Energy Efficiency and CO2 Emissions from the

Global Cement Industry

Michael Taylor, Cecilia Tam and Dolf Gielen

Energy Technology Policy Division
International Energy Agency

Draft version 30/8/2006.

Paper prepeared for the IEA-WBCSD workshop:

Energy Efficiency and CO2 Emission Reduction Potentials and Policies

in the Cement Industry,
IEA, Paris, 4-5 September 2006
Introduction

Cement accounts for two-thirds of total energy use in the production of non-metallic
minerals. In terms of CO2 emissions, cement production is by far the most important activity
in this category. Global cement production grew from 594 Mt in 1970 to 2 284 Mt in 2005,
with the vast majority of the growth occurring in developing countries, especially China. In
2005, China produced 1 064 Mt (47% of world cement production), while India, Thailand,
Brazil, Turkey Indonesia, Iran, Egypt, Vietnam and Saudi Arabia accounted for another 394
Mt (17%) (Table 1).

Table 1: Cement production, 2005 (USGS, 2006)

Production Share Cumulative
[Mt/yr] [%] [%]
China 1064 46.6 46.6
India 130 5.7 52.3
United States 99 4.3 56.6
Japan 66 2.9 59.5
Korea 50 2.2 61.7
Spain 48 2.1 63.8
Russia 45 2.0 65.8
Thailand 40 1.8 67.5
Brazil 39 1.7 69.2
Italy 38 1.7 70.9
Turkey 38 1.7 72.6
Indonesia 37 1.6 74.2
Mexico 36 1.6 75.7
Germany 32 1.4 77.1
Iran 32 1.4 78.6
Egypt 27 1.2 79.7
Vietnam 27 1.2 80.9
Saudi Arabia 24 1.1 82.0
France 20 0.9 82.8
Other 392 17.2 100.0
World 2284 100.0

The production of cement clinker from limestone and chalk is the main energy consuming
process in this industry. The most widely used cement type is Portland cement, which
contains 95% cement clinker. Clinker is produced by heating limestone to temperatures
above 950° Celsius. Cement production is an energy-intensive process in which energy
represents 20 to 40% of total production costs. Most of the energy used is in the form of fuel
for the production of cement clinker and electricity for grinding the raw materials and
finished cement. Since cement production consumes on average between 4 to 5 GJ per tonne
of cement, this industry uses 8 to 10 EJ of energy annually.

In early 2007 the IEA will publish an in-depth analysis of the energy demand, CO2 emissions
and CO2 emission reduction opportunities in industry. For each industrial sub-sector the goal
is to conduct an in-depth analysis of individual countries that account for at least 80% of total
production in that sub-sector. This paper sets out some of the initial data collected for the
analysis that will be conducted in the cement industry for this publication. Next steps are to
discuss this data and potential analysis with stakeholders to further refine the data and fill in
missing information. This analysis will also be an essential input into the IEAs ongoing
energy indicators programme of work, which hopes to engage developing countries as well as
IEA member countries.

Cement Production and Demand Drivers

Global GDP is assumed to grow by 2.9% per year between 2003 and 2050 in the ETP scenarios. That
is to say the global economy will be around 3.6 times larger in 2050 than it is today. In the past, the
global economy has experienced a relatively stable cement intensity per unit of GDP. The cement
intensity of the global economy remained within a relatively narrow band around 40 kg of cement per
US$ 1000 of GDP (1995 dollars at PPP) since the 1970s.

However, this relatively stable trend at a global level hides some quite dramatic differences between
countries and regions. There has been little change in the absolute demand in Europe and North
America since 1970, with virtually all of the growth in cement demand coming from developing
countries. More specifically, China’s demand for cement has exploded,and between 1998 and 2004 it
accounted for around two-thirds of the global growth in cement demand (USGS).

Chateau et al. (2005) show that the intensity of cement use, after initially rising, declines with
increasing GDP per capita, although this is not the case for all materials. The intensity of cement
demand per unit of GDP has declined in developed countries, at the same time as it has increased in
many developing countries. However, as their economies mature, the demand for cement per unit of
GDP is likely to decline in a similar manner to that experienced by developed countries. However,
given the different economic structure, natural resources available, levels of industrialization,
infrastructure development etc of developing countries no hard and fast rule can be applied to identify
when the intensity might peak. Our cement demand projections have been developed at a detailed
country, or regional, level. They assume a peak in the intensity of cement demand per unit of GDP
between 2015 and 2045 for developing countries, the resulting cement demand is shown in Figure 1.
China and India account for around half of global cement production.

Figure 1: Global Cement Demand by Region and Country (1970-2050)

6000

5000

4000

Other
OECD
3000
India
China

2000

1000

0
1970
1974
1978
1982
1986
1990
1994
1998
2002
2006
2010
2014
2018
2022
2026
2030
2034
2038
2042
2046
2050

Source: USGS and IEA.

The Clinker-to-Cement Ratio

Clinker production is the most energy intensive component of cement production, ordinary Portland
cement contains around 95% clinker (Table 2).

Certain cement types contain other feedstocks, such as pozzolana (volcanic ash), fly ash or granulated
blast-furnace slag. Producing these alternative cement types is far less energy intensive, due to the
lower clinker requirements per tonne of cement. This therefore reduces emissions from energy
consumption in the kiln and avoids the process emissions that stem from clinker production.
However, the availability of waste slag is limited, and pozzolana can be obtained only in certain
locations. Long-distance transportation of cement or cement feedstocks would result in significant
additional energy use, which is not an attractive option given the low value of the product.
Nonetheless, there is an appreciable potential for clinker substitutes, even in OECD countries.

Table 2: Composition of Different Cement Types

Cement type Portland Portland Blast- Activated
cement fly-ash furnace slag
cement cement cement
(%) (%) (%) (%)
Clinker 95 75 30 -
Fly ash - 25 - 45
Blast-furnace slag - - 65 -
Synthetic slag - - - 45
Quicklime - - - -
Water glass - - - 10
Sodium sulphate - - - -
Gypsum 5 - 5 -
Source: IEA, 2006.

The use of blended cement varies widely from country to country. It is high in continental Europe, but
low in the United States and the United Kingdom. In the long term, new cement types may be
developed that do not use limestone as a primary resource. The technological feasibility, economics
and energy effects of such alternative cements remain speculative and they will not be considered in
this analysis.

Figure 2 presents estimates of the clinker-to-cement ratio in a number of different countries and
regions. Significant uncertainty surrounds some of these estimates, and in many cases only a snapshot
of the industry is available from the 1990s. More accurate data for the current clinker-to-cement ratio
is needed if accurate projections of the clinker production to meet cement demand are to be made.

Figure 2: Clinker-to-Cement Ratio by Country and Region (1980-2004)

 100%

95% China
India
90%
EU-15
Russia
85%
USA
80% Mexico
Germany
75% Japan
Korea
70% Italy
Spain
65%
Brazil
Thailand
60%
Indonesia
55% Turkey

50%
19 0
19 1
82

19 3
19 4
85

19 6
19 7
88

19 9
19 0
91

19 2
19 3
94

19 5
96

19 7
19 8
99

20 0
20 1
02

20 3
20 4
05
8
8

8
8

8
8

8
9

9
9

9
9

0
0

0
0
19

19

19

19

19

19

19

20

20
Sources: Brazil - SNIC Soares, Centre for Clean Air Policy. China, Indonesia, Turkey and Russia - Worrell, et
al. 2001. Thailand – Siam Cement Company Ltd. Germany and EU15 – CEMBUREAU and EC. Italy – AITEC.
India – Coal Statistics. Mexico – IEA estimate. US – US Bureau of Mines and USGS. Canada – NRCAN and
USGS. Japan – Japanese Cement Association. Korea – Inha University.

An additional complicating factor is that in some countries, notably the US, significant blending of
clinker substitutes occurs when the concrete is mixed, rather than at the time of cement production.
This obviously has important implications for any analysis of the role that clinker substitutes can play
in reducing clinker production and hence CO2 emissions.

Clinker substitutes are not only used in cement production. Significant amounts of pozzolanic
materials are directly used for road bases etc., where they substitute cement and concrete. Also in a
few countries, notably the US and China, clinker substitutes are directly used in conceret production.
This is important because such practice reduces energy use and CO2 emissions in cement production,
but it does not show up when the energy use and CO2 emissions per tonnen of cement are compared.
In the US, 9 Mt of fly ash were directly used for concrete making in 1999. This represents about 17%
of all fly-ash. It should be compared to 86 Mt cement production, so this is equivalent to a lowering of
the clinker/cement ratio by 10 percentage points (NAHM, 2006). Also 2 million tonnes of blast
furnace slag were used in concrete. This is equivalent to the lowering of the cement/clinker ratio by 2
percentage points. In total this is equivalent to a reduction of the clinker/cement ratio by 12
percentage points. This results in important energy savings and CO2 reductions. This type of measures
must be considered when country energy efficiency and CO2 emission indicators are compared.

Technology and Energy Consumption in Cement Production

The technology used in the cement industry in developing countries (notably in China) differs from
the one used in industrialised countries. While small-scale vertical kilns predominate in China, large-
scale rotary kilns are most common in industrialised countries. Large-scale kilns are considerably
more energy efficient. The most widely used production process for Portland cement clinker is the
relatively energy efficient dry process, which is gradually replacing the wet process. In the last few
decades, pre-calcination technology has also been introduced as an energy-saving measure. Pre-
heating also helps to reduce the energy needed in the kiln.

Process types and fuel shares differ considerably by region (Table 3), which explains to a large extent
the regional differences in CO2 emissions per tonne of cement.

Table 3: Cement Technologies and Fuel Mix by Region

Process type Fuel share
Dry Semi-dry Wet Vertical Coal Oil Gas Other
(%) (%) (%) (%) (%) (%) (%) (%)
United States 65 2 33 0 58 2 13 26
Canada 71 6 23 0 52 6 22 15
Western Europe 58 23 13 6 48 4 2 42
Japan 100 0 0 0 94 1 0 3
Australia and New
Zealand 24 3 72 0 58 <1 38 4
China 43 0 2 55 94 6 <1 0
Southeast Asia 80 9 10 1 82 9 8 1
1
Thailand 80 4 0 16
Brazil 98 n.a. n.a. n.a. 1 66 1 32
South Korea 93 0 7 0 87 11 0 2
India 50 9 25 16 96 1 1 2
Former Soviet Union 12 3 78 7 7 1 68 <1
Eastern Europe 54 7 39 0 52 34 14 <1
Latin America 67 9 23 1 20 36 24 12
Africa 66 9 24 0 29 36 29 5
Middle East 82 3 16 0 0 52 30 4
Source: WBCSD, 2002, IEA estimates. 1 Siam Cement Industry Company Ltd. only.

Figure 3 presents the data for the average energy consumption per tonne of clinker in some of the
largest cement producing countries. The theoretical minimum energy use is 1.76 GJ per tonne of
cement clinker. Efficient pre-heater and pre-calciner kilns use approximately 3.06 GJ of energy per
tonne of clinker, while a wet kiln uses 5.3 to 7.1 GJ per tonne of clinker (WBCSD, 2002). In the
European Union, the average energy consumption per tonne of Portland cement is currently 3.7 GJ
per tonne. China, Canada and the US all require around 5 GJ/tonne of clinker. Together these three
countries accounted for around half of total cement production. For most of the other countries
presented the range is between around 3.15 GJ/tonne of clinker and 3.65 GJ/tonne of clinker.

The rapid economic development and demand growth for cement in many developing countries has
provided an opportunity, in some cases, for developing countries to achieve relatively high levels of
energy efficiency by building efficient dry-process plants to meet demand. Until recently, this wasn’t
the case in China, where small-scale kilns and shaft kilns dominated production. For instance, in 1995
the output from NSP kilns in China was only 6% of the total and large and medium-scale kilns
accounted for only 33% of total output (Cui, 2006). However, the situation is changing rapidly and by
2004 large and medium-scale plants accounted for 63% of production and that from NSP kilns for
around 45% of total output. Plans are that cement from large-scale NSP kilns should reach 80% of
total cement production by 2010 and 95% by 2030, reducing the contribution from shaft kilns to just
5%.
Figure 3: Energy Requirement per tonne of Clinker by Country including Alternative Fuels
China
7
India

6 USA

Canada

5 Mexico
GJ/Tonne of clinker

Germany
4
Japan

Italy
3
Spain

2 Brazil

Thailand
1
Indonesia

Theoretical
0 min.(incl. drying)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Theoretical min.
(excl. drying)

Sources: Brazil – SNIC Soares, Centre For Clean Air Policy and IEA estimate. Canada – NRCAN and Cement
Association of Canada. China – LBNL and IEA estimate. Germany – VDZ and CEMBUREAU. India - NCB.
Italy - AITEC. Japan – JCA and IEA estimate. Mexico – Jacott, et al. 2003. Thailand - Siam Cement Company
Ltd. US - Jacott, et al. 2003.
Note: The theoretical minimum energy requirement including drying is an upper limit, based on a very wet
feedstock. Most feedstock tends to be less wet than this, meaning that the actual theoretical minimum can be
below this.

Today’s state-of-the-art dry-rotary clinker kilns are fairly fuel efficient, consuming about 3.0 GJ/tonne
of clinker (Figure 4). The thermodynamic minimum to drive the endothermic reactions is
approximately 1.8 GJ/tonne. The superior performance of rotary kilns makes them the likely
technology for the next decades.

Figure 4: Energy Efficiency of Various Cement-Clinker Production Technologies

5
[GJ/t clinker]

0
Shaft kiln W et kiln Dry kiln (four Precalciner dry
stage preheater) kiln (six stage
preheater)

Source: WBCSD, 2002.

Electricity Consumption in the Cement Industry

Grinding is the largest electricity consumer in the cement industry. Currently around 100 kWh per
tonne of clinker is consumed in rotary kilns for grinding raw materials, at the kiln, and for grinding
the cement.1 The current state-of-the-art technologies, using roller presses and high-efficiency
classifiers, are much more efficient than previous ones. Current best practice is around 75 to 80 kWh
per tonne of clinker (Sathaye, J. et al, 2005). Still, the energy efficiency of grinding is typically only 5
to 10%, with the remainder converted to heat. There can be an enormous variation in the consumption
of electricity per plant, for example cement plants in Brazil were found to consume between 90
kWh/tonne of clinker and 200 kWh/tonne of clinker (Soares and Tolmasquim, 2000). Figure 5
presents electricity consumption for cement production by country.

Figure 5: Electricity Consumption per tonne of Cement by Country (1980-2005)

200
China
180
India
160

140 USA
kWh/tonne of cement

120 Canada

100 Mexico

80
Japan
60
Italy
40
Spain
20
Brazil
-
1980 1985 1990 1995 2000 2005

Alternative Fuel Use in the Cement Industry

Another way to reduce emissions and fossil fuel use is to burn waste or biomass as fuel. Cement kilns
are well-suited for waste combustion because of their high process temperature and because the
clinker product and limestone feedstock act as gas-cleaning agents. Used tires, wood, plastics,
chemicals and other types of waste are co-combusted in cement kilns in large quantities. Providers in
Belgium, France, Germany, the Netherlands and Switzerland have reached average substitution rates
from 35% to more than 70% of the total energy used. Some individual plants have even achieved
100% substitution using appropriate waste materials. However, very high substitution rates can only
be accomplished if a tailored pre-treatment and surveillance system is in place. Municipal solid waste,
for example, needs to be pre-treated to obtain homogeneous calorific values and feed characteristics.

The cement industry in the United States burns 53 million used tyres per year, which is 41% of all
tyres that are burnt and equivalent to 0.387 Mt or about 15 PJ. About 50 million tires, or 20% of the
total, are still land filled. Another potential source of energy is carpets, the equivalent of about 100 PJ
per year are dumped in landfills and could instead be burnt in cement kilns.

1
A small amount is also used for miscellaneous utilities on site and in packing.
In Japan, around 200 kilo tonnes (kt) of used tyres were burnt in 2005, 450 kt of waste oil, 340 kt of
wood chips and 300 kt of waste plastic (JCA, 2006). This is equivalent to around 42 PJ of energy
from alternative sources assuming 426 GJ/tonne for waste oil, 166 GJ/tonne of wood chips and 356
GJ/tonne of waste plastic.2
Although these alternative materials are widely used, their use is still controversial, because cement
kilns are not subject to the same tight emission controls as waste incineration installations. According
to IEA statistics, the OECD cement industry used 66 PJ of combustible renewables and waste in 2003,
around half of it industrial waste and half wood waste. Worldwide, the sector consumed 112 PJ of
biomass and 34 PJ of waste. There is apparently little use of alternative fuels outside the OECD,
although the comparison of country data from various sources with IEA statistics tends to imply that
alternative fuel use is under-reported. From a technical perspective, the use of alternative fuels could
be raised to 1 to 2 EJ, although there would be differences among regions due to the varying
availability of such fuels.

Cement Industry Scenarios:

Building on the IEA Publication Energy Technology Perspectives: Scenarios and
Strategies to 2050

The scenarios in this analysis build on the IEAs ACT MAP scenario, which assumes an incentive of
25 USD/t CO2 worldwide (IEA, 2006). The energy use and CO2 emissions can be calculated if the
cement production volume, the energy intensity of production, the energy mix, the clinker/cement
ratio and the use of CCS are known. Two scenarios have been analysed, one without CCS and one
with CCS. In the scenario with CCS, 350 Mt CO2 is captured by 2050. This equals the emission of
about 350 modern kilns, or 10% of all cement kilns in 2050. While more CO2 could be captured, this
would require a higher incentive level, unless new low-cost capture technologies are developed.

The general assumptions are that the world average clinker production fuel intensity declines to 3 GJ/t
clinker, the electricity use declines to 100 kWh/t cement, the shares of coal, gas and alternative fuels
are 70%, 5% and 25% by 2050, and the clinker/cement ratio declines to 0.7. This clinker/cement ratio
implies the use of about 1 500 Mt clinker substitutes in 2050. At that time between 200 and 250 Mt
blast furnace slag will be available and about 100 Mt steel slag, based on (Gielen and Podkanski,
2006). Also about 80 EJ of coal is used for power generation in the IEA ACT MAP scenario (IEA,
2006), resulting in about 300-400 Mt fly-ash. Especially in China and India, this could become a key
clinker substitute. However its use potential is limited as Portland fly-ash cement contains only 25%
fly ash. Certain other slag materials such as waste incineration bottom slag could also be used, but the
quantities are limited. Natural pozzolans are not widely available but their transportation over longer
distances may become an option once a CO2 incentive is introduced. In conclusion, the 1 500 Mt
clinker substitutes in these scenarios should be considered an upper limit.

The CO2 emissions almost double from today’s level to 2.4-2.7 Gt CO2/yr (Figure 6). This compares
to total world CO2 emissions in the IEA ACT MAP scenario of about 28 Gt by 20503. So 9-10% of
the world CO2 emissions in the ACT MAP scenario can be attributed to cement production. This puts
the cement industry in the category of key sources. Almost 2 Gt of CO2 (70% of the total cement
industry CO2 emission in ACT MAP) is process emission which can not be reduced through energy
efficiency measures. Therefore effective CO2 reduction policies must not only focus on energy
efficiency and fuel substitution, but also include process emissions.

2
These energy values for alternative fuels may be too high, as other data from JCA implies that 10% of energy
needs come from alternative fuels (incl. tires), or only around 20 PJ.
3
26 Gt energy related emissions, 2 Gt process emissions
 Figure 6: Direct CO2 Emissions from Cement Production (1970-2050)

3000

2500

2000
[Mt CO2/yr]

1500

1000

500

0
1970 1980 1990 2000 2010 2020 2030 2040 2050

25 USD/t CO2 w/o CCS 25 USD/t CO2 with CCS

Conclusions

More than half of all cement production comes from countries with an energy intensity of clinker
production significantly higher than today’s most efficient dry cement kilns with pre-heaters and pre-
calciners. There is significant variation in the amount of electricity used to produce cement, ranging
from around 100 kWh in Japan, Spain, India, Thailand and Italy to 160 kWh in Canada.

The scope for utilising clinker substitutes is large in some countries, but many large producers already
have quite low clinker-to-cement ratios. There could be an emerging constraint in terms of feedstock
availability for some of these clinker substitutes in some countries over. Significant uncertainty
surrounds some of the data and further investigation is merited.

CO2 emissions in the cement industry will continue to rise, even if new ambitious CO2 policies are put
in place. In the IEA ACT MAP scenario where global CO2 emissions in 2050 are at today’s level,
between 9 and 10% of the world CO2 emissions can be attributed to cement production. This makes
cement production a key emissions source.

Almost 2 Gt of CO2 (70% of the total cement industry CO2 emissions in ACT MAP) are process
emissions which can not be reduced through energy efficiency measures. Therefore effective CO2
reduction policies must not only focus on energy efficiency and fuel substitution, but also address
process emissions.

Options exist to reduce CO2 emissions further, such as a wider application of CCS and the production
of new clinker substitutes or pozzolan transportation. However, the feasibility and economics need to
be analysed in more detail.
References

Anand, S., Vrat, P. and Dahiya, R.P. (2006), Application of a system dynamics approach for
assessment and mitigation of CO2 emissions from the cement industry, Journal of Environment
Management 79, pp. 383-398.

Cement Association of Canada (CAC) (2006), 2006 Sustainability Report, CAC, Ottawa, ON.

Chateau, B., et al. (2005), VLEEM 2 (Very Long Term Energy Model), Study prepared for the
European Commission, EC/DG Research, ENERDATA.

Cui Yuan-Sheng and Wang Xiao (2006), The state of the Chinese cement industry in 2005 and its
future prospects, Cement International 4/2006, pp. 36-43.

European Commission (EC) (2006), Annual European Community Greenhouse Gas Inventory 1990–
2004 and Inventory Report 2006, EEA Technical report No 6/2006, EC DG Environment, European
Environment Authority, Brussels.

Gielen, D.J., Podkanski, J. (2006), Technological potentials for CO2 emission reduction in the global
iron and steel industry, Proceedings Ishii symposium on sustainable ironmaking, 2-3 March, 2006,
Sydney. Cooperative Research Centre for Coal in Sustainable Development CCSD, Brisbane,
Australia.

IEA (2006), Energy Technology Perspectives. Scenarios & Strategies to 2050, IEA/OECD, Paris.

Jacott, M., et al. (2003), Energy Use in the Cement Industry in North America: Emissions, Waste
Generation and Pollution Control, 1990-2001, paper prepared for the 2nd North American
Symposium on Assessing the Environmental Effects of Trade, May 30, 2003.

Japanese Cement Association (2006), Cement Industry’s Status and Activities for GHG Emissions
Reduction in Japan, presentation to the IEA-WBCSD Workshop “Energy Efficiency and CO2
Emission Reduction Potentials and Policies in the Cement Industry”, IEA, Paris, 4-5 September 2006.

NAHB (2006), Technical Information Resource, http://www.toolbase.org/index.aspx, NAHB

Research Center, Upper Marlboro, MD.

Natural Resources Canada (NRCAN) (2001), Energy Consumption Benchmark Guide: Cement
Clinker Production, Office of Energy Efficiency, NRCAN, Ottawa, ON.

Raina, S. J. (2002), Energy Efficiency Improvement in Indian Cement Industry, National Council for
Cement & Building Materials, paper prepapred for IIPEC Programme.

Sathaye, J., et al. (2005), Assessment of Energy Use and Energy Savings Potential in Selected
Industrial Sectors in India, Lawrence Berkeley National Laboratory, Berkeley, CA.

Siam Cement Company Ltd (2005), Sustainable Development Report 2005, Siam Cement Company
Ltd, Bangkok.

Soares, J.B. and Tolmasquim, M. T. (2000), Energy Efficiency and Reduction of CO2 Emissions
Through 2015: The Brazilian Cement Industry, Mitigation and Adaptation Strategies for Global
Change 5, pp. 297–318, The Netherlands.
United States Geological Survey (USGS), Minerals Yearbook, various years, USGS, Reston, VA.

WBCSD ( 2002), Toward a Sustainable Cement Industry. Climate Change, WBCSD, Geneva,
http://ww.wbcsd.org.

Worrell, E., et al. (2001), Carbon Dioxide Emissions from the Global Cement Industry, Annual
Review of Energy and Environment, Vol 26, 2001.