BRAZILIAN SUGARCANE ETHANOL: LESSONS LEARNED1 ...

BRAZILIAN SUGARCANE ETHANOL: LESSONS LEARNED1

ABSTRACT

Ethanol from sugarcane is an efficient and renewable biofuel that appears as a solution to the problems of rural development, diversification of energy sources, fossil fuel savings, as well as contributing to the reduction of local pollutants from vehicle exhausts and net reductions in greenhouse gas (GHG) emissions. During the 30 years of the Brazilian Alcohol Program, Brazil has developed a significant experience in the various aspects of sugarcane ethanol production. This paper discusses this experience and the lessons learned, with special attention paid to the following topics:

(i) private investments now fund alcohol mill construction, eliminating the need for subsidies or other government incentives (so there is no influence of the sunk capital investments made in the past); and it is competitive with gasoline without need for government subsidies to the industry; (ii) the favorable energy balance of ethanol; (iii) there is no competition for land with food; (iv) quality of jobs and social impacts; (v) compatibility of existing fleets with ethanol-gasoline blends; (vi) adequate legislation can control local environmental impacts associated with feedstock production and biofuel manufacture; (vii) perspectives for the replication of Brazilian ethanol program in other developing countries.

Keywords: Brazilian sugarcane ethanol, impacts, biofuels

1 Paper prepared for STAP workshop on Liquid Biofuels, Delhi, Aug 29-September 2, 2005. Authors: Suani Teixeira Coelho, José Goldemberg, Oswaldo Lucon, Patricia Guardabassi. Contributors: Plinio Nastari, Henry Joseph Jr, Luis Carlos Correa de Carvalho (Caio), Olimpio Alvares Jr., Renato Linke, Suleiman Hassuani. Revised version prepared for the Energy for Sustainable Development, September 2005 (Suani Teixeira Coelho, José Goldemberg, Oswaldo Lucon, Patricia Guardabassi) São Paulo State Environment Secretariat. Av. Prof. Frederico Hermann Jr. 345 054889-900 São Paulo SP Brazil tel (55)11 3030 6181 fax (55)11 3030 6185.

2

BRAZILIAN SUGARCANE ETHANOL: LESSONS LEARNED

INDEX

ABSTRACT ...........................................................................................................................1

1. INTRODUCTION ..............................................................................................................3

2. ETHANOL PRODUCTION COSTS VS. SUNK COSTS AND SUBSIDIES ......................4

3. ETHANOL ENERGY BALANCE......................................................................................6

4. LAND USE FOR SUGARCANE - COMPETITION BETWEEN BIOFUEL CROPS AND FOOD FOR LAND...............................................................................................................10

5. QUALITY OF JOBS AND OTHER SOCIOECONOMIC ISSUES ...................................12

6. COMPATIBILITY OF EXISTING FLEETS WITH ETHANOL-GASOLINE BLEND.....14

7. ENVIRONMENTAL IMPACTS ON ETHANOL USE AND PRODUCTION ..................17

8. PERSPECTIVES FOR THE REPLICATION OF BRAZILIAN ETHANOL PROGRAM IN OTHER DEVELOPING COUNTRIES.................................................................................19

9. CONCLUSIONS ..............................................................................................................23

REFERENCES .....................................................................................................................26

ANNEX 1. VEHICLE PERFORMANCE .............................................................................28

ANNEX 2. ETHANOL ENVIRONMENTAL IMPACTS .....................................................30

ANNEX 3. LAND USE SCENARIOS..................................................................................41

3

BRAZILIAN SUGARCANE ETHANOL: LESSONS LEARNED2

1. INTRODUCTION

Sugarcane ethanol is already well known as an efficient and renewable biofuel. In Brazil, it has promoted rural development, diversification of energy sources, lower dependence on oil imports, reduction in local pollutants from vehicle exhausts and net reductions in greenhouse gas (GHG) emissions.

These objectives were reached with the development of the Brazilian Alcohol Program (PROCALCOOL) through the several lessons learned, as discussed below.

PROALCOOL was created in 1975 to increase the production of alcohol for fuel purposes in face of rising oil prices on the international market. In the early stages of the alcohol program,ethanol use became viable to consumers through a pricing policy applied to fuels in Brazil. As the efficiency and cost competitiveness of ethanol production evolved over time, and fuel prices were liberalized, this support was no longer needed and was eliminated. Governmental incentive did exist in the past, but today the industry has matured significantly and relies exclusively on private investments.

The positive results of PROALCOOL were possible due to the technological achievements in both sugarcane and ethanol production. Due to these developments, Brazil is nowadays the benchmark in sugarcane-based ethanol production. As a consequence of the observed cost reduction, subsidies were fully eliminated by 1997 and are no longer applied on anhydrous nor on hydrated ethanol; the program is self-sufficient. Hydrated ethanol is sold to consumers for less than 70 per cent (by volume) of the gasoline price, corresponding to ethanol break-even price vis-à-vis gasoline. So alcohol is economically competitive without any subsidies.

2 Paper prepared for STAP workshop on Liquid Biofuels, Delhi, Aug 29-September 2, 2005. Authors: Suani Teixeira Coelho, José Goldemberg, Oswaldo Lucon, Patrícia Guardabassi. Contributors: Plinio Nastari, Henry Joseph Jr, Luis Carlos Correa de Carvalho (Caio), Olimpio Alvares Jr., Renato Linke, Suleiman Hassuani. São Paulo State Environment Secretariat. Av. Prof. Frederico Hermann Jr. 345 054889-900 São Paulo SP Brazil tel (55)11 3030 6181 fax (55)11 3030 6185.

4

Brazilian lessons can indeed be reproduced in other developing countries, contributing to a global expansion of ethanol biofuel, considering:

(i) Private investments now fund alcohol mill construction, eliminating the need for subsidies or other government incentives (so there is no influence of the sunk capital investments make in the past); and it is competitive with gasoline without need for government subsidies to the industry;

(ii) the favorable energy balance of ethanol

(iii) there is no competition for land with food

(iv) quality of jobs and social impacts

(v) compatibility of existing fleets with ethanol-gasoline blends

(vi) adequate legislation can control local environmental impacts associated with feedstock production and biofuel manufacture

(vii) perspectives for the replication of Brazilian ethanol program in other developing countries

2. ETHANOL PRODUCTION COSTS VS. SUNK COSTS AND SUBSIDIES

The success of sugarcane ethanol fuel in Brazil is often related to the sunk costs of the PROALCOOL Program (established in 1975) and to the “ongoing amortization of such hidden investments”.

The answer to this is simple: past investments in Brazil have improved the ethanol learning curve effect – and new players do not need to start from scratch. Ethanol has become fully competitive with gasoline in the international market, as shown in Figure 1. Sugarcane ethanol can be produced in several countries with potential resources, that can begin with advanced technologies.

5

1

10

100

0 50000 100000 150000 200000 250000 300000Ethanol Cumulative Production (thousand m3 )

(200

4) U

S$

/ GJ

Ethanol prices in Brazil Rotterdam regular gasoline pricetrend (Rotterdam gasoline prices) trend (Ethanol prices)

19862004

2002

1999

1980

1990

1995

Figure 1. Learning curve: Brazilian sugarcane ethanol competitiveness with Rotterdam gasoline. Source: Nastari, 2005.

In Brazil, the present wave of exclusivelly private investments – without further need of governmental assistance - in new sugar mills (around 50 new mill processing 3 million tones of sugarcane per year) demonstrate the economic competitiveness of sugarcane ethanol. During the 2003-2004 season, the country exported 2.5 billion liters of ethanol. According to the Ministry of Agriculture, Livestock and Food Supply (2005), the main importers of Brazilian ethanol in 2004 were India (480 million liters), the U.S. (425 million liters), South Korea (280 million liters), Japan (220 million liters) and Sweden (190 million liters).

In fact, sugarcane feedstocks represent a dominant share in the cost buildup of ethanol. The economic cost of production is in the range of US$0.18–0.25 per liter of gasoline-equivalent (average export price of ethanol in the period 2001-2003 was US$ 0.23 per liter). Nowadays, the initial investment for a compatible industrial plant (processing capacity of 2.16 million tonnes of sugarcane per year) is around US$ 60 million (in 2005 prices). Located in the Center-South of Brazil, this plant yields on average 79.39 liters of anhydrous ethanol equivalent (82.86 liters of hydrous) per tonne of sugarcane. Price paid per tonne of sugarcane is US$11.4 (UNICA, 2005). A simple calculation considering the price and a plant lifetime of 25 years would lead to a feedstock cost of US$ 0.143 per liter of

6

ethanol and an investment cost around US$ 0.017 per liter of ethanol. But investments are affected by the extremely high interest rates in Brazil: banks add their spreads to 19.75% per year, which is the basic official rate in August 2005. Brazilian ethanol is not anymore being subsidized3.

Biofuel production plants are capital-intensive, but not more than oil derivatives plants. Moreover, such investment has a longer life than the usual 15 year lifetime for oil wells. Initial investments represent 15% of the total production cost of alcohol in Brazil. This means it costs US$ 6.00 to produce one barrel of ethanol equivalent to gasoline (200 liters).

3. ETHANOL ENERGY BALANCE

Production costs of ethanol from sugarcane are low not only due to geographic conditions but also because of the extremelly favourable energy balance.4

The fuel life-cycle of ethanol in Brazil estimates that the greenhouse gases avoided emission for anhydrous ethanol is 2,7 kg CO2 equivalent/liter of ethanol. Sugarcane ethanol emits much less greenhouse gases than other biofuels during the whole life cycle (Figure 2). Table 1 shows the energy balance of sugarcane ethanol, demonstrating that more than eight units of energy are produced from each unit of fossil fuel consumed. The finding opens important opportunities for participation in the (new-born) carbon trade market that will certainly flourish when the Kyoto Protocol and its Clean Development Mechanism come into use.

3 from the Brazilian taxes, CIDE is an object of discussion CIDE (Contribution fot Intervention in Economic Domain) is a Federal tax on oil fuels. Natural gas and ethanol do not pay it. According to UNICA (2005a), “CIDE is the only tax which is given a differentiated value according to each fuel. The current level for CIDE on gasoline is R$ 280 per cubic meter; for diesel fuel is R$ 70 per cubic meter; and for hydrous ethanol is zero (Brazil, Decreto nr. 4565, de 03/01/03). The CIDE differential for gasohol and hydrous ethanol is, therefore, limited to only R$ 280 per cubic meter, or US$ 0.0957/liter in 2004 US dollars. It is difficult to admit that this difference in CIDE represents a tax differential, since the objective of CIDE since its inception has been to complement the price which is set by Petrobrás as the reference price of petroleum fuels. Proof to this fact is provided when a closer look is given to the values defined as the ‘reference prince’ of gasoline and the value defined as CIDE for gasoline. In 2004, the average prince of gasoline net of taxes was R$ 0.74837 per liter (Agência Nacional do Petróleo, “Preços da Gasolina A – Preços Médios Ponderados Semanais”, from www.anp.gov.br.), weighted averaged for Brazil. Adding the R$ 0.28 per liter value of CIDE, the total value attributed to pure gasoline which can compare to the price of ethanol net of taxes comes to R$ 1.02837 per liter, average value for 2004. This value translates into US$ 55.88 per barrel, which is practically identical to the average price of midgrade gasoline in the US during the same period, of US$ 56.28 per barrel (or US$ 1.32 per gallon) (Energy Information Agency, US Department of Energy, “U.S. Refiner Motor Gasoline Prices by Grade and Sales Type”, Washington, DC, 2005)”

7

0 10 20 30 40 50 60 70 80 90 100

Cereal ethanol

Beet ethanol

Wheat straw ethanol

Corn ethanol

Sugarcane ethanol

kg CO2eq./GJ fuel

Figure 2. Greenhouse gas emissions from different types of ethanol and conventional fossil fuels (Sources: Macedo et. alii, 2004, UK DTI, 2003 and USDA, 1995). Reference for petrol is 87 kg CO2eq/ GJ fuel and for diesel 95 kg CO2eq/ GJ fuel (DTI, 2003).

8

Table 1. Energy balance of sugarcane ethanol

Source: Macedo et. alii, 2004.

This favorable energy balance is mainly due to the fact that all energy needs in sugarcane mills are provided without any external energy source. Sugarcane bagasse5 is burned in boilers to produce steam and electric/mechanical energy to fuel the process (cogeneration process).

The potential to generate surplus electricity is huge. Presently up to 80 kWh/tonne of cane can be sold to the grid with substitution of low-pressure boilers (22 var, which yields 20 kWh/ tonne of cane) by high pressure ones (up to 80 bar). Outputs of 120kWh/tonne can be reached with better technology and recovery of sugarcane by-products. Gasification technology under development is expected to reach 300kWh/tonne of cane.

In fact, bagasse is the most important industrial by-product from sugarcane, available at the mills at no cost and sugarcane trash, when the sugarcane is harvested without burning. For example, midsize plants in Sao Paulo State,

5 Bagasse is the by-product from sugarcane crushing; it corresponds to 30% (in weight) of sugarcane, 50% wet.

9

processing 2 million tones of cane per harvesting season were able to increase their capacity for surplus generation from 2.3 MW to 9.3 MW, providing renewable electricity to local consumers.

Considering the energy balance for different crops, not whitstanding the variations in figures provided by different studies, there is no doubt that sugarcane is definetely an efficient feedstock in terms of replacement of fossil fuels (CO2), as shown in Figure 3.

0

2

4

6

8

10

12

Sugarcane Sugar beet Wheat straw Corn Wood

ethanol feedstock

ener

gy o

utpu

t/inp

ut r

atio

Figure 3: Energy balance of alcohol production from different feedstocks. Sources: Macedo et alii, 2004; UK DTI, 2003 and USDA, 1995

In fact, sugarcane has high photosynthesis efficiency and, because of its nature (grass species), can grow very well in an intensive culture, does not affect food production while generating competitive products. Its high intensity reduces environmental impacts and increases production longevity. Comparing different crops, some types of ethanol may have low – or even negative – energy balances. This argument is commonly raised against biofuels as a whole, not considering the extremely positive balance of sugarcane ethanol. Behind this consideration, in fact, there is the hypothesis that the only ethanol that could possibly be used by a country is that which is produced domestically. But this does, not taking in to consideration the option of importing biofuel.

10

Both developing and developed nations can benefit from ethanol, mainly because the use of this naturally clean fuel in their huge transport systems would significantly reduces net mobile source emissions of greenhouse gases. Ethanol life cycle assessment shows that a change from fossil fuels to biofuels could reduce CO2 emissions by a factor of five - provided that a high proportion of renewable energy is used at all stages in of the production process (Mänsson, 1998). However, sales of carbon emission reductions are just a part of the whole problem of global warming. A reduction in lifecycle GHG emissions in the transport sector is, according with many, the most relevant issue to be addressed in this century. The sector´s contribution is systematically growing and, keeping this pace, by 2030 the transportation sector will surpass the industrial sector as the leading CO2 emitter (IPCC, 2001).

4. LAND USE FOR SUGARCANE - COMPETITION BETWEEN BIOFUEL CROPS AND FOOD FOR LAND

Competition for land between biofuels crops and food crops is an issue often discussed and doubts have been raised about which is the better use for land. In this case too the Brazilian experience can contribute to the discussion.

The sugarcane average productivity in Brazil was around 65 t/ha by 1998 (Moreira and Goldemberg, 1999) but it was as high as 100-110 t/ha in the State of São Paulo (Braunbeck et al., 1999). Since the beginning of PROALCOOL yields have grown about 33 per cent in the State of São Paulo due to the development of new species and to the improvement of agricultural practices.

Sugarcane crops productivity gains from 1977 to 2004 allowed sparing two million additional hectares of land, equivalent to US$ 1.2 billion. Also genetic improvements allow cultures to be more resistant, more productive and better adapted to different conditions. Such improvements allowed the growth of sugarcane production without excessive land-use expansion.

Figure 4 confirms this statement since it shows that in Brazil food crops were not affected by sugarcane growth.

11

In São Paulo State, for example, sugarcane expansion during the period 2002-2004 ocurred mainly on land previously used for cattle(SMA, 2005). It must also be considered that, during every harvesting season, 20% of the sugarcane crop is cut and replaced with other crops like beans, corn, peanuts, etc. This procedure is mandatory and allows the soil to recuperate, being used throughout the country.

Harvested Area in Brazil

-

2.000

4.000

6.000

8.000

10.000

12.000

14.000

16.000

18.000

1930 1940 1950 1960 1970 1980 1990 2000

1000

ha

Rice Coffee Sugarcane Bean Corn Soybean Wheat

Figure 4. Area harvested of different cultures in Brazil. Source: Brazilian Statistcs Bureau, several years.

In Brazil there are soils that have been producing sugarcane for more than 200 years with ever-increasing yield. Good agricultural practices (direct plantation, protection against erosion, compactation and moisture losses, correct fertilization etc.) increase the sustainability of the culture.

Land area available for biofuels must not depend on deforestation nor competition with food, as happens now in Brazil. Sugarcane crops create no pressure on Amazon deforestation 6 , particulary because this crop is not adequate suited for production in that region.

6 Amazon deforestation is indeed a problem to be addressed but the main pressure on it is from soy crops and not from sugarcane.

12

Figure 5 below shows the map of Brazil, including the Amazon region; it shows also that most areas occupied with sugarcane are in Southeastern region and, moreover, the São Paulo State.

Figure 5. Map of Brazil locating sugarcane cultures (São Paulo State borders delimited around red area) and major ecosystems .

5. QUALITY OF JOBS AND OTHER SOCIOECONOMIC ISSUES

Job creation and job quality are other important subjects to be discussed, mainly when addressing possibilities for replicating the Brazilian Alcohol Program in other countries. In general, the number of jobs created in sugarcane sector is quite high, mainly in rural areas. In Brazil, for every 300 million tones of sugarcane produced, approximately 700,000 jobs were created. Also, workers in sugarcane industry are becoming progressively more skilled and better paid, sustaining several municipal economies.

It is important to note that Brazilian labor laws are strict and regulate all kinds of employment activities in the country.

Amazon Forest

Sugarcane cultures

Atlantic Rainforest

Pantanal grasslands

13

On the other hand, in São Paulo, the same legislation7 that established the mandatory mechanized harvesting of green cane phased-out (from 2002 to 2022), also mandates a program of professional re-qualification to those rural workers who used to harvest sugarcane and were replaced by the mechanical harvesting. By 2005, around 25% of the sugarcane in Sao Paulo State is harvested whithout burning, according to this law, and all workers involved received this re-qualification.

Although sugarcane production is seasonal (6 to 9 months per year), many jobs are formal and annual, due to constant activities throughout the year, such as equipment maintenance during off season.

Regarding economic development, regions where sugarcane is produced generally have high quality of life levels - this is particularly true in São Paulo State. Ribeirao Preto and other municipalities, where sugarcane/alcohol production is the main agro-industrial activity, present nowadays the highest quality life level in the state and perhaps in the country.

Other cultures (fruits, flowers, vegetables) may be more labor intensive by area, but generate far fewer jobs. In this sense, sugarcane, soybean, cotton, wheat, peanuts and rice are preferable for job creation.

Regarding the size of sugarcane producers in Brazil, almost 75% of the sugarcane land is owned by large producers. However, there are also around 60 thousand small producers in the Midwest-Southern Regions, organized in cooperatives with increasing negotiation power. A payment system based upon the sucrose content in sugarcane has been used for a long time and has promoted significant growth in agricultural productivity.

Considering the production of sugarcane in Brazil, there are two different situations. In Sao Paulo State, in most cases the sugarcane planted area belongs to large producers. Another situation is found in Parana State (Southern region, one of the highest sugarcane producers in the country) where most sugarcane producers are small and members of a cooperative.

7 State Law 11241, from 2002 establishes targets and timetables for the elimination of the sugarcane burning before harvesting.

14

Besides the social benefits existing in this sector, there are other socio economic issues. The capital investment for job creation in sugarcane sector is much lower than in other industrial sectors, as it is shown in Figures 6a and 6b below. The creation of one job in ethanol agroindustry requires on average US$ 11 thousand, while a job in the chemical and petrochemical industry costs 20 times more. Also, the rate of jobs per unit of energy produced is 152 times higher in ethanol than in oil industry.

152

1

3

4

0 50 100 150 200

Ethanol

Oil

HydroelectricPower

Coal

jobs/energy (oil=1)

220

145

98

91

70

44

11

0 100 200 300

Chem/Petrochemistry

M etallurgy

Capital goods

Automotive Industry

Intermediate Industry

Consumer Goods

Ethanol Agroindustry+ Indust ry

1000 US$/job

Figures 6a and 6b . Employment numbers (1980-84) from PROALCOOL, the Brazilian Ethanol Program; Jobs per unit of energy produced (6a, left) and Investment for job creation (6b, right).

Source: Goldemberg, 2002

6. COMPATIBILITY OF EXISTING FLEETS WITH ETHANOL-GASOLINE BLEND

Usually, the ethanol application as a fuel brings some concerns about its compatibility with existing fleets. Doubts are about compatibility of ethanol/gasoline blends regarding metallic materials of vehicle (corrosion), plastic and rubber materials of vehicle (chemical attack), as well as higher fuel consumption (due to low molecular energy content), losses in drivability (due to different air / fuel ratio for combustion) and cold start difficulties (due to lower vapor pressure). However, this depends on the amount of ethanol blended with the gasoline, on the ethanol fuel specification and quality and on the technological level of vehicle (vehicle age).

15

Even so, small blends of ethanol (5%) are recognized to not affect existing engines.

The World Wide Fuel Charter 8 establishes a maximum oxygen content in gasoline of 2.7% m/m, which allows blends higher than 5%.

Vehicle performance is a consequence of engine power and torque and it is limited by the characteristics of the fuel that was used by the automobile manufacturer to develop these properties.

However, with the introduction in many countries of low concentration ethanol-gasoline blends (up to 10%) as regular fuel for vehicles developed for neat gasoline, no remarkable performance loss is observed. Due to the ability of the electronic fuel injection - EFI System - to automatically correct the engine air-fuel ratio, the presence of ethanol up to 10% blended in gasoline do not affect vehicle performance and the increased fuel consumption was relatively low (below 3%).

The Brazilian Automobile Manufacturers Association (ANFAVEA, 2005) has summarized the necessary modifications in vehicles for the use of national sugarcane ethanol blends (Table 2). The same associated also provided comparative performance figures for the Brazilian new vehicles (Figure 7).

8 Alliance of Automobile Manufacturers (2005)

16

Ethanol Content in the Fuel

Carburetor

Fuel Injection

Fuel Pum

p

Fuel Pressure D

evice

Fuel Filter

Ignition System

Evaporative S

ystem

Fuel Tank

Catalytic C

onverter

Basic E

ngine

Motor O

il

IIntake Manifold

Exhaust S

ystem

Cold S

tart System

�5% NN for any vehicle

5- 10% PN NN for relatively new fleets (10 – 15 years old)

10-25% PN, Brazilian application NN

25-85% PN, U.S. application NN

� 85% PN, Brazilian application

Table 2. Modifications (not necessary – NN – and possibly necessary – PN) in vehicles for different ethanol blends.

Source: ANFAVEA, 2005.

103,

3 %

110,

0 %

102,

1 %

106,

4 %

103,

2 %

105,

3 %

95,5

%89

,3 %

105,

5 % 12

9,4

%

020406080

100120140

Power Max Speed Consumption(L/100km)

Gasoline 0% Gasohol 22% Ethanol 100%

Figure 7. Comparative performance of Brazilian similar direct injection new 1999 models, with pure gasoline (E0), gasohol blend (E22, pure gasoline with 22% anhydrous ethanol) and pure hydrated ethanol (E100), according to ANFAVEA (2005).

17

The recent alternative of “flex fuel” vehicles designed to use alcohols as automotive fuel allows the use of a variety of fuels such methanol, ethanol, gasoline and blends of these fuels at any proportion. In this case, the lambda factor is highly variable, but it is quite easy to correct it through the use of eletronic systems for engine management. Volkswagen, General Motors, Fiat, Ford, Peugeot, Renault are some of manufacturers producing flex-fuel vehicles in Brazil - a fleet of 700,000 units in 2005. The huge success of these flex fuel vehicles is due to the freedom of choice for the consumers, depending of the price of each fuel at the pump station. Alcohol prices can be up to 70% of gasoline prices, considering efficiency and low heating value of both fuels. So, a global ethanol market, allowed to the utilization of FFVs, can address eventual fossil and renewable fuel shortages.

Annex 1 discusses this subject in greater detail.

7. ENVIRONMENTAL IMPACTS ON ETHANOL USE AND PRODUCTION

Ethanol is well known as a renewable biofuel, essential to helping achieve the Kyoto targets for developed countries and to mitigate greenhouse gas emissions.

Local impacts have also been tackled efficiently. During the alcohol program, Brazil has developed significant expertise in reducing the environmental impacts of ethanol’s entire life cycle:

• legislation has been improved and enforced, specially in the State of São Paulo, where 60% of all Brazilian sugarcane is produced;

• discharge of effluents with high organic loads (mainly vinasse, a by-product) is being replaced by controlled fertirrigation practices (controlled by Cetesb, São Paulo State Environmental Agency), which reduce the needs for fertilizers;

• harvest burning practices are being phased-out, through a strict legal enforcement and allowing energy benefits of mechanization from higher surpluses electricity produced from sugarcane by products (State Law 11241/2002);

18

• land use is severily controlled, being mandatory to preserve riparian forests and other natural ecosystems and avoiding deforestation (sugarcane cultivation does not interfere with the Amazon, rainforest or any other large sentitive areas);

• industrial sugarcane and ethanol plants have their liquid and atmosferic emissions (particulates, nitrogen oxydes) controlled;

• due to the fuel characteristics and advances in vehicle technology (especially flexible fuels), ethanol combustion emissions are much less harmful than those from gasoline or diesel; even acetaldehydes emitted by ethanol are less toxic than formaldehydes from diesel or gasoline;

• spill offs and other accidents are rare and of low impact;

These items are discussed in more detail in Annex 2. However, it is important to stress that, with present technology, local emissions from fuel end-use are rather favorable to ethanol, even before tailpipe controls, as shown in the Figure 8, which summarizes tests conducted by the Brazilian Automotive Industry Association´s Energy & Environment Commission (ANFAVEA). Tests were conducted with similar (in terms of models and engines) new vehicles in 1999, direct injected, before the catalyst. ANFAVEA considers replicable the experiment. After catalyst, tailpipe emissions drop around tenfold.

Larger benefits of ethanol use in Brazil were due to total replacement of lead in gasoline without using MTBE (avoiding its environmental impacts) and reducing significantly sulfur oxides and particulate matter, mainly in large cities like São Paulo.

19

Comparative raw exhaust emission

100 100 100

8580

104

51 53

86

0

20

40

60

80

100

120

CO HC NOx

Gasoline 0% Gasohol 22% Ethanol 100%

Figure.8. Comparative raw exhaust emissions in Brazilian similar new vehicles, 1999 models (ANFAVEA, 2005).

8. PERSPECTIVES FOR THE REPLICATION OF BRAZILIAN ETHANOL PROGRAM IN OTHER DEVELOPING COUNTRIES

Brazil is endowed with vast agricultural areas. Plentiful land, favorable climatic conditions and low cost labor were indeed necessary for the success of alcohol program based on sugarcane. So these are significant considerations if other developing countries intend to start a program without assistance from developed countries.

Probably 10 or 20 tropical countries can handle a large alcohol program like Brazil. Although sugarcane is a highly intensive culture possible to be produced in regions that have an average temperature above 20oC and plenty of available sunlight and water, it does not imply that areas covered with forestry are the most suitable for the culture. Figure 9 below gives an idea of the regions that meet these conditions; from this figure it can be seen that most are developing countries.

20

Figure 9. Sugarcane potentials (SI, suitability index). Source: FAO (2005)9

Geographic and distributional considerations (such as “need for proximity to a sufficiently large market for the biofuel produced”) are a common failure in biofuel assessments. These arguments miss the point because they consider domestic production and consumption as as the only possible use for the fuel. Biofuels are liquids and can be transported by train, truck and boats over significant distances, as occurs today with all liquid fossil fuels (and with alcohol in Brazil, despite being a very large country). In fact, distances depend more on logistics and production scale. In Annex 3 there are some provided scenarios which show that the production of ethanol from sugarcane in developing countries is possible when considering land use aspects.

Regarding technogical aspects, the Brazilian experience shows that existing technology for alcohol production has had an increase of 3% per year on industrial productivity during the alcohol program and became commercially available. Brazil has a mature fermentation process, self-adaptative to different feedstocks and production conditions (startup, temperature changes etc), resulting in an efficient process with little undesirable byproducts, flexible, robust and inexpensive.

9 http://www.fao.org/ag/AGL/agll/gaez/ds/da.htm?map=24

21

The distillation process is also efficient (despite the fact that there is still spoace for further improvements in efficiency) and available at low costs.

The cogeneration process from bagasse has also undergone significant improvement in recent years and high efficient steam systems using sugarcane bagasse are commercially available.

Althought, in the case of Brazil, alcohol production for the blend of up to 25% is not a candidate for CDM projects, because this situation corresponds to a baseline before the base year for Kyoto Protocol. However, other developing countries can start such a program and would be excellent candidates for CDM.

9. LEAPFROGGING: PROPOSED STEPS FOR THE INTRODUCTION OF ETHANOL IN DEVELOPING COUNTRIES

From the above discussions in this paper, an interesting option for developing countries appears to be the local production of biofuels (considered the issues also discussed before).

Based on the Brazilian experience with sugarcane-ethanol, such countries could successfully introduce ethanol-biofuel in their economies. This introduction does not need to start from the very beginning, as happened with Brazil; they could start from an advanced step, avoiding past mistakes and benefiting from lessons learned by Brazil.

In any case, it appears as the best option the production of anhydrous ethanol to be blended to gasoline (in several countries the sugarcane option appears to be a good opportunity). Despite the fact that anhydrous ethanol production costs are around 10-20% higher than hydrous ethanol, there is no need for changes in existing vehicles, as it would happen for the use of straight ethanol (in this case it would be necessary to change vehicles’ engine to run with pure ethanol).

There are two main issues to be discussed:

1. Countries already producing some sugarcane for sugar and interested in producing ethanol for local consumption, reducing oil/derivatives imports:

22

these countries could start an alcohol program using part of the existing sugarcane production for alcohol production

2. Countries with no sugarcane production but with existing deforested land: these countries must start since the very beginning, including the choose for the best crop to be used

For countries in the first group, when interested in the alcohol production, it would be necessary a preliminary global (technical / economic / environmental / social) evaluation of the alcohol production. If perspectives are positive, existing policies could be discussed, together with perspectives for changes, including:

• Assessment of existing technical expertise, financing resources, policies, key economic players, other stakeholders;

• Development of an information exchange, technology transfer and capacity building program;

• To foster pilot projects for alcohol production for local consumption;

• Establishment of policies for a phase-in of (anhydrous) ethanol blends in gasoline without need of adaptation in the existing fleet, up to 5% in volume;

• Discussion of fiscal policies (if necessary) regarding economic competitiveness of alcohol fuel.

For countries in the second group, aiming to use existing degraded land, other issues must be addressed, besides a preliminary global (technical / economic / environmental / social) evaluation of the alcohol production, such as:

• Assessment of current and potential areas of arable land, sugar crops production, other cultures, rainfall and water demand and other physical conditions;

• Assessment of existing technical expertise, financing resources, policies, key economic players, other stakeholders;

• Development of an information exchange, technology transfer and capacity building program;

• Foster pilot projects, then pre-commercial scale plants;

23

• Establishment of policies for a phase-in of ethanol blends in gasoline without need of adaptation in the existing fleet, up to 5% in volume;

• Discussion of fiscal policies (if necessary) regarding economic competitiveness of alcohol fuel.

In both cases, issues regarding ethanol export could be addressed in a second phase program, including the discussion on existing trade barriers in developed countries, especially import taxes, quota allocation and harmful domestic subsidies against the WTO rules, as discussed in previously by the authors (Coelho et al, 2005).

In this context it is important to mention the opportunities for developing countries as discussed in the 2004 document (forthcoming) prepared by the World Bank Energy and Poverty Thematic Group. It is a quite valuable report called "Ethanol: Re-examining a Development Opportunity for Sub-Saharan Africa" that presents the opportunities and barriers to the implementation of ethanol production as a poverty alleviation vector.

10. CONCLUSIONS

Many biofuel programs in the developed world have benefited from policies designed primarily to support domestic agriculture. This is the case of the US corn producers and the EU rapeseed farmers. In any cases, agricultural support policies are an existing reality, a political challenge that has the World Trade Organization (WTO) as one of the possible discussion forums.

In spite of the benefits from biofuel production to sustainable development, exports to developed countries faces several barriers and local producers are against the removal of domestic subsidies. On the other hand, society, as a whole, benefit from trade liberalization, through a real introduction of an available renewable fuel. Local trading policies can balance quite well the supply of domestic and imported ethanol, in order to introduce an alternative to gasoline or diesel (Coelho et alii, 2005).

In brief, the Brazilian alcohol learning curve (figure 1) shows that industries may be supported in their developing stages but they do not need to follow the same path of Brazil. They can start from a more efficient level, using the so-called

24

“leapfrogging” 10 , saving development costs and assuring better levels of productivity with lower environmental impacts.

Biofuels are therefore: (1) a form of energy that is easily transported and stored, thus tradable; (2) environmentally friendly products and (3) obtained from agricultural feedstocks, which in many cases compete internationally with subsidised products. Biofuels are affected by protective legislations and subjected to debates at the World Trade Organization post-Doha work program. This includes negotiations on certain trade and environment issues, starting with the liberalization of trade in environmental goods and services (EGS), foreseen in the WTO Doha Declaration, Paragraph 31 (WTO, 2001). Developed countries and international organizations such as the OECD and UNCTAD have a key role to play in this regard. Multilateral Environmental Agreements (MEAs), such as the Kyoto Protocol or targets for renewables are very important driving forces for achieving sustainable development through the increased use of biofuels11.

It appears that there are also fears that sugar prices might soar because of production being directed primarily to ethanol. Biofuel prices are determined by world crop prices, which are subject to the volatility of agricultural output and crop prices. But there are always break-evens for sugar and ethanol production; producers will go to sugar only if prices allow. In a global market, it is very unlikely that adverse climate conditions may affect all countries (Colombia and Tailand, for example), at the same time and thereby impact overall biofuel supply. Also, ethanol blends can vary with flexible fuel vehicles, matching the market to the different fuel prices. Ethanol is not meant to totally replace gasoline in all cases.

The real issue is that there are importation barriers in developed countries to introducing biofuels, as it is with many other agricultural products. Biofuels are energy sources produced in rural areas to primarily service the energy needs of cities that can afford to pay. Often, biomass programs are created to produce energy for the poor living in rural areas. These programs have their merits but are limited economically because of size, given that the buyers have little purchasing power or are not even involved in the commercial market. Biofuel programs in developed countries are very often treated as a domestic issue only – a defficient strategy that provides ammunition for the defenders of fossil fuels in terms of reliability and potential.

10 “leapfrogging” means basically climbing faster the rungs in the energy ladder. More in UNDP (2002) World Energy Assessment 2000. United Nations Development Programme, Washington 11 Renewable energies are closely related to the Millenium Development Goals (MDGs). During the 2002 Johannesburg World Summit for Sustainable Development (WSSD) was proposed the so-called Brazilian Energy Initiative, a 10% global target for renewable energy (Goldemberg, 2002) . Such policy encompasses simultaneously cleaner energy, sustainable patterns of consumption, job creation, energy security and free trade. At regional level, the Latin America and the Caribbean region have approven in May 2002 a 10% target on renewable energy, the Brasilia Plataform, 2002 .

25

Although incentives to domestic production act in the learning curve effect, there is also a need to counteract the pressure from inefficient suppliers and to increase production efficiency worldwide, including progressive trade liberalization. This will allow developing countries to produce biofuels not only for domestic consumption but also to obtain revenues from biofuel exports (and even to import more value-added products from developed nations).

Considering the current investment, the main barrier to ethanol production in developing countries is the lack of funding. Developed countries, international agencies or international banks are options that could be explored. The opportunities from the Clean Development Mechanism also appear to be a good option for developed countries as a way of collaborating in the sustainable development of poorer countries.

26

REFERENCES

1. Abrantes, R., A Emissão de Aldeídos e Hidrocarbonetos Policíclicos Aromáticos de Veículos Comerciais a Diesel, CETESB, SIMEA 2003, São Paulo, Brazil.

2. Air Quality Impacts of the Use of Ethanol in California Reformulated Gasoline, California Air Resources Board, Sacramento, USA, 1999.

3. Alliance of Automobile Manufacturers (2005) World Wide Fuel Charter 1998 http://www.autoalliance.org/archives/000090.html

4. Anderson, L.G. et alii, Effects of Using Oxygenated Fuels on Carbon Monoxide, Formaldehyde and Acetaldehyde Concentrations in Denver, Air & Waste Management Association's 90th Annual Meeting & Exhibition, Toronto, Ontario, Canada, June 8 - 13 1997.

5. ANFAVEA (2005) Ethanol Fuel Vehicular Application Technology. Presentation of Henry Joseph Jr. (Brazilian Automotive Industry Association´s Energy & Environment Commission [email protected]) at CEPAL/CENBIO/USP Seminar São Paulo, August 17th, 2005

6. Apace Research Ltd., Intensive Field Trial of Ethanol/Petrol Blend in Vehicles, EDRC Project Nº 2511, Australia, December 1998. Available at http://journeytoforever.org/biofuel_library/EthanolApace.PDF

7. CARB (2004) California Air Resources Board data base 8. CAPCOA (1993) California Air Pollution Control Officers Association Air Toxics "Hot

Spots" Program, Revised 1992 Risk Assessment Guidelines. 9. CETESB (2003) Relatório de Qualidade do Ar no Estado de São Paulo. Companhia de

Tecnologia de Saneamento Ambiental 10. CETESB (2004) Air Quality Report 2003 11. Código Florestal (1965) Federal Law 4771/65 12. Coelho, ST; Lucon, O; Guardabassi, P (2005) Biofuels – advantages and trade barriers.

UNCTAD/DITC/TED/2005/1 www.unctad.org/Templates/Download.asp?docid=5741&lang=1&intItemID=1397

13. F.O. Licht (2005) World Ethanol Production 2001. Apud http://www.distill.com/berg/ 14. FAO (2005) Sugarcane potentials

http://www.fao.org/ag/AGL/agll/gaez/ds/da.htm?map=24 and http://www.fao.org/ag/AGL/agll/gaez/ds/ds.htm

15. FAOSTAT (2005) http://faostat.fao.org/faostat/form?collection=Production.Crops.Primary&Domain=Production&servlet=1&hasbulk=0&version=ext&language=EN

16. FAOSTAT (2005) Primary crops http://faostat.fao.org/faostat/form?collection=Production.Crops.Primary&Domain=Production&servlet=1&hasbulk=0&version=ext&language=EN

17. Goldemberg, J (2002) Brazilian Energy Initiative. www.worldenergy.org/wec-geis/focus/wssd/goldemberg.pdf

18. Goldemberg, J (2004) The case for renewable energies. Renewables 2004 Conference, Bonn, http://www.renewables2004.de/pdf/tbp/TBP01-rationale.pdf

19. EPA (2005) IRIS - Integrated Risk Information System: Acetaldehyde (CASRN 75-07-0) http://cfpub.epa.gov/iris/quickview.cfm?substance_nmbr=0290

20. EPA (2005) IRIS - Integrated Risk Information System: Formaldehyde (CASRN 50-00-0) http://cfpub.epa.gov/iris/quickview.cfm?substance_nmbr=0419 .

21. IEA (2004) Biofuels for Transport - An International Perspective. International Energy Agency, ISBN 92-64-01512-4

22. International Energy Agency (2003) Energy Statistics Of Non-OECD Countries, 2000-2001 - II.9

23. IPCC (2001) Third Assessment Report (TAR) "Climate Change 2001" http://www.ipcc.ch/activity/tar.htm and http://www.ipcc.ch/pub/online.htm

27

24. Macedo, I.C.; Leal, M.R.L.V. and Siflva, J.E. A.R (2004) Assessment of greenhouse gas emissions in the production and use of fuel ethanol in Brazil. São Paulo State Environment Secretariat. Also at www.unica.com.br/i_pages/files/pdf_ingles.pdf

25. Macedo, IC (2005) Evolução e Perspectivas do Etanol. Seminar “Brazilian Experience with Ethanol Fuel”, CEPAL - S. Paulo, 15-19 Aug

26. Magnetti Marelli (2005) Personal communication to Olimpio Alvares Jr, CETESB 27. Mänsson, T., Clean Vehicles with Biofuel - A State of the Art Report, KFB-Report

1998:18, Sweden, 1998. 28. Ministry of Agriculture, Livestock and Food Supply - Secretariat of Production and

Agrienergy (2005) Sugar and ethanol in Brazil, Brasilia - July 2005 29. MME (2005) Brazilian Energy Balance 2004, www.mme.gov.br 30. Nastari, PM (2005). Information Services on the Sugar and Ethanol Industries in Brazil.

Personal Communication from Plinio Mario Nastari. Email address [email protected] website http://www.datagro.com.br/ingl/index2.asp

31. National Renewable Energy Laboratory, US Department of Energy, USA, 2002. 32. O Estado de São Paulo newspaper (2005) Cana: formação de preço em debate, 17Aug

2005, p. G4 33. Sao Paulo State (2002) Decree 47.397 (4th December 2002) 34. Sao Paulo State (2002) Decree 48.523 (2nd March 2004) 35. São Paulo State (2002) Law 11241 36. São Paulo State Secretariat of Environment (2005). Personal communication, statistics

provided by Ms Elisabeth Kono, www.ambiente.sp.gov.br 37. Sher, E., Handbook of Air Pollution from Internal Combustion Engines, Academic Press,

USA, 1998. 38. The Brasilia Platform (2002) http://www.renewables2004.de/pdf/platform_declaration.pdf 39. U.S. Department of Energy, Ethanol for Sustainable Transportation, USA, April 1999. 40. UK DTI (2003), Technology status review and carbon abatement potential of renewable

transport fuels in the UK.Report B/U2/00785/REP www.dti.gov.uk/renewables/publications/pdfs/b200785.pdf

41. UN (2002) WSSD Final Plan of Implementation, Johannesburg, http://www.johannesburgsummit.org

42. UNDP (2002) World Energy Assessment 2000. United Nations Development Programme, Washington

43. UNICA (2005 a) Observations on the Draft Document entitled “Potential for Biofuels for Transport in Developing Countries”, prepared by Plinio Mário Nastari, Isaías de Carvalho Macedo and Alfred Szwarc for The World Bank Air Quality Thematic Group, July 2005 www.unica.com.br/i_pages/files/ibm.pdf

44. UNICA (2005 b) Personal communication, Alfred Szwarc 45. USDA (1995) Estimating the Net Energy Balance of Corn Ethanol. Report by Hosein

Shapouri, James A. Duffield, and Michael S. Graboski. U.S. Department of Agriculture, Economic Research Service, Office of Energy. Agricultural Economic Report No. 721. http://www.ers.usda.gov/publications/aer721/AER721.PDF

46. Whitten, G., Ethanol's Clean Air Impact, 9th Annual Renewable Fuels Association Conference, Miami Beach, FL, USA, February 18th 2004.

47. WTO (2001) Ministerial Declaration. Ministerial Conference, Fourth Session, Doha, 9 - 14 November 2001. WT/MIN(01)/DEC/1, 20 November 2001 (01-5859). http://www.wto.org/english/thewto_e/minist_e/min01_e/mindecl_e.htm

48. World Bank Energy and Poverty Thematic Group (2004, forthcoming). "Ethanol: Re-examining a Development Opportunity for Sub-Saharan Africa"

28

ANNEX 1. VEHICLE PERFORMANCE

In 1975, in the middle of the oil crisis, the Brazilian automotive industry developed a research program to evaluate the most suitable volume of anhydrous ethanol to be added to gasoline. At that time, the light duty vehicle engines were designed to work rich, looking for high power output, with high fuel consumption and tail pipe emissions. Performance and emission tests were done, adding increasing volumes of anhydrous ethanol to gasoline. The study concluded that the 20% +- 2 % ethanol blend yielded good results in terms of fuel consumption and emissions.

The energy content of ethanol fuel is 37% lower than gasoline and, consequently, to obtain the same output, it is necessary to burn more ethanol than gasoline, raising the fuel consumption. However, since ethanol density is 5% higher than gasoline, fuel consumption increment is not proportional only to the ethanol heating value but also to its specific weight, improving the final result to more acceptable levels. Energy conversion between gasoline and ethanol takes into account not only the lower heating values 12 . Equivalence must consider the final energy service provided (work) by each fuel, based on measured performance. In 1991, a Brazilian Commission for Re-exam of the Energy Matrix has assessed a ratio in consumption of 0.8067, comparing neat hydrous ethanol with gasoline. Flexible fuel vehicles, adjusted to have a similar power performance with both fuels, have lower ratios, of approximately 70% (apud UNICA, 2005a)

The Brazilian flex-fuel vehicles can operate with up to 100% hydrated ethanol. The concept is different from the US flex-fuel models that can use a maximum of 85% ethanol, due to problems at cold start and separation of phases (water/gasoline/ethanol) inside the tank at very low temperatures. When using 100% ethanol, any problem can be normally solved by means of the use of a little gasoline reservoir for instant automatically-activated gasoline injection during starts, as it is done in Brazil in any alcohol or flex vehicles.

In general the compatibility of ethanol (hydrous or anhydrous) with plastic and rubber parts is very good for almost all the materials usually employed by the

12 According to the Brazilian National Energy Balance 2004 (MME, 2005) the lower heating values (LHV) of diesel fuel, automotive gasoline, anhydrous ethanol and hydrous ethanol are respectively 35.50, 32.20, 22.34 and 21.33 MJ/liter of fuel

29

automotive industry13. Apart from pressures made by contrary economic sectors, phasing-in blends of up to 5% ethanol may not bring technological difficulties to any country. Major car manufacturers today are adjusting their assembling line to produce cars that are compatible with E10 fuels14 . The performance results showed in table 3 for a Volkswagen FFV vehicle, with similar behaviour in gasoline and alcohol.

Acceleration [ sec] Gasoline Alcohol

0 – 100 km/h 17"40 16" 92

Passig Time [sec]

40 – 80 km /h (3rd gear} 9" 70 9" 96

80 – 120 km/h (4th gear) 16" 03 16" 09

80 – 120 km/h (5th gear) 23" 10 21" 95

40 – 120 km/h (5th gear) 43" 38 44" 67

Table A1. Performance of a Volkswagen GOL Flex 1.0 liter, 16 valves, tested with two fuels, SFS technology Source: Magneti Marelli Controle Motor Ltda (2005)

13 The unique exception is the plastic polyamide 6.6 (Nylon®), which in presence of hydrous ethanol adsorbs water and swells, increasing the component dimensions; the polyamide 6.6 absorbs the water and, for this case, it is necessary to substitute it for another plastic (usually, polyamide 12). For the application with gasoline or ethanol-gasoline blends, there are not problems with the polyamide 6.6, since these fuels do not have significant water content. 14 Considering that ethanol is a “polar substance” and gasoline is an “unpolar substance”, there is not solubility between both, but only miscibility, which, in general, is good and stable. Two factors have influence on this miscibility: the presence of water and the characteristics of hydrocarbons of gasoline (resulting from the production process). However, with the use in low concentration of anhydrous ethanol (up to 10%) and with the increase of cracked gasoline application (instead direct distilled gasoline), there are not reports on separation problems in the countries were this blend is used, even in the countries with cold weather.

30

ANNEX 2. ETHANOL ENVIRONMENTAL IMPACTS

Exhaust emissions from ethanol use in light vehicles:

The first important environmental benefit from ethanol was the phase-out of lead in all gasoline produced in Brazil. The use of ethanol as a fuel anti-knock additive in gasohol avoids the need of organic lead compounds addition or even the increase of aromatics content to maintain octane levels. With the introduction of the alcohol blend, lead content in gasohol was lowered from 0.8 ml/l (1980) to 0.3 – 0.4 ml/l (mid 80’s) and since 1989 no lead has been added to gasohol in Brazil. The shift from leaded to unleaded gasoline has not resulted in any major engine problem, and most important, lead ambient concentrations in the São Paulo Metropolitan Region (SPMR) dropped from 1,4 ug/m3 in 1977 to less than 0,10 ug/m3 in 1991, far below the Brazilian air quality standard, 1,5 ug/m3.

Figure B1. Lead Quartely Average concentration in the SPMR (CETESB, 2003)

31

The use of gasoline-ethanol blends transfer to gasoline most of ethanol’s benefits, specially those regarding carbon monoxide emissions, that could be reduced up to 60 percent in older carburated vehicles. However, an increase in evaporative emissions occurs, depending on gasoline volatility and ethanol percentage. Emissions factors from neat alcohol and gasohol-fueled vehicles are presented below. The results indicate that substantial reductions in exhaust emissions have been achieved from both alcohol and gasohol-fueled vehicles, especially when comparing modern electronic injection and catalytic technology to the older carburated generation.

Although Brazilian new models are officially tested with ethanol and/or gasohol (E22), there are other results available, which compare pure gasoline (E0) with pure ethanol (E100) and with a fifty-fifty blend (E50), utilizing the Software Flexfuel Sensor (SFS), a computer engine control technology developed by Magnetti Marelli15. Table 4 below shows results of exhaust emissions of these tests.

Emissions Fuel consumption

CO HC NOx Aldehydes Evaporative Highway Urban

FTP 75 cycle

G/km g/km g/km mg/km g/test km / liter km / liter

Brazilian Legal

Limits

2.00 0.30 0.60 30.0 2.00 - -

Gasoline E0 0.65 0.10 0.15 10.0 0.55 16.50 11.0

E50 0.60 0.13 0.16 12.0 0.56 14.85 9.30

Ethanol E100 0.56 0.14 0.17 20.0 0.92 12.90 7.60

Table B1: Emissions and performance of a Volkswagen GOL Flex 1.6 liters, tested with three fuels, SFS technology, standard test FTP 75 cycle (recognized in the U.S. and Brazil). Source: Magneti Marelli Controle Motor Ltda

Catalytic converters manufacturers have developed new noble metals formulations in order to perform the exhaust gases catalysis efficiently in Brazil. In the first half of the 90’s, “tropicalized” catalysts for Brazilian market were based on paladium / rhodium for gasohol and paladium / molibdenium for (hydrous,

15 The SFS™ can also assure national commitments for renewable fuels use and greenhouse gases control programs. The SFS technology can also be extended to the use of natural gas, allowing an engine management by a unique electronic control unit.

32

pure) ethanol. This represents a significant cost reduction due to ethanol use, either as a gasoline additive or (mainly) as an alternative fuel, for paladium is cheaper than platinum. Ethanol reduces harmful pollutants from vehicle exhausts – a growing problem in major urban areas. In developing countries, it eliminates the need of leaded gasoline16, reduces significantly sulfur emissions17 (without high cost petroleum refinery upgrades) and reduces CO emissions. Even for PM and NOx, advanced vehicle technologies allow biofuels to operate with high efficiencies and low emissions.

Studies conducted in Brazil have shown significant environmental benefits due to the extensive use of ethanol-gasoline blends. A comprehensive Australian study18 has also shown positive results with low ethanol-content blends. The most obvious pollution reduction effects associated to blends containing up to 10% ethanol by volume (E10 blends) include reduction of carbon monoxide (CO), toxic hydrocarbons (such as benzene and 1-3 butadiene that are known carcinogens), sulfur oxides (SOx) and particulate matter (PM). Modern catalytic converters help significantly in the reduction of emissions, as shown in Table B2, which compares data from conventional fuel-dedicated and flexible fuel vehicles (FFVs).

Fuel CO (g/km) HC

(g/km) NOx (g/km)

Aldehydes

(g/km)

CO2

(g/km)

Autonomy

(km / l )

Gasohol (E22) dedicated 0.40 0.11 0.12 0.004 194.0 11.2

Ethanol (E100) dedicated 0.77 0.16 0.09 0.019 183.0 7.5

FFV with E22 0.50 0.05 0.04 0.004 210.0 10.3

FFV with E100 0.51 0.15 0.14 0.020 200.0 6.9

Range for FFV with E61 (50%

E22 / 50% E100) 0.15 – 0.74

0.038 –

0.14 0.06 – 0.19 NA NA NA

Table B2. Averaged emission factors for 2003 light vehicle models (1.6 and 1.8 liters) in Brazil. Source: CETESB, 200419

16 The use of ethanol as a fuel anti-knock additive in gasohol avoids the need of organic lead compounds addition or even the increase of aromatics content to mantain octane levels. With the introduction of the alcohol blend, lead content in gasohol was lowered from 0.8 ml/l (1980) to 0.3 – 0.4 ml/l (mid 80’s) and since 1989 no lead has been added to gasohol in Brazil. The shift from leaded to unleaded gasoline has not resulted in any major engine problem, and most important, lead ambient concentrations in the São Paulo Metropolitan Region (SPMR) dropped from 1,4 ug/m3 in 1977 to less than 0,10 ug/m3 in 1991, far below the Brazilian air quality standard, 1,5 ug/m3. 17 Denatured ethanol is the ethanol fuel added with small volumes of gasoline to avoid the human use of this fuel as an alcoholic beverage. The only way to denatured ethanol get some sulfur is from the added gasoline, since the sulfur content in ethanol is below 4 ppm. 18 Apace Research Ltd. (19398) 19 Weighed average by production volume, according to Brazilian standard NBR 6601. In 2003, for gasohol models 1.0 l engines are dominant; for ethanol, from 1.0 to 1.8 l. In 2004, for gasohol models there are engines between 1.0 l and 2.0 l; for ethanol, 1.0 l. In flex-fuel vehicles, engines from 1.6 e 1.8 l are dominant. Part of the production was tested with

33

There is an increase in exhaust emission of aldehyde when ethanol is blended to gasoline and, depending on engine characteristics and on ethanol content in the blend, exhaust emission of nitrogen oxides (NOx) may increase a little. However the significance of these two issues to air quality must be evaluated carefully to avoid misinterpretation. The case of aldehyde is a clear example. Total aldehydes emissions from alcohol engines are typically higher than those from gasoline, but the predominant acetaldehydes from the renewable fuel are less toxic than the formaldehydes from the fossil fuels (Table B3).

Healh risk parameter Acetaldehyde Formaldehyde

Risk level E-4 (1 in 10,000)

50 8

Risk level E-5 (1 in 100,000)

5 0.8

Air Concentrations at Specified Risk Levels (ug/m3)

Risk level E-6 (1 in 1,000,000)

0.5 0.08

Quantitative Estimate of Carcinogenic Risk from Inhalation Exposure (air unit risks per ug/m3)

0.0000022 0.000013

Table B3. Health risks of acetaldehydes compared to formaldehydes20. Source: EPA (2005a and 2005b).

Ambient aldehyde concentrations were measured in Denver, CO, USA, for the winters of 1987-88 through 1995-96 (before and after the introduction of E10) and no statistically significant differences were observed for both ambient acetaldehyde and formaldehyde. A study conducted by the California Air Resources Board predicted for E10 use virtually no increase for acetaldehyde ambient concentrations in 2003, relative to 1997 (when no E10 was used). Additionally, it was predicted a reduction of about 10% for formaldehyde, 30% for benzene and 45% for 1,3-butadiene. Rather, the California study identified aromatic compounds and olefins – basic constituents of gasoline as being

gasohol and the other part with neat ethanol. The largest differences due to engine size were observed in CO2 emissions. Gasohol for tests: blend of 78 % gasoline and 22 % anhydrous ethanol (v/v). Emission tests were performed according to the FTP 75 procedure. 20 Full summaries are also available. Health assessment information on a chemical substance is included in IRIS only after a comprehensive review of toxicity data by U.S. EPA health scientists from several Program Offices, Regional Offices, and the Office of Research and Development.

34

primarily responsible for the formation of formaldehyde and acetaldehyde in the air21.

Total aldehydes emissions from alcohol engines are typically higher than gasoline ones, but they are predominantly acetaldehydes, not formaldehydes. Acetaldehydes emissions produce less health effects than the formaldehydes emitted by gasoline and diesel engines22.

Recently, aldehyde emissions from high-content ethanol blends have been measured in Brazil and reach low levels. Typically, 2003 model-year Brazilian vehicles fueled with the reference blend for governmental certification (a blend with 22%v/v ethanol – E22) emit 0,004 g/km of aldehydes (formaldehyde + acetaldehyde)5, a concentration that is about 45% of the strict California limit that is required only for formaldehyde. On the other hand, emissions of aldehydes are not limited to ethanol use. Combustion of gasoline, diesel, natural gas and liquefied petroleum gas also generate aldehydes as well. Automotive use of diesel oil can be a more important source of aldehydes than gasoline-ethanol blends. Data from diesel vehicle aldehyde measurements show that emissions (formaldehyde + acetaldehyde) are 5.6 to 40.2 higher than those from vehicles running on E2223.

Nowadays NOx and VOCs may have negligible or even null increase with ethanol. Modern vehicle technology allows efficient NOx control, reducing tropospheric ozone. Compared to CNG, ethanol has less HCs fugitive emissions. Differently from gasoline, ethanol is an oxygenated fuel and therefore blending of ethanol to gasoline increases the available oxygen for combustion. This results in an effect that is called "air-fuel enlearnment" that may either increase or decrease NOx emission or even have no noticeable effect, depending on vehicle

21 The U.S. EPA has established a Reference Concentration (RfC) of 9.0 µg/m3 (5ppb) for acetaldeydes based on degeneration of olfactory epithelium in rats, and has not determined a Reference Dose. An acute non-cancer Reference Exposure Level of 3.7 x 102 µg/m3 (307 ppb) and a chronic non-cancer REL of 3.6 µg/m3 (3 ppb) are listed for formaldehyde in the California Air Pollution Control Officers Association Air Toxics "Hot Spots" Program, Revised 1992 Risk Assessment Guidelines. The toxicological endpoint considered for chronic toxicity is irritation of the respiratory system (CAPCOA, 1993). The U.S. EPA has not established a Reference Concentration (RfC) for formaldehyde. The World Health Organization Air Quality Guidelines establishes a reference maximum value of 100 µg/m3 for formaldehydes for a 30 minutes exposure period. 22 CETESB (2003) obtained in 1993 the concentration ratio acetaldehyde/formaldehyde based on ambient air monitoring data. The results were in the range of 1.7-1.8 and in 1996/1997, 1.6-2.1. Comparing these figures to the typical values encountered in Los Angeles, Atlanta and Chicago (0.18 - 0.96)’, the higher concentrations of acetaldehydes were observed in São Paulo due to the intensive ethanol use as an automotive fuel. It must be emphasized that during this monitoring campaign period, only a very small portion of the Brazilian light-duty fleet was equipped with catalytic converters - which help significantly in the reduction of aldehydes emissions. 23 Abrantes (2003)

35

characteristics and maintenance24. Another relevant aspect to be aware of is that ethanol has a higher “latent heat of vaporization” than gasoline what means that ethanol may contribute to lower combustion temperature and therefore help to reduce NOx. Also the ”solvent” effect of ethanol that contributes to clean the combustion chamber from deposits might be another factor to avoid the increase of NOx with mileage accumulation or even reduce NOx after cleanup is completed. In this regard it is necessary to recognise that cleanup of deposits may have a pronounced effect in the reduction of other important pollutants such as volatile organic compounds (VOC) over vehicle’s lifetime usage.

Ethanol-gasoline blends are frequently blamed for increased fuel volatility and this fact has been considered a difficulty because it may result in higher fuel evaporative emissions. Although it is true that blending ethanol with gasoline increases volatility25, a fair understanding of the case requires consideration of the following points that in general are overlooked. First, blend volatility varies as a function of base-gasoline composition. Gasoline types with lower concentration of light hydrocarbons will show a smaller volatility increase when blended with ethanol. Second, blend volatility varies as a function of ethanol concentration in the blend. For a certain gasoline there is a concentration of ethanol that results in a RVP peak (this peak may vary for different gasoline types). Once the peak is reached addition of more ethanol will actually reduce RVP. Depending on the ethanol content and base gasoline composition the RVP may eventually return to its original value.

On-board fuel evaporative emission control systems have been adopted in many countries and the substitution of carburetors by fuel injection systems has actually improved this emission control capability. Vehicles equipped with fuel injection systems and evaporative emission control, which have become the industry standard, are quite efficient in avoiding significant evaporative emissions26. Also, current international experience shows that gasoline vehicles fueled with ethanol-gasoline blends have in general not been affected by vapor

24 The Air-Fuel Ratio characteristic is more important for high ethanol content fuels (higher than 10%, since the influence of oxygen present in the molecule is more significant) and for the old carbureted vehicles because these engines did not have the stand-alone ability to correct the air–fuel ratio, like the electronic fuel injection (EFI) systems can do. This property of the EFI System comes from the continuous engine necessity of different air-fuel ratios (due to operation conditions and emission control) and allows to correct the air-fuel ratio in a range of ± 10%, what is enough to support ethanol blends up to 10%, even considering the vehicle emission control, performance and drivability requirements. 25 As expressed usually in terms of the standard Reid Vapor Pressure (RVP) and Distillation Curve (DC) measurements. RVP, which has been used as a popular parameter to evaluate ethanol-gasoline blends volatility, is a very limited indicator for this purpose because it is measured only at one temperature (37,8 ºC) and at an arbitrary air-to-liquid ratio of 4:1. Since fuel temperatures may vary widely as well as engine temperatures, a more precise evaluation requires an additional evaluation of the DC and other parameters such as the vapor-liquid ratio, at selected temperatures. 26 Taking Brazil as an example, fuel evaporative emission levels from 2003 model-year vehicles equipped with activated carbon canisters and fueled with a 22% ethanol-gasoline blend average an emission of 0,75 g/test (U.S SHED test procedure). This emission represents only 35% of current limit that is 2 g/test and is also adopted in most countries that have advanced emission control programs.

36

lock, an undesirable effect associated to excess vapor formation in the fuel supply system. Brazil, Australia and South Africa, countries that are subject to high ambient temperatures that may well exceed in some regions 40 ºC during summer, have not reported vapor lock occurrences with ethanol-gasoline blends27 . If necessary blend volatility can be controlled, through appropriate refinery operations, change of ethanol content in gasoline and use of additives that depress vapor formation are the main strategies that can be used in a combined and cost-effective way.

Depending on engine characteristics, reduction of exhaust emission of volatile organic compounds - VOC (frequently referred as hydrocarbons - HC) can also be accomplished. Reduction of VOC is important because VOC are both potent precursors of photochemical smog and toxic substances. A very comprehensive Australian study28 1 found that use of E10 decreased HC emissions by 12%; toxic emissions of 1-3 butadiene by 19%, benzene by 27%, toluene by 30% and xylene by 27%. The decreased carcinogenic risk was by 24%. CO emissions were reduced by 32%

The use of E10 blends to reduce harmful wintertime CO emissions has proven to be a very effective strategy in the USA. Tests at the National Center for Vehicle Emissions Control and Safety at Colorado State University document a 25% to 30% reduction in CO when automobiles burn E10. It is important to note that CO, in addition of being an important air pollutant by itself, also contributes to the formation of photochemical smog. Therefore, the reduction of CO may actually contribute to lower formation of ground-level ozone.

Regarding toxicity, ethanol has significant properties. It is a substance of moderate toxicity and, when ingested, is characterized by a depressant effect on the central nervous system. There is no evidence that repeated exposure to ethanol vapors results in cirroses of the liver. In general, it can be said that there is no evidence that the use of ethanol results in greater health risks than from the use of gasoline or diesel. In addition, ethanol is not known to be mutagenic or carcinogenic, in opposition to petroleum-based fuels, which are considered to have this characteristic, mainly due to the presence of aromatic hydrocarbons.

Studies performed with rats at the University of São Paulo have shown that ethanol combustion by-products and vapors are significantly less toxic than those from gasoline use. It is worth to mention that despite the extensive use of ethanol

27 Another valid example, although not at such high ambient temperatures, is the use of E10 at high altitude, in Colorado, USA, a situation that also favors enhanced fuel vaporization, and has not been associated with vapor lock events (CARB, 200...). 28 Apace Research Ltd. (1998)

37

in Brazil for more than two decades, occasional minor eyes, nose and throat irritation have been the only reported health effects in terms of general population and ocurred mainly in the early eighties. It is belived that the major cause was the high emission rate of aldehydes from the first generation of alcohol and gasohol vehicles. Since then, these emissions were reduced to low levels with the help of engine design optimisation, modern eletronic fuel injection systems and catalytic converters. The lack of evidence of health effects is also true for population groups that have been normally exposed to high levels of vapors and combustion products, such as traffic and highway-patrol agents, mechanics and service station workers.

Ethanol is much less toxic than methanol or common gasoline compounds such as toluene and xylene or benzene, which is also a well-known carcinogen. Although ethanol can also be absorbed through the skin, eventual inflammations are less serious than those caused by gasoline. The toxicity of gasoline-ethanol blends is not well known. In terms of skin contact, there is evidence that in case of blends, higher ethanol penetration occurs than of neat ethanol. Although skin contact with blends causes a burning sensation, this effect also acts as a warning before skin damage occurs to a greater extent. Generally speaking, similarly to neat ethanol effects, the Brazilian experience with gasoline-ethanol blends (up to 25 percent of anydrous ethanol addition, by volume) has not shown any deleterious health effects.

Impacts from ethanol agricultural and industrial production

There were indeed several environmental mistakes made in the past by the Brazilian Alcohol Program, a pioneer program. However, from the lessons learned, the process is nowadays much more environmentally sound. Also, it is worthwhile to note that environmental concern and sustainable development highlighted for the first time in 1972 and by that time it was just starting to be understood in developed countries. Nowadays, enforced environmental legislation is overcoming local pollution impacts.

Nearly 60% of all national sugarcane is produced in the State of São Paulo, the state that has the most strict environmental legislation in the country29. In Sao

29 Environmental licensing, all over the country, includes environmental control on land use (including strict control on deforestation), soil impacts, liquid and atmosferic effluents. Besides that, since 2002 Sao Paulo State law 47.397 (4th December 2002) requires the renewability of environmental permits, aiming to guarantee a better environmental enforcement and competitiveness improvements. In the case of São Paulo State, every licensing process is developed by

38

Paulo, the phase-out targets and timetables for sugarcane crop burning through São Paulo was established by the State Law 11,241/02, regulated by the Decree 47,700/03. This law establishes: (i) a timetable to the elimination of sugarcane burning. This activity will be fully eliminated in São Paulo State by the year 2031; (ii) that the entrepreneur must require the authorisation of the State Secretary for the Environment and communicate the date and hour of the burning to the State Department for Natural Resources Protection (DEPRN), The State Environmental Agency (CETESB) and the Environmental Police. Besides thie state law, also several municipalities in the State are prohibiting sugarcane crop burning prior to harvesting. With the elimination of sugarcane burning, sugarcane straws left over the field collaborate to control plagues, reducing herbicide applications by 50%.

Regarding land use, sugarcane plantations (or other crops) must provide at least 20% forestry cover on native trees (or reforested with native trees), according to the National Forestry Code30. Also, the reforestation of riparian forests is not mandatory by law, but the National Forest Code forbids the deforestation of these areas. Sao Paulo State has special requirements on riparian forests for licensing and since 2005 it was started a special program on recuperation of riparian forests with funds from World Bank/GEF.

Liquid and atmosferic effluents are also controlled. Wastes from sugar/alcohol mills are now better managed. All vinasse31 produced is utilized in sugarcane cultivated areas, eliminating the need for potash fertilizers. Filter cake32 is also totally used in the soils (eliminating 50% of phosphorus mineral fertilization needs).

Water usage in irrigation is small in Brazil, around 3.3 million hectares (compared to global 227 M Ha). Almost all Brazilian sugarcane is exclusivelly rain-fed. Water abstraction in sugarcane culture was reduced from 5 m3 / tonne cane (1990-1997) to 1.83 m3 / tonne cane (2004), in São Paulo, with intense water re-use (21 m3 / tonne cane). Wastewater treatment has efficiencies above 98%. Moreover, in the productive process of ethanol and sugar, vapour use is 500kg /tonne cane. Modern technologies reduced this figure to 300 kg vapor/tonne cane (Macedo, 2005). In Brazil, federal legislation forbids releases of vinasse and other polluting wastes in water bodies. Vinasse is now being used in environmentally

the State Secretariat for Environment and São Paulo State Environmental Agency (CETESB). Authorisation for water use comes both from CETESB and the State Department of Water and Electric Energy (DAEE). 30 Código Florestal, instituted by Federal Law 4,771/65 31 Vinasse or stillage is a by-product from ethanol distillation, extremely pollutant; however it presents a high content of minerals, being an important option to replace fertilizers in sugarcane crops. 32 A solid residue from cane juice filtration, rich in phosphorus and also used to replace fertilizers.

39

controlled fertirrigation of sugarcane fields, replacing traditional fertilizers with economic benefits. Anaerobic wastewater treatment systems are another good option, especially when attached to methane recovery. In Sao Paulo vinasse disposal is strictly controlled by CETESB, the state environmental agency.

Biocontrol, through the development of more resistant specimens of sugacane, has eliminated tonnes of pesticides from being released into the environment; sugarcane is one of the less chemical-intensive cultures. Other benefits are reduced erosions, improved topographical planning, better agricultural operations with less fuel use and lower associated costs.

In industrial processes, low-NOx burner systems and particulate filters are mandatory to reduce pollutant emissions from boilers. VOCs and odor can also be controlled by proper management. In more developed regions, like Sao Paulo State, air pollution control legislation requires innovative approaches, such as prevention at source and emission offsets through cap-and-trade systems. This is the case of saturated areas in São Paulo33. As mentioned before, bagasse is an important byproduct that supplies all needs for heat and electricity production in the mill, with carbon savings (255 kg CO2 equivalent per tonne of cane).

Impacts from ethanol transportation and storage

Ethanol is biodegradable, with low environmental persistence, with lower impacts than gasoline and mineral oils due to spill offs. In Brazil, ethanol and gasoline-ethanol blends are normally stored in closed storage tanks without any inert gaseous atmosphere, at normal ambient temperatures. Static electricity is less of a problem with ethanol than gasoline or diesel fuel. Therefore, the risk of accidental ignition is reduced. All common means of fuel transportation as pipelines, ocean tankers, inland barges, tank trucks and rail tank cars, have been safely used to transport ethanol and ethanol blends. In case of ethanol spills, risk of fire can be elminated by water, due to the miscibility of alcohol in water. Fires involving ethanol have visible flames and fire fighting does not pose any special problem. In general, ethanol handling and storage risks are equivalent to other currently used fuels, and require only common safety measures.

33 managed through State Decree 48.523 (2 March 2004)

40

The environmental impacts of neat ethanol spills or chronic leaks are expected to be much less serious than those experienced with crude oil, diesel fuel and gasoline because ethanol is miscible in water and degrades rapidly. The impact of a major ethanol spill would be in the short-term: the lenght of time during which ethanol remains in water at harmful concentrations can be measured in hours compared to years for crude oil, diesel or gasoline.

Recovery of the vegetation and population of animals and insects would occur more rapidly and completely at an ethanol spill site. Full recovery would occur within weeks, in contrast to months or years required for recovery from crude oil, diesel or gasoline spills. Clean up costs and other economic losses should be smaller for alcohol spills. In the event of an alcohol–gasoline blend spill, the environmental impacts are expected to be similar to those of a gasoline spill, and depend primarily on the volume of gasoline in the discharge.

41

ANNEX 3. LAND USE SCENARIOS

Regarding land use and the perspectives for ethanol production in developing countries, some considerations are important. Considering the agricultural area currently available in the world, it is important to note that the share of sugarcane areas harvested is very small compared to other primary crops – and even smaller if compared to total agricultural land, where cultures are already planted (Table C1). It is also significant the fact that even low income and African countries present agricultural efficiency lower than the world average and much lower than Brazil, showing some potential for improvements in agricultural sector.

Agricultural area (million ha) 2002

Primary crops area harvested

(million ha) 2004

Sugarcane area harvested (million ha)

2004

Sugar cane production (million

tonnes) 2004 Sugar cane yield (tonnes /ha) 2004

World 5012 1042 20.3 1324 65.2

Developed Countries 1826 329 1.1 84 76,4

Developing Countries 3186 711 19.2 1240 64.6

64 Low-Income Countries 1381 395 7.1 396 55.8

Africa 1111 164 1.4 85 60,7

Brazil 264 54 5,6 411 73.5

Table C1. Sugarcane in the world, 2004. Source: FAO, 200534

On the other hand, also according to FAO (200535), the worldwide area with potential for agricultural crops is 13.4 billion hectares (ha). From this total, 80 million hectares are considered “very suitable” and 303 million ha are “suitable”, 516 million ha as “moderately suitable” and more 556 ml ha “marginally suitable”. This assessment does not separate sugarcane from other crops, but there are

34 Source: FAOSTAT (2005) http://faostat.fao.org/faostat/form?collection=Production.Crops.Primary&Domain=Production&servlet=1&hasbulk=0&version=ext&language=EN 35 http://www.fao.org/ag/AGL/agll/gaez/ds/ds.htm

42

regions where sugarcane is the predominant crop. This is the case of South America, Central America and the Caribbean, Africa, Oceania and Polynesia, where there are conservatively (ie, only “very suitable” and “suitable” areas) 136.5 million hectares of potential sugarcane areas. Although North America can produce both sugarcane and sugar beet, the figures from the first culture are already impressive. In this context, it is important to know, from the world arable land, which area would be necessary to produce biofuels to be blended to gasoline. Table C2 provides an idea on this issue. It shows potential land to produce sugarcane crops aiming to produce ethanol to be blend in a proportion of 10% (in volume basis) to gasoline.

Unit World

OECD

countries

Non-OECD

countries

Gasoline consumption Billion litres/yr 1165 838 327

Ethanol 10% blend (a) Billion litres/yr 175 126 49

Sugarcane area necessary for E10 Million ha* 29 21 8

"Suitable” and “very suitable" sugar crops (FAO) Million ha 383 116 217

All sugar crops (all cultures, FAO) Million ha 1455 496 959

Table C2. Land use with to produce ethanol to be blend to gasoline in 10% (volume) basis) (E10 ). Source: FAO, 200536 Note: conservatively considered (a) 6,000 liters of ethanol/ha.yr; LHV (gasoline) = 33MJ/liter, LHV (ethanol) = 22 MJ/liter

From the Table C2 it can be seen that considering the blend of 10% (in vol) of ethanol to gasoline the sugarcane area necessary would be 29 million hectares, much less than the 383 million hectares of suitable and very suitable sugar crops. Another interesting scenario (Table C3) is the assumption of the expansion of ethanol by 10% p.y. by 2015, in order to catch up with a growth of 1% per year of gasoline consumption37.

36 Source: FAOSTAT (2005) http://faostat.fao.org/faostat/form?collection=Production.Crops.Primary&Domain=Production&servlet=1&hasbulk=0&version=ext&language=EN 37 Always remembering the co-benefit is the electricity production from bagasse and top leaves

43

World 2004 2015

Sugarcane Area Harvested (million ha) 20.3 47.3 a

Sugarcane Yield (tonnes/ha) 65.3 72.8 b

Ethanol production (billion litres) 16.0 44.0 c

Gasoline equivalent production (Mtoe) 8.4 23.1 d

Electricity generation (TWh/yr) 26 138 e

Gasoline consumption road transport (Mtoe) 849.8 948.1 f

Ethanol/gasoline 1.9% 4.6%

Electricity production TWh/yr 16017 17870 h

Cane electricity/ total 0.2% 0.8%

Table C3. Introduction of E10 (calculations based on IEA, 2003 and FAO, 2005 data). Assumptions: (a) 8%/yr increase in sugarcane; (b) 1%/yr increased yield; (c) annual yield increase per tonne of sugarcane 0.5% and 2004 data from F.O. Licht (apud http://www.distill.com/berg/); (d) ethanol 22 MJ/litre (1Mtoe = 2.388 x 10E-5 TJ); (e) based on high pressure steam turbine, conservative yield, linear growth from 20 kWh/ tonne cane to 40 kWh / tonne cane; (f) 2001 baseline 824.8 Mtoe (IEA,2003); 1%/yr increase; (f) 2001 baseline 15546 TWh (IEA,2003); 1%/yr increase. Data sources: IEA (2003) and FAOSTAT (2005)