogeneous hot paste (160200C).
Eolios at Qatalum paste plant
By Belbachir Salima,
Solios Carbone and
Alix Courau,
Solios Environnement
Green anode plant activities for
aluminum smelters generate noxious
vapors containing CTPV (Coal Tar Pitch
Volatiles) especially PAHs (Polycyclic
Aromatic Hydrocarbons). Fumes
collection and treatment prevent the
release of these pollutants into the
working environment and atmosphere.
Traditional pitch volatile capture has
been performed with coke injection
followed by filtration (conventional
dry scrubbers).
In recent years, RTOs (regenerative
thermal oxidizers) have been used
at some plants to improve PAHs
destruction particularly on the lighter
fractions characterized by their low
boiling point. In order to increase
destruction efficiency at lower operating
costs, Fives Solios provides a dual
approach based on the combination
of a conventional dry scrubber and an
RTO. In such an approach, the RTO is
dedicated to hot pitch fumes, while
the dry scrubber is dedicated to lower
light PAH fraction emissions
This combined solution is named Eolios
and has been successfully installed
in Mosjen, Norway and in Qatalum,
Qatar.
Upon reaching this temperature range,

the coal tar pitch releases a high amount
of CTPVs which are mainly composed
of PAHs. PAHs can be divided into two
groups: light and heavy fractions.
The EPA regulation has identified 16
PAH molecules as priority pollutants to
monitor, based on their carcinogenicity
and occurrence, including naphthalene
whose carcinogenicity for human
health is today contested by some
studies. The OSPAR regulation only
monitors 11 PAH molecules, which are
mainly the heaviest PAHs, i.e. the most
carcinogenic ones.
PAH Emissions in the Fabrication

Process of Anodes
In paste plants, PAH emissions occur
in all areas where coal tar pitch is used
or stored at a high temperature. The
main areas affected by these emissions
are the paste mixer, the paste cooler
and the anode forming area.
The proportion of light PAHs appears to
be 70% of total emissions, compared
to around 30% for heavy PAHs.
The anode paste cooler equipment
has been introduced in the anode
fabrication process to improve anodes
density. Using water to cool down anode
paste lowers the partial pressure of PAH
vapors generated by the paste, thus
realigning the equilibrium towards the
production of light PAH fractions.
Consequently, adding water into the
paste cooler drastically increases the
62 ANODE PLANT TECHNOLOGY
amount of light PAHs. This phenomenon was first confirmed at the Alcoa
Deschambault paste plant, (Canada)
where light PAH emissions were multiplied by six after paste cooler installation.
Conventional Coke Dry Scrubbing

System
The dry scrubbing system is derived
from the potline alumina dry injection
scrubbing technology. This treatment
system was developed by Fives Solios in
1977, and has been installed since then
in more than 70 plants in the world.
It consists of injecting coke fines in
a gas stream loaded by pitch fumes:
causing an adsorption phenomenon
to occur between the coke and pitch
fumes. Then, the fines fraction of coke,
readily available for the preparation of
the anode paste, is injected countercurrently to the fume-laden stream.
Turbulence and highly efficient contact
between the pitch fumes and aerosols
are thus promoted inside a Venturi
reactor. Pitch loaded fines are then
collected through the dust collector
and reintroduced into the anode
paste recipe with their condensed
hydrocarbons. Finally, the clean gases
are released to the atmosphere through
a discharge stack.
The global PAH capture efficiency of
dry scrubbing varies between 90%
and 98%, depending on the treatment
temperature and the list of PAHs
involved.
This efficiency ratio is higher for heavy
PAHs. For these components, the
common efficiency is about 99.5%.
The dry scrubbing system philosophy
is based on the ability of PAHs to
condensate when establishing contact
with coke fines inside the reactor: this
allows their adsorption on the surface
of coke fines. That explains the higher
efficiency of the process on heavy PAH
fractions, which condensate even more
easily in comparison to the lighter PAHs.
The most toxic PAHs being the heaviest
fractions (B(a)P or equivalent), they are
extremely well handled by conventional
coke dry scrubbers.
Coke + Pitch
Paste Mixer
+Water
Paste Cooler
Coke Fines Injector
Anode Former Area
Typical Dry-Scrubbing - scheme
RTO: New Technology for a Better

Destruction of light PAHs
The RTO Principle
PAHs, mainly composed of C-H bonds,
are easily broken by oxidation as
follows:
CaH2b + (a+b/2) O2 -- aCO2 + bH2O
+ Heat
valves to isolate inlet & outlet duct

fumes and one purge valve per chamber.
Eolios at Qatalum Paste Plant: An

Innovative Solution for Optimum
Emission Performances
The oxidation temperature is about

850-900C, depending on the nature
of PAHs. Good combustion practices
include management of the 3Ts:
Temperature, Turbulence and Time.
Once the system was fine-tuned,

performance tests were undertaken
at the main stack by a third party. PAH
emissions were measured as per ISO
11338-1-1 standard.
RTO Description
RTO at Qatalum paste plant
The RTO is a compact equipment

composed of 3 main elements:
Stringent emission requirements have

led Fives Solios to install the Eolios
system to treat coal tar pitch fumes.
The combustion chamber, where

the oxidation reaction takes place
measurements were performed in order

to estimate the relationship between
RTO temperature and destruction
efficiency. The FID measures the
concentration of Total Volatiles Organic
compounds and they show clearly that
a higher combustion temperature
enhances RTO efficiency due to
an increase in organic compounds
destruction: 91.2% efficiency measured
at 840C against 96.2% at 870C.
Ceramics beds, used as heat

exchangers: which store and
recover heat to preheat the inlet
gas flow, allowing for energy
savings,
The high concentrated fumes collected

from the paste cooler, which contain
mainly light PAH fractions and water
vapor, are specifically treated by an
RTO unit. Fumes emitted by the
remaining pieces of equipment (mixer,
vibrocompactor, etc.) are treated by a
dedicated dry scrubber.
The valve box, which includes two
A set of FID (Flame Ionization Detector)
Results display a 0.96 mg/Nm 3

concentration for 16 PAH (Norwegian
Standard 9815) corresponding to a
global destruction efficiency for the
Eolios system of 99%. This value sets
a new benchmark for Eolios designed
without pre-filter.
Eolios at Mosjen: Good Results

after 6 years
Fives Solios supplied its first Eolios
at Alcoa Mosjen Aluminium Smelter
in 2007.
AWJ 2014 63
Stack
Dry Scrubbing System
Calcined
Petroleum Coke +
Liquid Pitch
Paste Mixer
Electrical Hot Air

Generator
Coke Fines
Injection
Offers Suction Points

(Anode Forming Area
Mixer)
RTO
+Water
Paste Cooler
Eolios at Qatalum paste plant
On this reference, the ventilation of

the green anode plant is split into two
separate lines: the wet fumes line,
treated with a dry-scrubber followed
by the RTO and the dry fumes line,
treated with a dry-scrubber only. These
two lines are implemented in parallel
and join each other at the stack.
On both lines in-line coke injections are
installed which allow for the catching
of pitch vapors on the coke fines close
to the location of emissions. The dry
line generates a low concentration of
pollutants mainly taken in charge by
a Venturi reactor with coke injection
followed by filter bags. The wet line
collects much more gaseous tars and
PAHs treated by RTO associated to a
pre-filter with coke injection.
Eolios performance has met the PAH
plant requirements of 0.05 kg/h (NS
9815) or 0.8 mg/Nm3 for the whole
plant.
The first important maintenance intervention on the RTO appeared after 6
years operation. Regarding the dryscrubber, regular preventive maintenance has ensured the correct running
of the plant over this whole time.
Eolios vs. Full RTO

The introduction of the RTO in the
Eolios technology can raise a question
about the possibility to use it for all
paste plant streams.
The advantage of such a configuration is
to have a diluted stream that is optimal
for RTO operation. However, as the RTO
is more efficient on light PAHs, the
global destruction efficiency for PAHs
in whole (light and heavy) is limited at
95%. This limited efficiency is due to the
adsorption/desorption phenomenon of
heavy and intermediate PAHs that occur
in heat exchangers during a cycle. This
configuration seems totally unable to
achieve the 99% efficiency observed
with Eolios at Qatalum and Mosjen
Paste plant.
An alternative to the full RTO is the
Hybrid RTO which conveys highly
concentrated streams directly to the
combustion chamber via a specific
burner.
The energy consumption for such configurations (full RTO and full Hybrid
RTO) is another parameter to take into
account: A more diluted stream
Mosjen Paste Plant RTO
decreases pitch fumes concentration, taking the RTO away from auto
thermal mode. A full RTO of 50,000
Nm3/h, requires around 100 m3/h of
gas, whereas a RTO integrated in the
Eolios technology will only consume
about 20m3/h of gas. The hybrid RTO
requires even higher gas consumption, up to 120m3/h (hybrid flow rate:
3,000Nm3/h & main RTO inlet: 27,000
Nm3/h).
EOLIOS
RTO: 5 000
Dry Scrubber: 45,000 m3/h
m3/h
97/98% (Proven)
1mg / Nm3 (Measured)
RTO: 50,000
m3/h
Full RTO
PAH16 Destruction / Capture Efficiency (NS9815)

95% max guaranteed
>5mg/Nm3 (Published)
By pass: treatment by the DS during

RTO maintenance
<20 m3/h
(44,000 /year)
170 T/Year
By pass not possible: = redundancy

required or prod. Stoppage
Natural Gas Consumption
100 m3/h
(220,000 /year)
CO2 Emissions
850 T/Year
Full RTO vs Eolios
Moreover, the flow rate passing through

the heat exchanger is used to size the
RTO. RTO dimensions will increase
with flow rate, therefore considerably
increasing CAPEX.
by the coke dry-scrubber and keep

the RTO ready to run on dirty fumes
without delay when the problem is
solved and inlet conditions are back
to normal operational ones.
In both configurations, the system (full

RTO and hybrid RTO) cannot benefit
from the biggest asset of Eolios: flexibility. RTO direct by-pass to the atmosphere does not exist in the Eolios
solution, as the RTO by-pass is directed
to the dry-scrubber. It means that pitch
fumes are at least treated by the coke
dry-scrubber in case of an unexpected
problem that would require stopping
the RTO. This offers an attractive possibility to switch on the dry-scrubber
by-pass which remains quite clean in
terms of emissions. This is very comfortable for operation.
Last but not least, Eolios offers a

higher treatment capacity. To limit full
RTO CAPEX, RTO suppliers tend to
reduce the treated flow rate as much
as possible, where temperature is high
enough for condensation problems.
Eolios does not need to heat fumes
on its dry-line, as coke injection of the
dry-scrubber adsorbs tars at an ambient
temperature. It is quite interesting to
be able to increase ventilation of the
paste plant, as the possibility to treat
more air with same inlet data is directly
linked to safety issues (ambient air
quality in paste plant). The possibility
to increase the treated air flow is also
useful for revamping considerations
and much easier to perform on Eolios
than on a full RTO solution.
Whereas a direct by-pass must be

used at the strict minimum in case
of emergency, Fives Solios by-pass
including fumes treatment can also be
used during maintenance or operation.
An RTO has important inertia due to its
ceramics and can be protected, using
the by-pass, from abrupt and impetuous
variations at the inlet. Therefore, any
of these variations can be absorbed
from pitch vapor. Dry scrubbers remain

the most efficient technology to treat
heavy PAHs, however the introduction
of an RTO to treat lighter PAHs is a
radical improvement for the whole
pitch fumes treatment system.
The Eolios solution, which combines
dry scrubbing and an RTO, allows
operation of the green anode plant while
maintaining PAH emissions at a level
below such stringent requirements.
Eolios exhibits a lower operating cost
and a smaller carbon footprint than
alternative technologies, such as full
RTO for instance, while providing very
comfortable flexibility of operation
thanks to its convenient by-pass to
dry-scrubber.
The implementations of Eolios at
Mosjen Paste plant and recently
at Qatalum Paste plant confirm
the benchmarking performance
and benefits associated with this
technology.
Conclusion
Environmental emissions treatment
requirements are becoming more and
more stringent, especially for suspected
carcinogenic substances such as PAHs
AWJ 2014 65
Hydro Aluminiums historical evolution of closed

type anode baking furnace technology
carbon plants in Sunndal and rdal
to supply the company smelters with
sufficient quantities of anodes in the
desired dimensions.
The plan was to design and construct
two proprietary prototype sections.
These were tested in the baking furnace
at Sunndal in 1983.
The main design criterion for the
test sections was to adapt the pit
geometry to new anode dimensions
and to maintain or increase production
capacity.
The key aims during the reconstruction
of the furnace was that it should be
done within the existing building,
without major dimension changes to
the existing concrete tub and that the
distance between the headwalls was
maintained.
Figure 1. Construction site of the Riedhammer furnaces #1 and #2 in rdal (1958).
Authors: Michal Tkac1, Anders Ruud1,

Inge Holden2, Hogne Linga1
1. Hydro PMT, Primary Metal Technology,
rdal, Norway
2. Hydro Aluminium rdal Carbon,
rdal, Norway
Abstract
This paper summarizes the historical
evolution of the closed anode
baking furnace technology from the
Riedhammer design to the Hydro
Aluminium concept in the Norwegian
carbon plants; rdal and Sunndal over
the last 50 years.
The increasing demand for higher
production and larger anodes during
the last 30 years has required Hydro
Aluminium (HAL) to design a proprietary
high capacity HAL baking furnace
concept. Some major aspects and
challenges connected to the rebuilding
of the furnaces are described, including
maximum utilisation of the existing
factory space, allowing a low CAPEX
per annual production capacity.
Development of new repair and
maintenance methods for critical
refractory parts were essential in order

to maintain a high anode quality and to
extend the furnace service life.
Main improvements related to the
process control, process safety and
performance data of the current
technological status are presented.
Introduction
Historical Development at Sunndal
Furnace
The production of prebaked anodes
in Sunndal was initiated in connection
of the paste plant start-up in 1968.
The first anodes were produced in
a closed type Riedhammer furnace
designed for an annual capacity of
54,000 tonnes.
Until 1984, Norsk Hydro (SV) had five
anode baking furnaces of the closed
Riedhammer design in operation in
Norway. The total annual production of
these furnaces was 160,000 tonnes.
The need for higher metal production
volumes and consequentially bigger
anodes initiated a strategic plan for the
After a one year successful test period,

the entire Riedhammer furnace was
rebuilt according to the design of
the prototype sections. The Sunndal
furnace was rebuilt divided in to 28
sections, each with 7 pits and was
operated with two fires.
The annual production prior to
reconstruction in 1984 (54,000
tonnes of baked anodes) increased
to 63,000 tonnes (about 17%) after
the rebuilding.
With the next furnace retrofit in 1997,
the Sunndal furnace achieved a
further increase in capacity to 80,000
tonnes.
Historical Development at rdal
Furnaces
The very first production of prebaked
anodes in SV rdal is dated back to
1958. Two closed Riedhammer type
furnaces #1 and #2 were built with an
annual capacity of 18,000 and 15,000
tonnes. Due to the demand for larger
anodes, the height extension of furnace
#1 was done in two steps; in 1971 and
then in 1980. Furnace #2 was modified
in a similar way in 1977.
Sunndal furnace
SNC
Start Operation
Technology
Annual capacity (t/y)
Ardal furnace #1
AAKI
Ardal furnace #2
AAK2
Ardal furnace #3
AAK3
Ardal furnace #4
AAK4
1968
1958
1958
1970
1971
Riedhammer
Riedhammer
Riedhammer
Riedhammer
Riedhammer
54,000
18,000
15,000
34,000
34,000
Number of fires
Number of sections
28
24
24
30
30
Number of pits/section
Number of anodes/section
108
60
60
90
90
Production rate (kg/pit/per hour)
514
257
214
388
388
1984
HAL 1st. gen
1971 and 1980
1977
1989
high extensions
high extensions
HAL 1st. 80,000 gen
63,000
22,000
22,000
53,000
Start Operation
Technology modification
Number of fires
Number of sections
28
24
24
30
126
60
60
105
514
314
314
605
prod. stop 1997
prod. stop 2010
Start Operation
Technology modification
1997
HAL 1st. gen
2004
HAL 1st. gen
1998
HAL 2nd. gen
80,000
112,000
102,000
Number of fires
Number of sections
28
30
30
126
168
168
652
913
832
Table I. Overview of historical evolution of baking furnace technology in SV furnaces in Sunndal and rdal
During the operational period both

furnaces were also used for calcining
of cathode blocks.
rdal Furnaces #1 and #2 were
eventually closed down in 1997 and
2010 respectively, mainly due to
operational economy issues and low
pit capacity utilisation.
Due to the metal production expansion
in rdal, two new Riedhammer furnaces
#3 and #4 were built in the beginning of
the 70s. Each of the new furnaces had
double the capacity of furnaces #1 and
#2, mainly because of the increased
number of sections, one extra pit and
a shorter fire advance cycle.
In 1988, based on the operating results

from the redesigned furnace at Sunndal, it was decided to rebuild furnace
#3 in rdal to the same concept, which
was denominated as the HAL concept.
Baking furnace #3 had 30 sections and
was operated by two fires. After the
reconstruction a production increase,
from 34,000 tonnes to 53,000 tonnes
(about 56 %), was achieved.
New designs of the refractory brickwork
structures were also developed in order
to achieve the ability to produce even
larger anodes within the increased
section load.
The reconstruction projects mentioned
above are described as 1st Generation
HAL Baking Furnace Technology, which
were implemented within the existing

concrete tub and without alteration of
the factory buildings.
The gas cleaning and transportation
facilities were upgraded to coincide
with increased production and modification of the anode dimensions. The
total reconstruction times (i.e. furnace
shutdown) varied from 80-130 days,
except with rdal Furnace #3 where
rebuilding was carried out in stages
during a period of 10 months and adjusted to a nearly normal production
level.
The last projects carried out were the
rebuilding of rdal Furnace #4 in 1998
and modernization of rdal Furnace
#3 in 2004, to what we
AWJ 2014 67
denote as 2nd generation HAL Baking

Furnace Technology. Unlike with the
1st generation, the existing concrete
tub was fully or partly renewed. The
concrete tub was still located within the
existing buildings, but it now allowed
for adjustments of the length and
width so that the pit geometry could
be tailored to the necessary production
capacity and the anode dimensions
required.
Section B-B
rdal Furnace #4 had 30 sections
and was operated by 2 fires and the
initial production capacity in 1971 was
34,000 tonnes/y. Since the rebuild in
December 1998, designed capacity was
for 102,000 tonnes/y. Reduced output
due to limitations in pit utilization and
periods when the furnace was operated
by one fire (50 % production), resulted
in an average production of 86,000
tonnes/y.
rdal Furnace #3 was restarted after the
reconstruction of HAL 2nd generation
concept in April 2004. Design capacity
was increased to 112,000 tonnes/y, but
the average annual production has
been 96,000 tonnes. Furnace #3 in
rdal achieved a production increase of
110 % since the previous rebuild (HAL
1st) and an increase of 230 % compared
to the original furnace design.
Currently the total production of the
3 furnaces with HAL technology in
operation (furnaces #3 and #4 in
rdal and the Sunndal furnace) is
approximately 290,000 tonnes/y.
Main Features of Riedhammer
and Hydro HAL Baking Furnace
Technology
Both Riedhammer and HAL concepts
are so-called closed type furnaces with
a vertical flue gas pattern.
The principle of the flue gas pattern for
the Riedhammer furnace is shown in
Figure 2. The exhaust gas enters from
the previous section underneath the
headwall and turns upwards through
the firing shafts with counter flow firing.
Under the section cover, the gas is
Figure 2. Schematic drawing of the flue gas

pattern for closed type Riedhammer baking
furnace
dividing wall. After an energy input

from downstream firing, the flue gas is
channelled first upwards through the
A part of the flue wall. Approximately
60% of the fuel energy required per
section is supplied in such a way.
Under a section cover, the rest of the
fuel energy input is injected from
the vertically oriented cover burners
and the flue gas is sent downstream
through the B part beneath the pit floor.
The exhaust manifold is connected
directly from the section cover to the
ring main at the first section of the
fire zone.
Distribution of gas and energy input
between the A and B parts is the key
feature of the HAL concept which
enables operation of the furnace with
a high production rate per pit as shown
in Table I [1].
Section B-B
Figure 3. Schematic drawing of the flue
gas pattern for closed type HAL baking
furnace.
distributed downwards into the flue

wall channels and transferred via the
bottom channel system into the next
section. At the front of the fire zone,
the exhaust manifold leads off gas
from the exhaust take-off duct to the
ring main (Figure 6). In this way flue
gas passes the whole fire zone and
heats up the furnace sections.
The Hydro closed type furnace design
was developed with a different flue
gas pattern, as shown in Figure 3. The
exhaust gas enters the section through
the bottom part of the headwall directly
to the flue wall. Flue walls and the
bottom part of the pit are physically
divided into A and B parts by the
Rebuild to HAL 1st Generation Baking

Furnace Technology
The main requirements and major
aspects connected with the rebuild
of the Riedhammer furnace to HAL
technology are further discussed.
The most challenging part during the
rebuild was to adopt the new furnace
design into the existing building and
the same concrete tub.
This solution included step changes on
previous essential furnace construction
parts with a simultaneous focus on low
investment cost. All these changes
were adjusted in order to avoid major
changes in the existing infrastructure to
achieve the shortest construction time
without a large production loss.
The major cost saving was partly
achieved by reducing the amount of
installed refractory compared to the
previous furnace design (Table II). This
was achieved through a new headwall
design which included
Riedhammer
HAL 1st. generation
Sunndal furnace
11,500
10,000
rdal furnace #3
12,500
11,000
rdal furnace #4
12,500
Installed Refractory (t)
HAL2nd. generation
Table II. Overview for tonnage of installed refractory material for furnaces.
10,800
10,500
Figure 4. The new cover design for HAL

furnace. The exhaust take off hatch is
on the short side of the cover. Auxiliary
equipment is integrated in the cover.
removal of the firing shafts and the

exhaust take off ducts [2]. In this way,
the section width increased for two
additional pits (Figure 5 and Figure
7). New headwall brickwork was more
simple and required less brick types
and tonnage installed. The increased
section volume allowed loading of
several larger anodes which resulted
in a higher section load and improved
furnace productivity.
The extension of section dimensions
and the new concept of exhaust gas
take off required construction of new
section covers and exhaust manifold.
The new HAL section cover design
enabled even distribution of the
flue gas flow due to the rectangular
geometry of the cross sectional area.
Consequently, the flue gas velocity
gradients in the corners over the outer
and inner pits, were minimised [3, 4].
The main operational advantage of
the HAL section cover design is that
all peripheral equipment (gas burners,
thermocouples, additional air fans) are
integrated in the cover.
An increase of the cover width and
length required modification of the
construction design. The original rigid
cast iron frame was replaced with a
flexible steel frame construction
supported by a torsion stable steel
structure. The new cover was designed
with the intention to withstand
mechanical and thermal stresses
during the whole fire cycle. Thermal
expansion measurements and thermomechanical analysis were done prior
to modifications of the cover bottom
frame [5, 6]. As a result, expansion
Figure 5. Top view of the Riedhammer design

furnace #3 section from 1970. Exhaust take
off duct is integrated in the headwall.
Section A-A
Figure 6. Cross sectional view of the Riedhammer design furnace #3 section from
1970. Section with 5 pits, exhaust manifold
is connected to the exhaust outlet.
Figure 7. Top view of the HAL 2nd generation

design furnace #3 section with 7 pits after
retrofit in 2004.
Section A-A
Figure 8. Cross sectional view of the HAL 2nd
generation design furnace #3 section after
retrofit in 2004. Direct exhaust take off is
from the section cover.
joints at the long side of the steel

frame were introduced. This measure
also helped to reduce the mechanical
stresses on the cover arch insulating
brickwork.
Rebuild from HAL 1st Generation to

HAL 2nd Generation Baking Furnace
Technology
Increasing amperage in pot rooms set
persisting demand for larger anodes
during the 90s, which resulted in
further evolution of HAL 1st generation
baking furnace technology. Further
retrofit of the furnaces involved a
gradual rebuild of Sunndal and both
rdal furnaces #4 and #3.
Major modifications of the furnaces
comprised dimensional changes in
the concrete tub, both in length and
width. The most remarkable change
was the retrofit of furnace #3 where the
amount of pits increased from 5 to 7
per section. Increased loading tonnage
per section required calculations of
strength and an evaluation of concrete
tub thickness with respect to the wall
stability and thermal stress [7].
The 3D finite element (FE) modelling
of the furnace bottom structures was
done in order to ensure sufficient
bearing capacity and optimised flue gas
flow distribution [8]. The same study
evaluated several cases with various
widths (cross sectional opening) of the
flue wall bricks to show the effects on
the flue gas pressure loss and the flow
distribution in the flue wall. Heat profile
calculations, confirmed by temperature
measurements, enabled reduction
in the sidewall thickness and furnace
substructures height which in turn
increased the pit depth [9, 10].
Modelling and subsequent optimisation
of the important furnace constructions
allowed for a reduction in the amount
of installed refractory without losing its
functional value. Lighter construction
allowed operation with a faster fire
advance cycle and increased the
furnace productivity. After 4 years of
operation of furnace #3, extensive
modelling was conducted. 3D finite
element (FE) model of a baking
furnace and thermal and structural
analyses helped to understand the
mechanisms for stress build-up in
the baking furnace. The FE-model
was also used to compare the pros
AWJ 2014 69
and cons of different furnace design

solutions [11].
Refractory Maintenance
The Hydro furnace concept has adopted
the continuous refractory maintenance
strategy. This approach assumes
that the maintenance of refractory
parts is carried out without loss of
the furnace production. The furnace
is in continuous operation during
the regular maintenance; such as a
change of the flue wall or pit floor.
Production is stopped only during
the major rebuild periods when the
main furnace structures like sidewalls,
bottom insulation and headwalls are
replaced.
The evaluation of maintenance and the
lifetime of a baking furnace very often
depend on the amount of fire cycles.
For example furnace #4 in rdal was
originally built as a Riedhammer type
furnace in 1971 and was in continuous
operation until the rebuild in 1998. This
means that the original headwalls,
sidewalls, bottom insulation and covers
were 27 years old. Assuming an average
17 cycles per year, the furnace had
passed a minimum 460 cycles during
its service lifetime.
Another important fact was that
the complexity (amount of special
brick shapes) of the HAL design
for the critical refractory structures
was simplified. Thus less refractory
tonnage was installed which reduced
the heat capacity of the whole furnace
structure and enabled operation with
faster fire advance cycles. Improved
heat exchange between the anode
and the flue wall was reflected in
lower energy consumption. Lifetime
of furnace refractory brickwork depends
mainly on:
Refractory quality
Heat stress under normal
operation condition
Mechanical load combined with
operational routines
Routines deployed for continuous
maintenance under furnace
operation
When the furnace structures of the

new HAL furnaces were designed
selection of high refractory quality,
according to the installation position,
enabled customized solutions for the
high heat and mechanically loaded
structures such as the headwalls and
bottom pillars.
Selection of the refractory quality
resulted from an extensive test program
which included the characterization and
the evaluation of previous operational
experience. The most important
testing methods and the refractory
properties were selected [12]. Testing
included both the characterization of
the new materials and the analysis
of used refractory which reflected
the ageing effect of thermal cycling.
Results gained from testing helped to
identify and increase understanding of
the thermo-mechanical and chemical
stresses influencing the critical
brickwork structures of the furnace.
In connection with the introduction
of the HAL 2nd generation furnace
design, new maintenance and repair
methods were adopted. Evaluation of
the baking furnace brickwork condition
is based on the inspection of sections
with varying extent of damage.
General condition monitoring is done
at least every 2 years. Sections with
potential for large deformations and
displacements, especially the headwall,
are checked at shorter intervals (4-6
months). Damage mechanisms
and causalities for observed critical
deformations were explained and the
maintenance method solutions were
proposed [13].
Routine follow-up and evaluation of the
brickwork condition is an important tool
to predict and determine a timeframe
for a new renovation or modernization
of a furnace. Therefore, it is crucial to
focus on the construction portions
that are time-consuming and costly
to maintain. Hence, the condition of
the headwalls is crucial in determining
when to rebuild a furnace.
Figure 9. Schematic drawing of a surgical

reparation of the flue wall cracking in the
bottom corner of pit. The damaged part of
the brickwork is demolished and rebuilt while
the rest of the wall is stabilised by a chain
jack and maintenance platform.
Flue Wall Reparation Methods

During operation, packing coke tends
to stick to a flue wall and forms slag
around anchor slots in the headwall.
Blockage of expansion joints in a
headwall will prevent the flue wall from
free movement and cause deformation
and cracking. A deformed flue wall
causes narrowing of the pit and makes
loading of anodes difficult. In addition,
large cracks in the bottom of the flue
wall will cause problems with excessive
airburn of the anodes and packing
coke. Packing coke runs under the
pit floor and causes restrictions in flue
gas flow and prevents optimal heat
distribution.
With an introduction of so-called
surgical reparation of the flue wall
and pit bottom, it is possible to replace
only the damaged part of the brickwork
without changing the whole wall. This
reduces maintenance time and cost
significantly.
Refractory Lifetime and
Consumption
The combination of improved
understanding of stress build-up
mechanisms (FE-modelling) and
operational experience has resulted
in a major leap in the flue wall lifetime.
This is exemplified by the lifetime
distribution figures for the latest rebuild
of furnace #3 in 2004. The foreseen
Amount of flue walls
120
Lifetime distribution 1st generation flue walls

Lifetime distribution 2nd generation flue walls
Operational time distribution 2nd generation flue walls
100
80
60
Benefits from the continuous refractory

maintenance strategy are reflected in a
positive way by the extended refractory
lifetime of flue walls and low specific
consumption of refractory material.
40
20
0
This includes routine follow-up of the

brickwork condition, and precise planning of maintenance work with minor
production disturbance. Developed
surgical maintenance reparation methods assure maximum extension of
the refractory brickwork lifetime and
minimal maintenance cost.
500
Figure 10. Lifetime distribution of
1000
1st
1500
2000
[days]
andLifetime
2nd generation
of the flue walls at furnace #3.
refractory maintenance consumption

over the furnace lifetime of 20 years is
close to 5 kg/tonnes baked anodes.
The surgical maintenance of flue walls
gives a potential for further reduction
of maintenance cost and an increased
lifetime of furnace brickwork.
Furnace Design vs. Process Safety
The inherent design of the closed top
baking furnaces enables fire zones
to be set to a safe state in case of
loss of draft situations. The routines
include flaring of the pitch volatiles
until an oxygen excess situation is reestablished in the fire zones.
When the surplus air is established in
all fires, normal operation can be reestablished without any risk of forming
flammable gas mixtures by restarting
the draft. The inherent design feature
of the furnace, which allows the fire
zones to be set to a safe state, forms
the basis for the SIL21-classified safety
system of the HAL-furnace [14].
Conclusion
Increased potline amperage and
expanded metal production at SV
during the 80s resulted in a demand
for larger anodes. As a consequence, a
new HAL anode baking furnace concept
was developed and introduced through
1st and 2nd generation evolutionary
phases. All rebuild changes were done
with a strong focus on the investment

cost. The main dimensions of the new
baking furnaces were adapted to the
existing building with minor changes to
the existing infrastructure. The reduced
amount of installed refractory and
the short reconstruction time without
production loss had a positive effect
on the overall project cost.
Design modifications for the HAL 1st
generation furnace were done without
major dimensional changes on the
existing concrete tub or the distance
between headwalls.
Main features of the HAL 2nd generation furnace comprised of a partial
renewal of the concrete tub. Adjustments to the length and width of the
pit geometry were tailored to the anode dimensions required, without alteration of the factory building. After
retrofit, the new HAL furnace section
dimensions allowed operation with a
higher section load and production of
larger anodes. New design of brickwork
structures and less refractory tonnage
installed enabled operating with faster
fire advance cycles. As a result, a considerable increase in furnace capacity
was achieved. New section covers with
integrated auxiliary equipment were
developed in order to fit the new furnace section dimensions. The Hydro
furnace concept has adopted a continuous refractory maintenance strategy.
References
[1.] Holden et al., New process control system
applied on a closed baking furnace, TMS Light
Metals 2006, 603-608
[2.] Anders Ruud, Baking furnace concept
new headwall design, TEK95/015, Internal
report, 1995
[3.] Nigel Anderson, Differences in flue gas
distribution between take off from headwall and
cover, TEK93/110, Internal report, 1993
[4.] Nigel Anderson, Anders Ruud, Direct take
off from section cover for HAL furnace. Test
summary from Dec. 93 - Feb 94, TEK 94/023,
Internal report, 1994
[5.] E. Sandvik, J. H. Skaar, Development of
cover for baking furnace with larger section,
TEK96/156, Internal report, 1996
[6.] Anders Ruud, Cover design HAL7P
Dimensional stability and temperature profile in
steel frame, TEK97/149, Internal report, 1998
[7.] Fredleiv Fosse, AAK baking furnace 3,
Concrete tub for baking furnace, Building
evaluation of construction, Urheim AS
Consultant engineering, External report, 2002
[8.] Aage Jsang, Pressure loss calculation
HAL7P and HAL7Ps furnace 3 AAK, Research
centre Porsgrunn, Internal report, 2001
[9.] Aage Jsang, Heat balance sidewall
HAL7P furnace 3 AAK, Research centre
Porsgrunn, Internal report, 2001
[10.] Aage Jsang, Heat balance bottom pit
and sidewall HAL7P furnace 3 AAK, Research
centre Porsgrunn, Internal report, 2002
[11.] Henrik Bruzell, FE analyses of AAK
Furnace 3, Validus Engineering AB, document
ref: R0501-02_revA, External report, 2008
[12.] Anders Ruud, Refractory materials for
baking furnace, testing criteria with respect
to selection of material quality, TEK90/100,
Internal report, 1996
[13.] Anders Ruud, Maintenance of brickwork
for furnace AAK3. Recommended activities and
reparation methods, Internal note, 2004
[14.] Holden et al., Safe Operation of Anode
baking Furnaces, TMS Light Metals 2008,
905-911
AWJ 2014 71
An Intermediate Step in Cost Reduction for Inert Anodes
Process steps from idea to deployment

IDEA GENERATION
PROOF OF CONCEPT
Gate
2
Gate
1
Generate idea and
assess potential
PROTOTYPE
Develop the idea into

a feasible technology
Gate
3
Evaluate and
validate at
pilot scale
DEPLOYMENT
DEMONSTRATION
Gate
4
Demonstate and
optimize in lead
plant
Gate
5
Encapsulate the
technology in a
commercial
package
Process steps from idea to deployment
The Aluminium Industry has seen many

innovations in recent years developed
for smelters, especially in the areas
of pot design concepts and dealing
with ever higher amperages, and more
recently in the inert anodes of Rusal
(still to be commercially proven). The
inert anodes are not yet operational
at a large scale and will likely require
many years of further development
before they can be applied in smelters
worldwide.
A transitional phase will take place to
reduce the current costs occurring in
smelters. One area of cost reduction
relates to the mechanical and electrical
connection of the anodes to the rod
stub in the rodding shop. The current
connection method has a number of
drawbacks that contribute to the high
cost of aluminium production.
The mechanical connection of anodes
to the rod stub by means of grouting
the rod stub to the anode stub hole with
liquid cast iron has the disadvantage
that the seal created by this method
shrinks during the cooling process. Due
to this shrinkage an optimal electrical
contact between the anode and the
rod stub can not be achieved. Further
disadvantages caused by forming stub
holes with splines in the paste plant are
the risk of crack formation and broken
out pieces which lead to an increase
of rejected anodes.
A new design concept of plug and play
stub holes would reduce the amount
of rejected anodes caused by faulty
stub holes as well as reducing the
cost incurred in anode production,

especially taking into consideration
that the application of liquid cast iron is
expensive and involves a certain degree
of risk for the operator. A possible
solution is at hand.
the principle drawing fig 1). A major

advantage of the M-connection method
is making the formation of a stub hole
a lot easier as it eliminates the need for
splines. Consequently, the risk of crack
formations is largely reduced.
spiral spring
which establishes the

mechanical connection 2
electrical conductive
uid X
expands by heat
Anode
Fig 1. Mechanical and Electrical connection
A New Connection Method Concept

Mechanical Connection
Creating a plug and play connection for
the mechanical connection of anodes
to the rod stub by using liquid iron is
no longer necessary. I propose a new
method for the mechanical connection
of the rod stub to the anode stub hole,
here referred to as the M-connection
method.
This method can be achieved by
mounting a spring on the existing
rod stub and machining a groove
in the lower part of the stub hole.
These modifications will allow the
M-connection method to work as a
plug and play system (as shown in
The groove in the stub hole required for

the spring needs to be milled (forming
is not recommended).
Adapting the M- connection
method would increase the cost
of manufacturing anodes only
marginally as the initial cost would
be outweighed by the resulting
cost savings that could be achieved
in the rodding shop (20 to 30%).
Since tapped holes occupy an elevated
position above the anodes, the required
amount of butts is significantly reduced,
- in effect allowing for cheaper anode
production cost and expanding anode
pot life.
An Intermediate Step in Cost Reduction for Inert Anodes
Electrical Connection
Anode connected to a rod stub
The electrical connection can be

accomplished by an application
of electrically conductive fluid (X)
which will expand after the use of an
activator that will react upon reaching
a specific temperature (Y), becoming
one hundred percent conductive. Once
the connection is established voltage
loss becomes minimized, resulting in a
fixation of the rod stub to the anode.
The Principle of New Style

Rodding
Stubhole Spline
The modified rod stub will slide into the

stub hole and the spring will click on
the anode connecting with the groove
in the stub hole, thus establishing the
M-connection. Simultaneously the
conductive fluid (X) will flow around the
rod stub and start to expand and harden
upon reaching the correct temperature
(Y). One important property of the
conductive fluid (X) is the ability to
penetrate the porous surface of the
stub hole, thus establishing an optimal
electric conductive connection.
The advantages that can be gained
are:
A new, fast, safe and effective way
of connecting a rod stub to an
anode
Stubhole Defects A.
A connection method offering

excellent electrical conductivity
and significant cost savings for
rodding shop anode production.
Looking forward to the future, it will
be extremely interesting to discover
whether the new plug and play type
M-connection method concept has
sufficient commercial value for
further development according to
the principles and model introduced
here.
Author: Louis Dekker
Contact Information:
louis.dekker@zeelandnet.nl
Stubhole Defects B.
AWJ 2014 73
Operational and environmental benefits of the

new baking furnace at Boyne Smelters by use of
an advanced firing technology
This paper will outline the special
technologies used and demonstrate
the results that can be achieved to allow
for smarter and cleaner baking cycles
in combination with relevant safety
standards and system availability.
Figure 1. BSL on Boyne Island
Abstract
In February 2012, a new baking furnace
was commissioned at Boyne Smelters
Limited (BSL) on Boyne Island,
Australia. This new furnace replaced
the production of two existing closed
type furnaces and set new benchmarks
in production performance.
The advanced firing technology was
based on latest safety requirements
of the Australian Gas Authority (AGA)

gas standards. In parallel, new selfrecovery network structures behind
the wireless network have been
developed to maximize the redundancy
and availability of the system. Finally,
intelligent control modules have been
implemented for on-line optimization
of the baking process.
Introduction
Now part of Pacific Aluminium, Boyne
Smelters Ltd. (BSL) is a primary
smelter put into operation in 1982
by Comalco Aluminium and its joint
venture partners. BSL has undergone
extensive expansion over time. The
smelter underwent a A$1 billion
expansion in 1997 introducing a
third reduction line which increased
aluminium production from 260,000
to more than 550,000 tons per annum.
The company has further undergone
a significant modernization with the
re-building of Carbon Baking Furnace
3 (CBF3) and the construction of a
new Carbon Baking Furnace 4 (CBF4)
to upgrade the baking technology.
The new baking furnace replaced the
two existing closed type furnaces that
were obsolete and fully outdated in
terms of energy efficiency, emissions of
greenhouse gases and high refractory
maintenance and operational cost. An
aerial view of the smelter is provided
in Figure 1.
Details of the Carbon Bake
Furnace 4
The new baking furnace is an open-type
Rio Tinto Alcan AP design. It consists
of 66 sections, each with 9 flues and 8
pits. BSL operates 4 fire groups on this
furnace. The configuration includes 3
sections in the preheat area, 3 sections
in the firing area and 6 sections in the
cooling area. Figure 2 shows the actual
fire configuration.
The fire cycle time can vary between
24 and 32 hours, depending on the
production requirements.
Advanced Firing Technology

The Advanced Firing Technology is a
tailored system to suit the requirements
of BSL. The control philosophy is based
on the implementation of intelligent
Figure 2. New Firing System for CBF4 at BSL
Figure 3: Fire configuration for CBF4
optimization modules on top of the

basic automation for an optimum
system using todays state-of-the-art
and future oriented technology.
The firing system provides the following
features:
1* Fully automatic system
operation
2* Preheat Control with complete
internal pitch burn
3* Automatic Cross Over strategy
4* Safety interlocks according to
Australian Gas Association
(AGA) standards
5* High pressure pulse burner
technology
6* Preparation of the equipment
for fire direction reversal
7* Simple and easy operation
8* Advanced Control Modules for
Firing Optimization.
In addition to the firing control system
the following Auxiliary systems were
implemented:
9* Main Gas Supply Skid for
furnace gas supply, including
emergency stop (E-stop) circuit
10* Fume treatment centre (FTC)
data interface including main
draught interlock
11* OPC (standard data

communication protocol)
interface to plant wide
Manufacturing Execution
System (MES)
12* Ring main duct explosion vent
monitoring
13* Start-up Burners for firing
control below 750 C
An overview of the operator control
screen is shown in Figure 3.
Safety Requirements for AGA
The firing and control system for
the Boyne Smelter upgrade project
requires extensive safety interlocks as
per AGA and Queensland Government
requirements.
Each of the local programmable logic
controllers (PLC) executes process
control functions. All (classified) safety
interlocks are handled independently
from the process PLCs and are realized
by separate safety integrity level (SIL)
2 safety hardware. Safety signals are
forwarded independently from the
network infrastructure via a hardwired
daisy chain to the respective system
unit, as a result of a risk assessment
and the HAZOP study with the AGA
Queensland inspector.
The significant safety interlocks

required to comply with AGA standards
are:
Hard-wired flue draught release for
the burner ramps of the same fire
High and Low Temperature
interlocks at each burner ramp
E-Stops for the Main Gas Supply
Skid along the furnace bay
Main draught release (from FTC)
for the Main Gas Supply Skid.
Beside these primary interlocks
listed above, the following safety
functionalities have been implemented
for maximum operational safety:
14* Tightness test of primary
safety shut-off valves at the
Burner Ramp before fire start
and after each fire move
15* Pipe tightness test at Burner
Ramp before fire start and
after each fire move
16* Low Gas pressure supervision
and cut-off at the Burner
Ramp
17* EmergencyStop at each
Burner Ramp
18* PLC watchdog on each Ramp
19* Automatic calibration of all
draught / pressure sensors
every hour
20* Life Zero check for the draught
AWJ 2014 75
control system and the MES located

in the central control facilities. The
communication system becomes an
important key component of the firing
control system.
Wireless communication became
the logical state-of-the-art concept
as the equipment in the field needs
to be moved every day and a wired
communication concept immediately
creates reliability, operability and
maintenance issues. However, the
demand of a control system for
anode baking furnaces results in very
specific design criteria for the wireless
communication system and equipment.
The design as developed for CBF4
ensures a maximum availability of the
system by implementing redundancy
in combination with industrial Ethernet
and wireless equipment.
The wireless network in the field is
executed by four (4) stand-alone
industrial access points (AP) which work
independently to each other to provide
maximum redundancy. In the worst
case, one AP is sufficient to provide
the necessary communication between
the ramps. Additionally the APs are
connected to each other via selfrecovering fibre optic ring architecture
through industrial Ethernet switches.
The control architecture is illustrated
in Figure 4.
Figure 4: Self recovery network structures
/ pressure sensors every hour.

The combination of these additional
safety features and interlocks leads to a
significant increase of operational safety
and a more reliable and sustainable
production of anodes with respect to
the baking furnace environment.
Self-recovery Control Architecture on
Wi-Fi Bases
Due to the nature of the anode baking
process the firing system consists
of movable ramps performing the
required firing discipline while the
product remains at a constant position
during the baking process.
Each ramp operating on the furnace
is controlled by a local industrial PLC
located inside the control cabinet of

this particular ramp. All sensors and
actuators are connected to this PLC. For
local visualisation and backup operation
an Operator Panel (HMI) is installed
which interacts with the PLC and the
connected instrumentation executing
the required control functions for that
particular ramp. With the help of the
local HMI the operator is able to interact
with the ramp locally in different
operation modes on the furnace floor
and monitor process and equipment
status information (Level 1 control).
The automated control of the
anode baking process requires data
communication between each of the
ramps forming a fire group as well
as to the Level 2 Advanced firing
Compared to public Wi-Fi zones where

the wireless infrastructure mainly
provides access to the internet via
a common gateway, in an industrial
network mainly control components
(PLCs and PCs) need to communicate
to each other to exchange process
and safety data within an industrial
environment.
Since control components such as
PLCs usually do not have a built-in
wireless interface, special wireless
client modules (WCM) are required to
provide wireless connectivity of the firing
ramps. Such WCM clients have been
developed for the particular demands
of industrial control components and
industrial environments, so these
Figure 6: Firing Index Module
industrial WCMs have become the

benchmark technology around the
world.
Each local control cabinet on the
firing ramps contains a WCM. The
WCM creates the connection from the
PLC to the radio network. The WCM
transmits all data sent from the PLC
to the stationary network (backbone)
via a corresponding wireless access
point.
These wireless components require
an industrial design for mounting
inside the climate controlled electrical
control panels as well as performing
required functionality within an
industrial automation system. Due to
the daily fire move / crane handling, the
equipment (especially for the ramps of
the firing system) needs to be installed
in accordance to industrial standards
as an integral component of the firing
control system, for example:
21* Mountable on standard DIN rail
22* Heavy duty / industrial power
and signal connectors
23* Suitable for industrial control
voltage (24V DC) incl. standard
allowable tolerances
24* Vibration / shock resistant
Figure 5: Flooding Index Module
25* Suitable for industrial control

panels
26* Easy to troubleshoot, analyze
and maintain.
Intelligent Control Modules
The design objective at BSL is to
maximize the production performance
and achieve this in parallel with cleaner
baking cycles. For this purpose the
anodes must be produced with:
27* Low energy consumption
28* Low emissions, and
29* High consistency of heat
treatment.
The following intelligent control
modules have been implemented
which ensure an on-line optimization
of the process even under changing
production conditions.
Oxygen Control by
Flooding Index
With the on-line calculation of a
Flooding Index [1] the actual oxygen
levels in the firing area of the furnace
can be evaluated, especially at the
front burner ramp. Any lack of oxygen
is automatically prevented via the
Flooding Index Module as shown in
Figure 5 by dynamic limitation of the
burner capacity. Due to this module the

system ensures a complete combustion
of the gas fuel fed into the burners.
Firing Index
Changes to the fuel input as a result
of the flooding index module make it
necessary to leave the target baking
curve temporarily. In order to ensure
the same heat supply to all pits and
thus a maximum consistency, the
heat supply to the pits is recalculated
by the Firing Index Module [1] and
compensated by the control system
as shown in Figure 6.
The Firing Index Module allows even
greater deviations from the baking
curve during the pitch burn phase. Here
the burner capacity must be reduced
in order to guarantee sufficient oxygen
for the volatile combustion.
Pitch Burn Module
The Pitch Burn Module requires a
specific recognition of the oxygen
levels during the pitch burn phase.
The Pitch Burn Module incorporates the
opacity reading in the collection pipe of
the exhaust ramp and the temperature
gradients in the preheat sections, as
illustrated in Figure 7.
AWJ 2014 77
Figure 7: Pitch Burn Module
As a result of these observations

the pitch burn module dynamically
increases the draught (volume) in the
individual flues and reduces the fuel
gas consumption at the burner ramps
in a two-step strategy.
Zero Point Control
The zero point control is also one of
the essential technologies to minimize
energy consumption and maximise the
anode quality consistency. Therefore
it utilizes one controlled blower ramp,
a second measurement ramp for
sensing the pressure behind the heating
area and one uncontrolled cooling
ramp. The furnace and equipment
arrangement for zero point control is
shown in Figure 8.
Between the last burner ramp and

the controlled blower manifold all
peepholes remain closed. So the
cooling air can only leave the cooling
area via the flue gas channels into the
firing area. The preheated air enters the
firing area and can be utilised for the
firing/combustion process. For CBF4,
the air volume of the blower manifold
is produced from one central fan and
controlled for each of the nine flues
by individual motorized dampers to
an accuracy of +/- 1Pa.
Since the flue gas volume that is
necessary for the heat exchange
process is normally larger than the
necessary combustion air for the
burners in the firing area, exclusively
4
Section
Figure 8: Cooling principle with Zero Point and controlled blower ramp
Precise control of the combustion

air into the flues also minimises the
potential for excess combustion of
the packing coke in the pits. Too much
pressurised air introduced into the flues
can result in cooling air passing through
the openings in the refractory walls and
in coming in contact with the packing
coke at temperatures above 650C.
Blower manifold
(controlled)
Zero Point
Ramp
Firing Area
preheated air from the cooling area will

be used for the combustion. This heat
recuperation from the baked anodes is
one of the largest contributors for the
energy efficient operation of an open
type anode baking furnace.
Cooling manifold
1
Last filled section
CBF4 natural gas consumption [GI/mt baked anode]

2.2
2.1
2
1.9
1.8
1.7
1.6
1.5
Week 1
Week 2
Week 3
Week 4
Figure 9: CBF4 natural gas consumption
Results
Operational Safety and
System Reliability
All existing baking furnaces at BSL
have a hard-wired connection between
each of the ramps of the fire group to
connect the draught safety release
signal. This system has been in use
for many years and provides a high
level of inherent safety to the fire
control system due to the physical
connection of equipment with a lead.
The disadvantage is the need for leads
to be placed across the floor of the
furnace, and mechanical faults with
the plugs and leads as the equipment
ages.
For CBF4, a full wireless fire control
system would not satisfy the AGA
regulations. The design was adapted
to incorporate a hardwired connection
between the Measurement ramp and

the Burner ramps. For BSL, the new
design allowed the elimination of
some plug and lead operational issues,
while being compliant with the gas
regulation.
The introduction of a wireless network
communication between other ramps in
the fire group was a new development
for BSL. A second wireless network
is in use for the CBF4 building, for
communications from the furnace
cranes. Testing of the strength and
reliability of the fire control system
network was performed with the cranes
in operation and travelling to all areas
of the furnace building to simulate all
expected operating scenarios.
Gas Consumption and

Off-Gas Emissions
Commissioning of the first fire group
commenced in February and the fourth
and final fire group was completed
in mid April. By May, the furnace was
fully operational and supplying all
anodes previously supplied by the
existing closed baking furnaces. Early
gas consumption for the new furnace
was in the range of 1.9 to 2.1 GJ/mt
of baked anodes (Figure 9). Further
opportunities have been identified
to reduce consumption by tuning of
the zero point and adjustment of the
target temperature curves.
Sampling of the off-gases from the
furnace duct revealed low levels of
volatile organic and polycyclic aromatic
hydrocarbon (PAH) species, as well as
Stack CO Concentration [mg/Nm3]

40
35
30
25
20
15
10
5
0
27/05/12
29/05/12
31/05/12
02/06/12
04/06/12
Figure 10: Carbon monoxide in the off-gases from CBF4
06/06/12
08/06/12
AWJ 2014 79
Parameter
Baking level
Units
Typical Value
1,225 (40)
Electrical resistivity
O.m
57 (4.5)
Baked apparent density
kg/m3
1560 (7)
CO2 Reactivity Residue
94 (2)
CO in furnace off-gases
mg/Nm3
25 40
200 250
GJ/mt baked anode
1.9 2.1
Anode temperature at time of unload

Natural gas consumption
Table I Early CBF4 furnace and anode property data. Standard deviation
shown in brackets
carbon monoxide (CO), an effective

indicator of pitch volatile combustion.
Carbon monoxide readings (Figure
10) in the range 25-40 mg/Nm3 were
measured in the off gases during
performance testing of the furnace
and fume treatment centre.
Selection of the target temperature
curves for the new furnace was a key
contributor to early achievement of
the low natural gas consumption. In
particular, the complete combustion
of the pitch volatiles and the use of this
heat energy in pre-heating the anodes
is a key contributor to low overall natural
gas consumption. The pre-heat target
temperature curves were selected
based on the collective experiences
of the commissioning team, made up of
Innovatherm, Rio Tinto Alcan, and BSL
process engineers. Once the curves
were selected, the advanced Intelligent
Control modules of the firing control
system allowed precise and repeatable
control of the operating window around
these target temperatures.
Anode Quality
For BSL, the goal with the CBF4 startup was to commission all four of the fire
groups and achieve an acceptable initial
baked anode quality in the shortest
possible time frame. BSL were not
able to sustain the operation of three
furnaces due to pressure on green
anode supply and manning, so it was
important to have enough confidence
in the early baked anode quality to shut
down the existing closed furnaces.
A summary of the early anode quality
data for CBF4 is given in Table I. At

BSL, final anode temperatures and
pit temperature profiles are assessed
at using the equivalent temperature
technique. A small crucible of green
petroleum coke is placed in the stub
hole of the anodes in the pits, and the
crystal structure analysed after the
baking process and converted into an
equivalent temperature [2].
challenges that will no doubt arise

in the highly competitive aluminium
smelting industry.
The early results from CBF4, along

with a small-scale trial of the anodes
on the Reduction Line gave BSL the
confidence to commence a rapid
de-commissioning of the existing
closed baking furnaces. Further, the
initial results have given the plant
process engineers a basis for further
optimisation of the target temperature
curves and burner set-up to reduce
variation in temperatures within
the pit and reduce the natural gas
consumption.
Glenn Gordon; Sathya Moodley,

Boyne Smelters Limited,
Handley Drive,
Boyne Island, QLD 4680,
Australia
Summary
This paper describes the results that
can be achieved when a commissioning
team comprised of experienced start-up
and operations personnel is combined
with a bake furnace fire control system
with advanced process control
functionality. It also demonstrates
that the adoption of modern Wi-Fi
communications technology and more
stringent furnace safety requirements is
no barrier to high system availability.
BSL has entered into a new era of
anode baking capability, and is well
prepared for the present and future
Authors
Andreas Himmelreich, Detlef Maiwald;
Domenico Di Lisa,
innovatherm GmbH&Co.Kg
Am Hetgesborn 20,
Butzbach 35510, Germany
Integrated Technology
Firing and Fume Treatment for Anode Baking Furnaces
ProBake
Advanced Firing Systems
Lowest energy consumption
Total pitch burn
Higher quality consistency
ProClean
Fume Treatment Technology
Higher adsorbtion ratios
Lower emissions
Higher reliability
Your Sustainable Partner
ddilisa@innovatherm.de
www.innovatherm.de
One Design One Technology One Company

AWJ 2014 81
Upgrade of an existing Fume Treatment Plant at

ALUAR to cope with higher production in the new
open type anode baking furnaces
gas volume of 90.000 am3/h and a
maximum exhaust gas temperature
of 130 C. It was a three chambers Bag
house Filter Concept with Alumina as
adsorbent without cooling tower due
to the low exhaust gas temperatures at
the exit of the closed type furnaces.
For the higher production of anodes
in the new open type furnaces, more
exhaust gas volume is generated
at temperatures up to 220C which
implies the installation of a new FTC to
maintain the higher volumes and the
higher exhaust gas temperatures. The
specific volume per ton of produced
anodes is now in the range of 5.000
Nm3/t of baked anodes compared to
2.900 Nm3/t before.
Fig.1 Former Fume Treatment Centre at Aluar
Abstract
Aluar Aluminio Argentino has been
operating with two closed type furnaces
for the production of anodes for more
than 35 years. Due to the continuous
expansion by amperage increase in the
potlines, these furnaces were replaced
by two new open type baking furnaces,
due to present anode requirements
only one was started. As a result, the
existing Fume Treatment Plant had to
be upgraded to cope with higher volume
and temperature requirements.
This paper explains the most
economical solution for an FTC
concept upgrade, which integrates
major existing components but in
parallel enhances the plant by new
equipment and process technologies to
attain the targets. Finally it outlines the
actual results achieved by presenting
key performance figures, including
emission levels.
Introduction
The Aluar Aluminium Smelter at Puerto
Madryn (Argentina) had a production
capacity of 250.000 tons/year before
the last expansion project was initiated.
But once the ramp-up of all new pots

and the amperage increase process
in the existing ones was finished,
the production output was raised to
430.000 tons/year.
Other facilities of the Smelter
like the anode plant had to adsorb
these changes in production. Higher
aluminium production output also
enforced production of more anodes.
The green anode plant was expanded
and designed with some spare or extra
capacity for maintenance purposes.
Regarding the baking furnaces, two
of the existing three furnaces were
of closed type technology. The
anode quality requirements for the
new process conditions in the pots as
well as the difficulties to perform the
refractory maintenance routines were
causes, which drove the decision to
replace the 2 closed type furnaces by
open type technology. In addition, this
modification has lead to an increase
of the baked anode production.
The existing FTC as shown in Fig.1 had
been designed for a nominal exhaust
Targets for a solid state upgrade

The driving idea for the new FTC
design: try to keep and use as much as
possible from the existing installation
in order to get the best economical
concept for minimum CAPEX and
still reach the performance and the
environmental limits.
Further targets for a solid-state upgrade
of the FTC were:
1. Enlarge the surface of the
filter chambers to serve for the
higher capacity of 155.000 am3/h
without adding new chambers or
compartments inside the
shape of the existing baghouse.
2. Insurance of a maximum
reliability and operational safety
to run 24-7-365 without any major
interruptions.
3. High filtration efficiency of
harmful components with
low adsorbent consumption and
filtration ratios of < 1,05 m3/m2
and minute.
4. Spare capacity and redundancy
for compensation of plant aging
effects and for maintenance
works to be performed without
production stop, operation with
n-1 chambers and n-1 fans to
ensure 2.
5. A continuous high performance

to ensure anode production at
high quality with minimum OPEX
and minimum emissions.
6. High-level FTC process
automation concept in order
to guarantee an optimized and
smooth FTC operation without
any interruptions incl. automatic
switch-over to multiple modes of
operation without any interaction
of the operator.
The new FTC process technology

The new concepted FTC as shown
in Fig.2 was completely embedded
around the former FTC (baghouse). A
new cooling tower was added. Around
this, the plant was designed to provide
ideal movement of the gases from the
furnace to the stack. This is achieved
by means of calculation of optimum
main component dimensions, including
those of the cooling tower and fabric
filter, as well as the minimisation
of diversions. Pressure losses were
minimized using computer modelling
to ensure maximum flexibility even at
high baking furnace pressure losses.
The plant is operated under constant
and controlled draught conditions.
This negative pressure ensures that no
crude gas or contaminated adsorption
media are emitted to the environment.
The baking furnace off-gases are taken
from the end of the furnace building
into a new crude gas channel and then
routed to the conditioning cooling
tower.
Three advanced features are
implemented in the crude gas channel
for operational safety and reliability
(24-7-365) before the gases reach the
cooling tower inlet:
1. The crude gas channel is
designed as a special emergency
stack, including an emergency
damper at the highest point of
the gas channel.
2. A diesel generator supplied
fan (SDS=Safety Draft System)
is installed for aspiration of the
furnace flue gases in case of
Fig.2 The new FTC in operation
breakdown of the electrical

power supply.
3. A spark detection system is
installed in the crude gas channel
followed by automatically
operated extinguishing ring
nozzles.
In case of breakdown of electrical and
air supply, the FTC will automatically
switch over to the status Natural Draft
and together with the SDS-System to
the Mode Forced Natural Draft. This
mode will keep the furnace building
free of pollutants; further movement
of the furnace equipment will be
possible.
The spark detection and extinguishing
system operates independently from all
other systems and ensures that sparks
emitted from the bake furnace will be
automatically extinguished without
any interruption of FTC operation as a
preventive fire protection system.
Cooling tower
The conditioning cooling tower is
designed as a direct current cooler.
The off-gases are guided to the top of

the cooling tower (gas inlet) and leave
the cooling tower at its base. A special
design ensures that the off-gases are
treated with atomised water produced
by a two-phase nozzle technology in
an area where plug-flow is reached.
The cooling tower design is optimized
with regards to the following tasks:
1. Adequate evaporation distance
2. Even plug-flow within the
evaporation distance
3. No deposit build-up at the inner
surface of the cooling tower
4. Dry condition at cooling tower
hopper - outlet
5. Dry discharging of heavy particles
at tower hopper
These tasks are verified and optimized
by a CFD study of the cooling tower
design, as shown in Fig. 3.
The off-gas temperature at the
conditioning tower outlet is maintained
constant at 0.5C by automatic control
of the water flow injection. The towers
conditioning system is provided with
AWJ 2014 83
full redundancy and automatic handling

of breakdowns. This allows lances
and nozzles to be inspected and/or
maintained without interruption or
setting the FTP to bypass mode.
Fig.3 CFD Study Cooling Tower
A final, special surface coating protects

the tower against corrosion, acid attack
and deposits. Deposits and particles
such as packing material from the
baking furnace have already been
extracted in the lower part of the tower
by means of a cyclone and a double
pendulum damper. Fig. 4 shows the
new cooling tower in operation.
The performance of the cooling tower
is also dependent on the quality of
evaporation. The lance system provides
a dual-phase nozzle technology, which
generates finest droplets of less then
25 m. Figure 5 shows such a spray
nozzle system, consisting of 5 nozzles
per lance.
Fig.4 The new Cooling Tower in operation
the alumina silo, is directly injected

after the cooling tower and entry of the
main reactor duct by a special alumina
injector. The injection of fresh alumina
early after the cooling tower is a key
parameter for the performance of
the FTC, especially for the bonding
of preliminary phases of acids.
Fig. 5 Dual phase spray nozzle system
The pre-cleaned and conditioned offgases are channelled through the lower
side conical outlet into the main reactor
chamber. Fresh alumina, sourced from
The enriched off-gases are then

guided into individual secondary
reactor chambers and finally to the
fabric filters. The secondary reactor
chambers are also charged using the
alumina injector. This injector is fed with

recirculated secondary alumina from
the individual fabric filter chambers. As
a result, the recirculation takes place
at high concentrations of alumina,
resulting in high adsorption rates for all
kinds of pollutants. Finally due to the
long sections of the reactor chambers
and intensive mixing of the adsorbent
media, the aerosols not only adsorb
off-gases but fines in the media also
absorb them and both are extracted
at the fabric filter cloth as filtration
cake.
As a secondary benefit, the consumption

of primary alumina is very low (< 800
kg/h) and still achieves a maximum offgas efficiency in terms of HF, organics
and other acids.
Duct systems and Dampers

The duct system guides the gases
through the FTC. The existing Duct
Systems and Dampers had to be
checked carefully. Non-optimized
layouts with multiple bends and
elbows lead to high pressure losses
(hpl). The same applies for special
types of dampers. A proper design
of the duct system is the basic task to
minimize these pressures losses. Also
the change of a damper design can
improve the performance of the FTC
tremendously. For the overall system,
consisting of ducts, dampers and filter
chambers, a compromise between
low-pressure loss and longest dwell
period of alumina enriched with flue
gas had to be found. Figure 6 shows
the optimized lpl design to optimize
flows and minimize pressure losses.
Filter chambers
The adaptation of the existing filter
chambers is strongly depending on the
existing infrastructure and the existing
design. The process requirements for
a redesign of an existing filter chamber
was easy to formulate, but not easy
to realize. The design target was the
enlargement of the fabric surface in the
space of the existing filter chambers.
An additional chamber was no option
due to economical aspects. The future
value of the filter surface load should
be favourable in a range of < 1,0 m3/
m2 and minute. High gas flow rates
in a range of 18 m/s in the ducts and
reactors have to be reduced in steps
to less than 1 m/s in the area of the
filter bags. The separation of heavy
particles (agglomerates) needs to be
executed prior to the filtration cake.
In this area gas speeds of < 2m/s are
obligatory.
new pre-separation chamber was

designed and installed in front of each
filter chamber to ensure the staggered
deceleration of the gas speed.
It would have also been possible to
raise the filter chambers and to prolong
the filter bags. But the maximum
length is limited by the efficiency of
the pulse jet cleaning system and the
maximum physical load on the filter
bags during the cleaning cycle. Filter
bags up to 6m are feasible, longer
bags need to be examined with care.
Figure 7 shows the inner part of the
modified filter chamber.
For daily operation and fast maintenance
of the filter bags two main topics have
to be considered:
The gas flow onto the filter bags has to

be oriented in a wide area from bottom
to top.
Fig. 7 Inner part of the filter chamber
Fig. 6 Optimized lpl design of the duct system
In order to ensure a good alumina

transport without sedimentation within
the ducts, a minimum flue gas speed of
> 12 m/s is necessary. Especially during
different phases of FTC operation with
only one or two furnaces in operation,
the minimum flue gas speed has to
be maintained. For these dynamic
conditions an automatic controlled flue
gas recirculation system was designed
and implemented.
Technical solutions, which contain

a horizontal flow, generate an early
wear and tear of the fabrics by partial
overloads and in parallel inactive areas
inside the filter chambers.
In a first approach all disturbing
installations had been eliminated.
After that the filter chamber was fully
furnished with a maximum possible
amount of filter bags. In addition a
1. The Production must be possible

even with n-1 chambers in operation, meaning with 1 chamber
isolated.
2. The change of one complete
set of filter bags for one chamber
should be possible during one
production shift.
Consequences for the fans

The existing fans had a power of 160
kW each. To ensure the desired flow,
3 new fans with a power of 200 kW
each for nominal and 1 spare fan for
extra capacity had to be installed to
reach the necessary volume flow
and draft with enough spare power
for compensation of future plant
AWJ 2014 85
Fig. 8 Fan system in operation
aging effects. An operation mode of

n-1 is implemented, because every
malfunction of a fan would disturb the
furnace production immediately.
Systematic maintenance incl. turndown of fans is mandatory to reach the
desired performance and reliability.
Figure 8 shows the new fan system
in operation.
Conclusion
After all these aspects have been
technically realized, the desired
performance for higher production
was available. As a positive effect the
dosing of fresh alumina after finetuning was minimized to 800kg/h;
this corresponds with a fresh alumina
load of 7,8g/Nm3 flue gas.
If the right balance is found between
fresh and recirculated alumina, the
wear and tear in all aspects of the FTC
will be minimized in the long term

and the adsorption ratio, respectively
the cleaning effect is maximized.
This leads to minimum emissions in
the clean gases, which are vented to
atmosphere.
Operational results
The fluorides and total dust values
were obtained following the US EPA13a
method. Each monthly value was
obtained as an average of two or three
individual values.
The condensed soluble tars values were
obtained following EPA429 method
and correspond with international
standard test measurements.
The total dust concentration obtained
by this method during these measurements was 0.54 mg/Nm3, which is
equivalent with the monthly average
obtained by the other method.
The very low condensed soluble tar

values (< 0,008 mg/Nm3) correspond
with the design criterias but also with
an optimum pre-process condition for
the FTC, i.e. the excellent adjusted
interactions between furnace, firing
system and a professional baking
process conduction through the carbon
team.
Summary
The results presented above show
impressively that the upgraded
FTC technology even in the given
boundary conditions can achieve or
surpass emission values stipulated
by international regulations by far.
The project was executed in time
and in budget which finally assured
the technological and economical
benefits for the carbon plant of ALUAR
Aluminium.
For gaseous fluorides the following emission values were

measured:
Fig. 9 Measurement of gaseous fluorides.
For the particulate fluorides the following emission values were

measured:
Fig. 10 Measurement of particulate fluorides.
Total dust at stack measured and is showing values far below the
environmental limits:
The emissions of condensed soluble tars is shown in the following

figure:
Fig. 11 Measurement of total dust .
Fig. 12 Measurement of tars:
Authors
E. Cobo; L. Beltramino; J. Artola,
Aluar Aluminio Argentino
Dr.-Ing. F. Heinke; Dipl.-Ing. D. Maiwald;
D. Di Lisa;
Innovatherm
Prof.-Dr. Leisenberg GmbH & Co KG,
Butzbach, Germany
AWJ 2014 87
Trusted by more smelters

than any other.

For over forty years Alesa has been supplying
aluminium smelters with systems to handle and
convey bulkmaterials such asfresh and reacted
alumina, crushed bath, coke and fluoride.

Our knowledge goes beyond materials handling.
Being an aluminium producer we understand
how oursystems impact yourprocess. Alesa is a
full servicetechnologies supplier thatalso offers
automation,pot controls systemsand
reductiontechnologies.
#(("'"#"
$#(" ($!"
!$)(!!(
)"##"
("#!("'"#"
!$!(#"
) #'( $#(
%')"'"#"(
(

(
(
RTA Alesa
&&&!#"
((((
TECHNOLOGY
8 88 ANODE
GLOBALPLANT
ISSUES
MATERIALS, HANDLING & TRANSPORTATION

Vigan Engineering
Pneumatic unloaders, a most convenient way for handling alumina
p. 90-91
Siwertell
Siwertell unloaders offer industry changing advantages for
p. 92-95
alumina handling
AWJ 2014 89
VIGAN pneumatic unloaders, a most convenient

way for handling alumina
Pneumatic suction of alumina by VIGAN continuous ship unloader
The trade of alumina remains strong

and very competitive: therefore
port operators continue to focus on
achieving low unloading costs as well
as minimizing the environmental
impact which is becoming an evermore
important topic.
The very free flowing and abrasive
characteristics of this white powder
make its handling a rather challenging
goal in order to achieve these two main
targets.
Most frequently, alumina is unloaded
from sea vessels into a specially
designed conveying system for
transport to storage facilities, or into
a hopper from which it is loaded into
trucks, but it can also be directly packed
on the quays, into 1 or 2 ton big bags for
instance, to be transported to smelting
plants later on.
Although the traditional unloading
method is done by grabs and simple
hoppers on the quay, this is not the
most effective one, namely due to
dust pollution and spillage:
Dust pollution is a serious problem,

mainly in urban areas. Any moderate
breeze will blow away alumina, thus
contaminating the air and surrounding
areas which must then be continuously
cleaned.
Spillage represents a material loss of
0.2 to 0.5 %. Considering that generally large quantities of this valuable
product are conveyed, spillage adds
to the costs incurred in unloading operations. The use of hoppers with dust
suppression devices such as suction
fans and air filters is feasible, but the
investment cost involved can be relatively high. Furthermore the time spent
for completing each complete cycle
of the grabs can significantly reduce
the average unloaded yield.
Pneumatic unloading: a proven

method with up-to-date
technology
Pneumatic unloading is the most
suitable way for handling alumina:
its free flowing characteristics make
its suctioning very easy, and because
90 MATERIALS HANDLING AND TRANSPORTATION
the product is conveyed in a totally

enclosed system, the issues of
pollution by escaping dust and spillage
are avoided.
VIGAN expertise in alumina pneumatic
unloading include:
Optimized design of the suction line
with unique high wear-resistant alloys;
Specific know-how for separating the
alumina from the conveying air without
any airlock, which provide maintenancefree operation;
State of the art filter design and filtering materials to guarantee dust-free
air exhaust.
Self-propelled on rails or on rubber
wheels for instance, VIGAN pneumatic
unloaders are able to move with ease
along the ship and no additional
equipment (such as hoppers) is
required for ensuring quick and reliable
unloading operations.
The direct drive is another major

improvement because it reduces not
only the number of bearings compared
with a traditional drive of belts, but also
the mechanical wear on the turbine
shaft.
Launched in the same time period as
the turbine control with inverters, it has
become a standard device for all large
size VIGAN pneumatic CSUs.
Continuous monitoring of the
performance and operating parameters
of the machines has revealed the
power consumption of this type of
equipment is quite comparable with
the method using grabs for instance,
especially when the average unloading
performance and the directly related
operating costs or consecutive savings
are taken into consideration.
Optimum alumina unloading

equipment choice
Old and new designed TURBO blower assembly with direct drive.
A most frequently commented aspect

of pneumatic CSUs in comparison with
the traditional method performed
with grabs is the power consumption
required.
However, most recent technological
developments such as the installation
of frequency inverters and the direct
drive of the turbo blowers, both
duly correlated with an optimized
design of the suction nozzle, are
consequently enabling much lower
power consumption than in the past.
In comparison with 12 to 15 years ago,
power consumption of pneumatic
equipment in use for unloading
agribulks has decreased from 0.9 to 1.0
kWh/t down to 0.6 to 0.8 kWh/t.; and
similar power consumption decreases
have been measured for alumina.
Frequency inverters (also called
speed variators) allow continuous and
precise control of the suction pattern.
The air flow is now controlled by the
inverter.
Air over flow is regulated in order to

limit the power absorbed by the main
electrical motor.
The power delivered to the blower by
the electrical motor is automatically
controlled with the inverter by limiting
the torque at a maximal value. When
only air is sucked, the torque limitation
will automatically reduce the electrical
motor velocity.
Also when the suction is stopped by
the operator or by the interlock system
of the storage system, the inverter will
stop the blower, delivering additional
power saving during downtime.
Inverter systems installed on VIGAN
equipment have been proven to add
25% to 30% power saving.
For more than 12 years VIGAN has
installed frequency inverters on
pneumatic ship unloaders with an
in-house expertise for optimized
calibration.
Indeed, it is a real challenge for many

executives in the alumina sector.
A cost effective pneumatic technology
for unloading remains the best
performing alternative for many
ports and the alumina processing
industry.
VIGAN team commitment with about
25 years expertise in various countries
such as Finland, France, Poland, China
and Iran, is to supply its customers
with the best expertise by providing
the most suitable solutions for their
requirements.
Contact Details:
Vigan Engineering s.a.
Rue de lIndustrie 16
1400 Nivelles, Belgium
Phone: + 32 67 89 50 41
E-mail: info@vigan.com
Web: www.vigan.com
AWJ 2014 91
Siwertell unloaders offer industry changing advantages

for alumina handling
Alumina unloading is one

of the few industrial areas
which have not been touched
by modern advances, and
Cargotec believes that it
would benefit from the
substantial advantages
inherent in its Siwertell screwtype unloaders
There is always one market where
you know that you could make a real
difference, and for us that market is
alumina unloading, says Per Karlsson,
Managing Director of Siwertell bulk
handling solutions at Cargotec. While
we are well known in alumina loading
with numerous successful references,
we have not yet established ourselves
as the first choice for unloading equipment; pneumatic and grab systems
are currently favoured by the industry,
but both of these technologies have
major drawbacks, such as high power
consumption and spillage.
We know that our screw-type unloader

is ideal for discharging alumina and will
deliver huge cost savings. However,
the industry is resistant to change.
Therefore, to prove without doubt that
we can offer the best system, we are
very keen to perform a full scale test
with one of our mobile unloaders to
demonstrate how incredibly effective and environmentally-friendly our
system is.
We are so confident of our claims that
we will undertake one of these tests
for a potential client later this year and
demonstrate how our mobile unloader
discharges a vessel. We will do this,
free of any charge. Perhaps there are
others that would also like to see a
demonstration? I hope that someone
reading this article will take the lead. I
know they will not be disappointed.
Approximately two tonnes of alumina
is needed to produce one tonne of aluminium. It poses specific and complex
handling problems due to its highlyabrasive and extremely free-flowing
properties. This means that unless the

ports unloading equipment is totally
enclosed, it will face problems with
spillage and dust.
Alumina is a very expensive material,
currently about US$400/tonne and
therefore any loss of material should
be avoided as it has a huge impact on
the actual running cost, explains Mr
Karlsson. For example, a port that
imports 1 million tonnes of alumina
per year and uses a grab crane will
lose about 1.5 percent of material
due to spillage; the value of the lost
material will be approximately US$6
million each year that this operational
set-up exists. Based on a typical ship
unloading equipment lifetime, about
20 years, the total loss of the operation will be around US$120 million.
A figure of this magnitude should be
impossible to ignore.
A totally enclosed Siwertell screw
type unloader completely eliminates
spillage. Lifetime cost savings of this
magnitude easily justify the initial in-
vestment in a Siwertell unit, notes Mr

Karlsson. In addition to these substantial cost savings, dust creation is
reduced to a minimum, providing a
healthier working environment and
avoiding conflict with the increasing
levels of environmental regulation.
While pneumatic unloading systems
do not suffer from spillage, they have
other characteristics which make them
far from ideal for handling alumina.
The high velocity inside the conveying
tubes leads to high rates of wear on
the conveying line. More importantly,
high speed conveying results in a high
degree of crushing for the material
being transported. While this may not
seem to be a particularly significant
problem, it is in the case of alumina
because of its effect on the smelting
process.
The high conveying speed and consequent crushing cause approximately
5-6 percent degradation of the alumina. This increased percentage of
fines in the alumina has a negative
impact on the already energy-intensive
aluminium smelting process, as fine

alumina requires significantly more
energy to process. Therefore, use of
a pneumatic unloader, which itself
has a high power consumption, also
incurs significant additional costs further down the line in the production of
aluminium. Mr Karlsson says this can
be as much as a factor of three or four
compared to processing undamaged
raw material. To put a figure on this,
if the port is still unloading about 1
million tonnes of material a year and
the price per kWh is about US$0.15,
the cost for the additional energy will
be in the range of US$1.2-1.5 million
per annum.
Given that using a Siwertell screw-type
ship unloader for handling alumina will
address the problems of spillage and
crushing, along with the substantial
associated costs, it is hard to understand why the industry has not eagerly
taken up the technology. Mr Karlsson
believes that this reluctance may have
a lot to do with peoples difficulties in
accepting that a screw-type unloader
can have such a gentle touch on the

material passing through it.
Everyone can see that there is no
spillage and hardly any dust; that is not
an issue. And the quiet, economical
operation is very easy to verify. I believe
the sticking point for many operators
is the issue of cargo degradation.
Intuitively they feel that a screw-type
unloader must be harsh on the cargo.
While intuition certainly has its place in
business, it should be rejected when
the facts and figures clearly show that
it is wrong. And we have overwhelming
evidence to prove our claims in the
context of alumina unloading.
The cargo-friendly nature of Siwertell
screw-type unloaders and conveyors
has been repeatedly proven both by
both full-scale tests and testimonials
from satisfied customers throughout
the brands 40 years experience of
bulk handling.
Tests include those carried out over a
ten-year period in Bosnia-Herzegovina
at Aluminij Mostar d.o.os alumina in-
AWJ 2014 93
take at Port of Ploe in Croatia, using

a Siwertell 5000 S-GT unloader. The
results provide clear evidence of cargo
degradation at less than 0.2 percent. A
full-scale test using a Siwertell 640-D
at Fremantle, Australia in December
2006 controlled by SGS confirms negligible degradation of less than 0.1
percent. In contrast, the degradation
typically experienced by operators
using pneumatic unloaders is in the
region of 5-6 percent.
Siwertell is continuously developing

its technology to further improve efficiency and reduce operational costs;
however current performance is already
extremely impressive. Siwertell has
undertaken performance tests where
the unloader has exceeded 75 percent
through-ship efficiency. Siwertell offers guaranteed lifetimes for unloaders handling abrasive materials of up
to 10,000 hours for Siwertell screws
and tubes.
In conclusion, Mr Karlsson says he believes that by offering alumina handling

operators guarantees relating to material breakage, power consumption and
spillage, along with lifetime warranties,
Siwertell will take alumina handling
to a new level that will generate huge
cost savings for those with the vision
to put the past behind them and adopt
the optimum technology.
Cost-efficiency comparison: Siwertell unloader and pneumatic unloader
Per Karlsson
We know how to load and unload
alumina, and our units meet all the
known regulations related to safety
and the environment, says Mr Karlsson. Over the years Siwertell screwtype ship unloaders have become the
industry-standard for handling many
kinds of dry bulk materials; 400 installations in operation worldwide speak
volumes about the quality, reliability,
cost effectiveness and minimal environmental impact of our technology,
along with our maintenance and service
offerings.
Yearly intake
One million tonnes
Operating hours/day
20 hours
Ship size
65,000 dwt
Ship cost/day
US$25,000
Rated capacity, pneumatic
600t/h
Efficiency, pneumatic
55 percent
Rated capacity, Siwertell
1,000t/h
Efficiency, Siwertell
70 percent
Unloading days, pneumatic
152 days
Ship cost, pneumatic
US$3.8 million/year
Unloading days, Siwertell
72 days
Ship cost, Siwertell
US$1.8 million/year
A saving of US$2 million per year is possible in shipping costs alone using a highly
efficient Siwertell unloader.
Contact Details:
Email: blksales@cargotec.com
Tel: +46 4285800
Website: www.siwertell.com
Say goodbye
to crushing,
dust, spills,
and additional
fines in
alumina
unloading
Take advantage of a free
demonstration at your
facility and let us show you
what advanced alumina
handling really looks like.
There is nothing to lose and
so much to gain. We are
waiting to hear from you.
blksales@cargotec.com
www.siwertell.com
1974
40 2014
AWJ 2014
YEARS
95
Stressometer Systems. From hot to cold.
The long standing problem of flatness measurement and control in aluminum hot
rolling mills is now solved by ABB. Achievements in the field show yield improvements
of 2 % and rolling speed improvements of 10%. Altogether this corresponds to
millions of dollars on the bottom line. One reason for this is, as one of our customers
put it, Good tail-out from the hot rolling mill results in high quality head-in to the cold
rolling mill.
The range of applications where the Stressometer system sustainably improves
your business includes all kinds of cold rolling mills; and now also aluminium hot
rolling mills. www.abb.com/pressductor
ABB AB
Force Measurement
Phone: +46 21 32 50 00
96 COMPANY
MATERIALSPROFILES
HANDLING AND TRANSPORTATION
COMPANY PROFILES
ABB
Making your processes measure up
p. 98-99
Hycast
Game changer casthouse solutions
p. 100-102
AWJ 2014 97
Making your processes measure up
We are helping thousands of clients all over the world to boost their productivity and yield
Stressometer Flatness Control System installed in an aluminium rolling mill.
ABB Measurement Products

Force Measurement
Vasteras, Sweden
At the heart of ABB technology is the
understanding that flatness, tension,
pressure, position and dimension can
be sensed accurately, reliably and repeatedly on a continuous basis. The
data generated by such sensing devices can then be used to control external equipment in such a way that
process parameters are kept constant.
And as a result, operators can increase
productivity and achieve higher levels
of consistency in product quality. Our
products not only measure the forces within a process, they help make
sure that production measures up to
expectations. Using state-of-the-art
technology, ABB provides purpose
built solutions for your force and dimension measurement needs. Making
it possible for your production output
to accurately match the most varying
and demanding requirements.
Challenge the hidden potential within
your application with the leading measurement technologies:
Pressductor Technology
ABBs well-known Pressductor
Technology is a measurement principle based on the magneto-elastic
effect the magnetic properties of a
metal are influenced by the mechanical force applied to it. Because the
signals produced are not reliant upon
physical movement or deformation,
the load cells combine sensitivity with
extraordinary tolerance to overloads
and virtually no built-in limit to the
number of load cycles.
ABBs Pressductor transducer stands
for unbeatable load cell performance,
thanks to its unique combination of
accuracy, overload capacity and ability
to withstand harsh environments. By
using this technology you will achieve
higher quality and reliability, especially
under demanding conditions.
Pulsed Eddy Current Technology
ABB has developed a completely new
way of performing measurements
with eddy current technology. It is
a method that makes it possible to
measure, in real time and in line, dimensions and other attributes with
exceptional accuracy.
98 MATERIALS
COMPANY PROFILES
HANDLING AND TRANSPORTATION
The new ABB technology is based on

measurements of the voltage pulse
induced in the coil when the current
is suddenly interrupted. By measuring this value at three different times
three parameters can be derived; the
distance, the electrical resistivity and
the thickness.
Since this is a non-contact technology it eliminates the drawbacks of
common systems based on X-ray or
isotope contact measurement.
Flatness Measurement & Control

Based upon our experience from more
than 1200 installations of flatness
measurements and flatness control
systems worldwide we continuously
develop new generations of the market leading flatness system the
Stressometer System FSA.
This system will not only make your
mill more competitive today, it will
also provide you with tools and methods to ensure that you keep in front of
your competition in the future. Internet
technology gives you a non-proprietary platform - independent way of
ABB Millmate Thickness Gauging System installed in an aluminium rolling mill.
expanding the system functionality

and the CPU power when your needs
are growing.
Advanced control methods, such as
the Extended Singular Value Decomposition (ESVD), bring rewards to our
customers exceeding 100,000 USD
per year.
Some of the benefits with the Stressometer System are:
Improved yield out of each coil

Improved strip quality
Shorter production cycles
Reduction of number of strip
breaks
Reduction of cost for process
development
Reduction of cost for
maintenance
Lab accuracy in the mill
The Millmate Thickness Gauging
Systems (MTG), utilizing the Pulsed
Eddy Current Technology, open up a
new dimension in metal strip gauging
with superior features:
Millmate Strip Scanner System installed in a rolling mill,
Contact-free and yet

material independent
Gauging for non-ferrous
metals
Robust and completely
insensitive for conditions
in the measuring gap,
such as oil, water, coolants,
steam, etc.
Accuracy to a level of 0.1%
Poses none of the risks
associated with X-ray or
isotope gauges
Measures as accurately
in production as in the
laboratory
Modern rolling mills often use a variety of measurement and control

facilities. One of the most important
parameters is the position of the strip.
If this is not correctly measured, there
is a great risk of rolling the strip under
incorrect presumptions. This leads to
poor flatness, especially at the edges,
and to a final product of inferior quality
or even strip breakage. In addition to
this, the strip width must be kept within
the specified tolerances essential for
good economy of production.
A non-contact, non-optical
measurement system
The Millmate Strip Scanner System
determines the edge position of a
metal strip in a rolling mill down to
millimeter resolution. High, consistent
quality is always the aim when producing steel, aluminum or copper strip.
This is equally true in rolling mills and
process lines.
AWJ 2014 99
100 COMPANY PROFILES
Hycast
Game Changer Casthouse Solutions
In 1990 Hycast AS was established by

Hydro to be a commercial manufacturer
and market organisation for casthouse
technology developed in-house at
Hydros R&D centre at Sunndalsra.
The first product was an inline degasser
but shortly after casting technology
for extrusion ingots, and stations for
pot line metal fluxing followed. Most
of the Hycast products have been
captive during these two decades.
Today the Hycast product portfolio covers
the whole casting centre and includes;
RAM Removal of Alkali Metal.

Crucible treatment for alkali metal
removal, the removal of alkaline metal
is carried out by the introduction of
aluminium fluoride (AlF 3) to the
liquid aluminium through a rotor
using argon as a transport medium.
The RAM system is a customized
system that is tailor-made for the
tapping system on the smelter.
Hycast RAM - characteristics and

results
High removal efficiency for alkaline
metal and carbides
Environmentally friendly - no need
for chlorine in the cast house
process line
Hycast SIR - characteristics and

results
Excellent removal of Hydrogen
and inclusions (especially Oxides
and Carbides) - especially important in locations with a high
humidity
Mould carriage can be customized

for extrusion ingot as well as
rolling slab casting
State of the art internally guided
stroke cylinder
Breakaway-torque limiter for
cylinder guide protection
For extrusion ingot lines, no

mechanical filtration system is
required downstream the I-60 SIR
Easy access to critical components,

e.g. torque limiter
Exceptionally low dross generation

due to inert atmosphere in the
reactor
Platen with open steel frame,

industry-proven safety system in
an automated system
Low operational cost - few rotors

and very low Argon consumption
Fully automated casting sequence

obtainable in combination with
Hycast CCS
Fully automated - environmental

and operator friendly
Small footprint in terms of floor
space requirements
Unique and patented melt refining
concept
Drain free reactor unit
Available with Chlorine for improved Alkali removal efficiency
CMV Casting Machine Vertical.
GC Gas Cushion / LPC

Low Pressure Casting,
extrusion ingot casting technologies.
GC/LPC
The Hycast GC (Gas Cushion)

Billet Casting Technology is ideal for
production of high quality extrusion
billets with excellent surface quality
and extrudability. The system is
proven worldwide over more than
two decades in casthouses owned by
Hydro Aluminium and its partners.
Hycast LPC (Low Pressure Casting)
Billet Casting Technology is ideal for
the production of larger diameters
and hard alloys.
Optimized furnace cycle time due

to minimized or eliminated infurnace melt treatment
Hycast CMV internally guided casting

cylinder and mould carriage system.
The Hycast CMV comprises of a
vertical casting machine with a single
acting internally guided hydraulic
casting cylinder for extrusion ingot and
sheet ingot casting. The mould carriage
system is supplied as a tilt frame on a
fixed foundation, or a rollaway mould
carriage system with or without an
integrated tilt-frame.
Patented rotor design and injection

system
Hycast CMV - characteristics and

results.
Other diameter available upon

request
Founded on solutions proven to be

successful in full-scale operation
Patented dual graphite rings for

optimal oil and gas distribution
Mould carriage tailor-made to

customer-specific requirements
High pit utilization due to optimized mould arrangement
Reduced inclusion level in the potline metal transferred to the

casting furnace
Dross formation in casting furnace
is reduced due to low Na content
SIR Inline Melt Refining.

The Hycast SIR sets the benchmark
for the in-line refining of molten
aluminium at casting centers.
Hycast GC / LPC - characteristics

and results
GC diameter range 127mm 405mm (5 - 16)
LPC diameter range 127mm 735mm (5 - 29)
AWJ 2014 101
Hycast
Game Changer Casthouse Solutions
High productivity, pit recovery,

excellent and consistent surface
quality
stantly achieve better quality at lower

operation cost and thereby increases
the competiveness of its customers.
A fully automated casting sequence can be obtained by combining

the Hycast Billet Casting Technologies with the Hycast CCS
Technology and process development
AFM Adjustable Flexible Moulds / FM Flexible Moulds, sheet ingot casting

technologies.
Complete casthouse support for

customers
AFM/FM
The Hycast AFM (Adjustable

Flexible Mould) Rolling Slab Casting
Technology is developed to reduce the
number of moulds needed to produce
ingots with optimized geometry and
surface quality. Hycast FM (Flexible
moulds) have been used within Hydro
Aluminium for more than a decade.
Flexible moulds with fixed width and
moulds with fixed geometry may also
be provided an option for smaller
dimensions.
Hycast AFM - characteristics and

results
Geometry range: Thickness
330mm -700mm, Width 1100mm2500mm
Adjustable range: 300mm -400mm
Unique and patented flexible mould technology to ensure optimal
ingot geometry from start to end,
and for varying casting speeds
High productivity, pit recovery and
excellent and consistent surface
quality
A fully automated casting sequence can be obtained by combining
the Hycast AFM with the Hycast
CCS and the Metal Level Control
System
Hycast Service Knowledge and

Competence.
Hycast supports customers to con-
102 COMPANY PROFILES
Project management, engineering

(mechanical and electrical/automation)
Close link to Hydro Aluminiums
R&D facilities, including Reference
Centre
Cooperation with NTNU and SINTEF in Trondheim, Norway
The business model of Hycast is to be
an engineering company for aluminium
casthouse projects. Manufacturing is
done by a limited number of qualified
suppliers. Innovation is a highly prioritized task and new product development is done in close collaboration with
Hydros R&D and production systems,
technology partners, customers and
suppliers. The core competence of
Hycast is actually multi-disciplinary.
Aluminium metallurgy knowledge
is the basis for all our products and
services but process understanding,
mechanical and electrical engineering, automation and project execution
are equally important competences at
Hycast. Over the years a substantial
number of small and large projects
have been completed in most corners
of the world. HES have been the backbone of Hycast from day one. Since
the start in 1990, no Hycast employee
has been injured at work! No fatality related to the operation of Hycast
equipment has ever been reported.
Hycast was the first to deliver a technology portfolio that eliminated the
need for Chlorine in the cast house.
This was made possible by combining
the AIF3 pot line fluxing (RAM) with
the SIR inline melt refining unit. The
casting processes delivered by Hycast
are highly automated and minimize
the need for operators to be in close

proximity to the casting pit during all
phases of the cast. The Hycast Safety
Philosophy is embedded in the automation system, including hardwiring
of all emergency functions.
In the past Hycast has introduced new
technologies to the market. In 2012,
Hycast introduced the Adjustable
Flexible Moulds (AFM) for sheet ingot
casting. The new AFM technology is
based on the proven Hycast flexible
mould technology a unique and
patented technology that ensures
optimal sheet ingot geometry from
start to end, and for varying casting
speeds. Additionally the AFM reduces
the number of mould sets needed in
the casthouse as the technology has
an adjustable range of up to 400mm
depending on the mould width. During
TMS 2014, Hycast presented the Low
Pressure Casting (LPC) technology for
extrusion ingot casting; the foundation
of this new technology is based on
proven Hycast GC technology. LPC is
a new direct chill casting technology
for applications where surface quality
matters. Its also ideal for production
of larger diameters and hard alloys.
The LPC provides a new standard
in surface quality as well as thinner
inverse segregation zone compared
to other technologies.
Service and support is a challenge
when working globally. To meet this
challenge Hycast is staffed with highly
skilled, experienced and mobile service
engineers. In addition, in many cases
customer assistance may be done from
the Hycast office by accessing the cast
house systems online for investigations
and trouble-shooting. Hycast supports
customers to constantly achieve better
quality at lower operation cost and
thereby increases the competiveness
of its customers.
For more information see;
www.hycast.no
Contact:
hycast@hydro.com
Some things live forever. Like the famous photo of

Marilyn Monroe. Like aluminium, which can be
recycled and live on in new products. And like Hydro,
which has been renewing itself for more than a century.
www.hydro.com
AWJ 2014 103
Advertisers and Web Index
PAGE No.
COMPANY
WEB ADDRESS
ABB AB Force Measurement
www.abb.com/measurement
96
ABB Switzerland Ltd.
www.abb.com/aluminium
Alcoa Inc.
www.alcoa.com
1/4
Cargotec/Siwertell
www.siwertell.com
95
ECL
www.ecl.fr
20
Fives
www.fivesgroup.com
FLSmidth
www.flsmidth.com
49
Hycast A/S
www.hycast.no
100
66/103
60/106
Hydro
www.hydro.com
Innovatherm
www.innovatherm.de
81
Power Jacks
www.powerjacks.com
56
Rio Tinto Alcan
www.riotintoalcan.com
18
RTA AP-Technology
www.ap-technology.com
46
RTA Alesa Ltd.
www.rta-alesa.com
88
Sensotech
www.sensotech.com
52
TMEIC
www.tmeic.com
UC Rusal
www.rusal.ru/en/
Vigan
www.vigan.com
104
ADVERTISERS
COMPANY PROFILES
AND WEB INDEX
8
14
90/105
:PVS1BSUOFSJOCVMLIBOEMJOH
UPOTIPVS
*.".,)0.&*/**SBO
UPOTIPVS
-*"/:6/("/($IJOB
UPOTIPVS
(%:/*"1PMBOE
'SPNJOJUJBMQSPKFDUEFTJHOUPDPNQMFUFUVSOLFZ
JOTUBMMBUJPOTXJUINFDIBOJDBMBOEPSQOFVNBUJD
NPTUSFMJBCMFBOEQSPUBCMFTPMVUJPOT
7*("/&OHJOFFSJOHTB
3VFEFM*OEVTUSJF
/JWFMMFT #FMHJVN
1IPOF
'BY
*OUFSOFUXXXWJHBODPN
&NBJMJOGP!WJHBODPN
3*;)"0$IJOB
UPOTIPVS
1/&6."5*$6/-0"%&34'03"-6.*/"
AWJ 2014 105

71
ANODE PLANT TECHNOLOGY

Aluminiumworldjournal2014 141022112624 Conversion Gate02 PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aluminiumworldjournal2014 141022112624 Conversion Gate02 PDF

Uploaded by

Copyright:

Available Formats

Global Media Communication Ltd.

Charles Martin Hall

Maximize your return on investment?

Global Competence Center Aluminium

Aluminium World Journal 2014

Foreword By Christopher Fitcher-Harris

Charles Martin Hall had a purpose to his life.

Mr. Hall revealed that probably his

on his experiments, then asked others

himself in the woodshed, combining

Alcoas lightweight aluminum helped

tion process had scientists the world

To make his efforts profitable, Hall

Halls process is a new discovery. It

He used to read the Encyclopedia

Learning From The Best

The New Kensington office building of Pittsburgh Reduction Company. (1891)

Authors: Investors Business Daily

Primary Smelting and Processes

Anode Plant Technology

Materials Handling And Transportation

Advertiser and Web Index

TMEIC Serving the Aluminum Industry

15 ALUMINUM MILLS AUTOMATED

TMEIC Serving the Aluminum Industry

quality and most of all, complete coil

TMEIC the Company

Advanced technology, excellence in

Recent Aluminum Projects

through the mill. Success is measured

require very close cooperation

solutions. Detailed discussions are

Mill Control System

First Coil from Mill

TMEICs AC main drive motors are

TMEIC Serving the Aluminum Industry

TMEIC Main Rolling Mill Motor

TMdrive-70e2 Variable Frequency Drive

TMEIC has a global network of offices

Our modern control systems include:

Inner-stand tension control

Roughing mill setup

Finishing mill setup

Work roll coolant control

Finish temperature control

1 Gbps Ethernet communications

Coiler temperature control

Hot backup level 2 strategy

Roll thermal wear

On-line and Off-line model operation modes

Strip crown and flatness control

Automatic gauge control

Graphical interface that allows operators to

new products. TMEICs technical

364"- New Horizons

Boguchansk HPP, 50% owned by RUSAL

China, the worlds largest aluminium

surging 14.9% after record sales of

total aluminium consumption, thus

VAP production at RUSALs Bratsk smelter

In particular, the National Development and Reform Commission (NDRC)

In the first two months of 2014, the