|No 50 Em Português|
Carbon Balance in the Transformation Centers
Carlos Feu Alvim
Omar Campos Ferreira
In the No 48 issue of this periodical were published the first results of the study made under the agreement between the MCT and Economy and Energy and they were presented to that Ministry in February 2005. The carbon atoms are conserved in all steps of its cycle. This is also true in what concerns energy in the National Energy Balance (BEN) because what happens is the transformation from one form of energy into another and the energy used in the transformation is accounted for in the energy sector.
The transformation centers are not treated in a homogeneous way in BEN. In some cases it is possible to carry out separately the carbon balance of the raw material and its sub-products (emissions are accounted for in the Energy Sector as a whole); in other cases, emissions must be calculated by the center itself. In what follows, the structure and nomenclature adopted by BEN are listed.
Types of Transformation Centers
BEN has the following types of transformation centers:
· Petroleum Refineries,
· Natural Gas Plants,
· Gasification Plants,
· Coke Plants,
· Charcoal Plants,
· Public Service Power Plants,
· Auto-producers Power Plants,
· Nuclear Fuel Cycle,
· Other Transformations
From the carbon balance point of view we can distinguish three (or four) types of transformation:
· Transformation units where carbon enters as a component of the raw material and comes out in the form of a product where the emission of gases containing carbon is not accounted for; in this type of transformation there is no gas emission to the atmosphere (the first five types above);
Finally, there are units such as hydroelectric and nuclear generation plants where there is no carbon both in the input and output and which should not be considered in the carbon balance. The emissions from ancillary equipment of these units such as diesel generators and others should be accounted for in other units such as those corresponding to the Energy Sector.
We will verify that in transformation units of the first type there is an accounting artifice where the emissions are accounted for in the Electric Sector.
In what follows some types of transformation centers will be presented in order to illustrate the adopted approach. It is interesting to notice that even in the case of some transformations where emissions do not appear in the accounting, the balance calculation can be useful because in the process it is verified the coherence between the input data (used in the approach that accounts for the intermediary products) and the results obtained from coefficients determined from the national emission calculation inventory that is assumed to be based on emission measurements (bottom-up) evaluated for a fuel type in the sector, except for cases in which the lack of data has required the use of generic coefficients for its determination.
Figure 1 presents the calculation scheme of Carbon Balance in a petroleum refinery.
The petroleum compounds, used as energy source in the refinery itself, are accounted for as product (output) even though energy consumption and emissions occur physically in the refinery itself. The part consumed in the refinery is recorded as “input” in the Energy Sector accounting, where emissions are recorded as well. The carbon balance must equate for the system composed by the refinery and the Energy Sector.
A similar case occurs in the (humid) natural gas processing units where some liquid sub-products are extracted and that can be directly incorporated to the commercialized products as liquefied petroleum gas (LPG), gasoline or naphtha, a fraction that is treated by refineries and natural gas denominated “dry” (fundamentally methane and ethane).
The “Energy Sector” entity is still used to record all consumption in units connected with it – including transformation centers. This consumption does not include the ‘transformed” energy which is treated in the corresponding accounting centers.
Figure 1: Balance regarding refineries and the Energy Sector
It is also interesting to notice that for petroleum and its “non energy” products that are part of the energy balance, this procedure will be followed in the carbon balance. The fact that the treatment is not uniform for all sources requires some methodological adaptations that will be mentioned in due time.
In Figure 2, it is shown the scheme adopted by BEN for distilleries and the carbon balance calculation is illustrated.
Figure 2 Carbon balance scheme for distilleries, showing that the “sugarcane products” (BEN’s denomination) are considered as different forms of primary energy. Bagasse is accounted for in the Energy Sector but vinhoto (waste by-product from alcohol distilling) is not explicitly mentioned in BEN. Molasses (sugarcane concentrate present in distilleries annex to sugarcane plants) is also considered as primary energy.
In coke plants (coke production from mineral coal) carbon is accounted for as coke products and the gas that is partially consumed in the process is accounted for in the energy sector. So, it is hoped that the carbon balance can be carried out at this level.
It is hoped that residence gas production (a more noble gas than natural gas distributed for consumption) can also be treated as a carbon balance center. The diversity of raw materials used along the years in this process and the variation of their composition should however cause difficulties in the accounting of this type of unit. It should be remembered that this fuel is disappearing in Brazil because of the dry natural gas that is directly distributed to consumers (adaptations in the network and in the appliances were necessary)
In the case of charcoal plants the primary energy (firewood) is transformed into charcoal through partial burning. There is no intermediary product that is adequate to transfer consumption to the energy sector (a part could be credited to carbon monoxide – CO, but this is already done). Furthermore, all gas is used in the transformation process or directly emitted. The carbon balance must be calculated by adding the carbon contained in the product to that constituting the emissions, as indicated in Figure 3.
Figure 3: Carbon Balance in a Charcoal Plant.
For electricity generation plants the input values corresponding to the fuel used are available. The carbon balance – which we are dealing with – can be calculated in these units in the same way that it will be done in the consuming sectors because the product (electricity) does not contain carbon. The carbon balance scheme is shown in Figure 4.
Figure4: Carbon balance scheme in electricity plants; in the case of auto-producers the balance unit can be “virtual”, located, for example, in an industrial plant. The hydroelectric and nuclear input and the “electricity” product that are part of the energy balance do not contribute to the carbon balance The dashed lines products indicate, in this case, the values not included in the balance. In the case of losses, they should be included in the differences found in the carbon balance.
Based on what was described above, the carbon balance calculation in some transformation units does not involve emissions and it can be carried out from the results of the contained carbon and they will be presented in what follows for the 1970-2002 period.
For five transformation centers it is possible to calculate the carbon balance before emissions. The main centers are petroleum refineries.
Table 1 shows the energy and carbon balance for 2002 and also the coefficients in tC/toe (toe of 10000 kcal) or in tC/TJ (1 cal = 4.1855 J).
The carbon balance value has the same precision as that of the energy balance. This can be seen in Table 2; this is also true for the previous years. Furthermore, the carbon balance is negative, which is compatible with transformation losses that are not recorded in BEN. For the petroleum processed between 1970 and 2002 the carbon balance has a difference of only – 0.8% and an average quadratic deviation of 1.0% in the annual values.
Table 1: Energy and Carbon Balance in Petroleum Refineries in Brazil in 2002
In Table 2 are presented the carbon balances for refineries in chosen years. Besides the round years (final of decade) the 1990 and 1994 (extreme years of the inventory calculation), 1999 (year when there exist some evaluations) and 2000 (last calculated year) years were chosen.
Table 2: Carbon Balances for selected years (Carbon Mass in thousand Gg)
Carbon and energy balances have, besides that, a very similar behavior along time (Figure 5) what indicates that the coefficients used seem to be valid, in spite of the large variations of the fuel characteristics used in Brazil (mainly in what concerns diesel oil) in the studied period. This change in the fuel characteristics occurred mainly in the period of the petroleum crisis when alcohol fuel (substituting gasoline) and the vigorous process of fuel oil substitution caused an increase in the diesel oil participation in petroleum products consumption. Furthermore, there was (and still there is) a substantial advantage in the price of this fuel by traveled kilometer. Figure 5 shows that in spite of the fuel variations along time, the carbon balance is very similar to that of energy. That is, the variations in the carbon balance are due mainly to variations of the energy accounting and not to the parameters relative to the carbon content.
It should be noticed that BEN presents data concerning carbon content of fuels along the years (toe per m3 or kg). These data are used in the calculation of carbon content and partly correct these variations. The same may occur when – as it happens in the case of humid natural gas – the transformation from “natural units” (of mass or volume) uses a constant coefficient for converting energy along the years.
Figure 5: The energy and carbon balances show a similar behavior along the years and a low mass deviation. This indicates the good choice of the carbon content coefficients and the little influence of the different characteristics of fuels in the refineries’ carbon balance.
In the NGPU liquids are extracted from raw (humid) natural gas which condensate at room temperature and (dry) natural gas is produced, mainly methane and ethane. The liquid fractions can be directly incorporated to some products (LPG, naphtha, etc) or processed in refineries. As can be observed in Figure 6, the last option seems to be the preferential destination in recent years, probably because it facilitates the homogenization of the commercialized products and simplifies the NGPU operation.
Figure 6: From1992 on about half of the NGPU products in Brazil are constituted by the fraction of natural gas condensates that are listed in BEN as refineries raw material inputs.
The energy and carbon balances are shown in Table 3 for 2002, as well as the factors used to calculate them.
Table 3: Energy and Carbon Balances in the NGPU in 2002.
Table 4: Carbon Balance in chosen years (mass in Gg)
In Figure 7 it can be observed than for most of the years the energy balance does not present deviations above expectations except for the last three years. This variation can be considered important and should be due to inadequate values of the lower calorific value in the energy balance itself. Actually it is expected that in the graphic representation the variations due to the energy balance are shown in “parallel” (same direction) curves. In what concerns carbon content, it would be shown in approximation or distance of the curves. This fact (approximation of the curves) seems to be the case in the mentioned figure and it can be explained by the variation of carbon content in the humid gas.
Figure 7: Carbon and energy balances in the NGPUs in Brazil, showing that some deviations in what concerns carbon can be due to the differences of that of energy. It seems also that there is a systematic shifting component between the two balances due to inadequate carbon coefficients.
In 2002 the quantity of processed gas in the NGPUs (6,7 thousand Gg) corresponds to 4.6% of the total carbon mass of the fuels used (gross internal offer) in Brazil. Errors around 6% (as those of 2000) represent 0.3% inaccuracy on the whole. As natural gas is a fuel of growing importance, it would be necessary to pay more attention to the subject .
Carbon Balance in the Coke Plants
Metallurgical coal is converted into coke (used for steel fabrication) in a distillation process without oxygen. As sub-product are produced gases, which are used as fuel in steel plants and in the coking plant itself, and liquids. The energy and carbon balances are shown in Table 5 for the year 2002.
Gases are listed in the balance as coke plants gas and the liquids are listed as tar. BEN has opted to consider them as fully consumed in the steel industry (and nothing in the energy sector). Since a specific coefficient is not available, it was chosen the default coefficient for petroleum for the inventory. The same coefficient was used for carbon balance.
Carbon balance for the year 2002 is negative as expected since losses are not accounted for. A 6% deviation, as those observed in the carbon and energy balances, point out to inadequate coefficients or errors regarding raw materials or products calculations. From the point of view of carbon emission importance, the process involves about 7.4 million tons of carbon that represent 4.7% of the carbon gross internal offer. A 6.3% error in the carbon balance means a 0.3% error in the emissions. It should be remembered that, at least in the present case, the error is in the energy balance itself.
Table5: Energy and Carbon Balance in Coke Plants2002
The carbon balance for the chosen years are shown in Table 6. The evolution of energy and carbon balance in coke plants in the 1970 / 2002 period can be seen in Figure 8 .
Table 6: Carbon Balance in Coke Plants in Chosen Years
Figure 8: Carbon balance shows a systematic error probably due to inadequate coefficients concerning carbon mass/energy ratio. Balances are also important in the energy source but the differences can be imputed, at least in part, to losses.
Data of Figure 8 confirm problems concerning carbon balance that could be assigned to the use of inadequate coefficients.
The use of residence gas in 1970 was practically restricted to Rio de Janeiro and São Paulo. Gas for distribution in the existing network was produced in gasification plants. The raw materials used in its fabrication changed between 1970 and 2002 from mineral coal (most of it metallurgical) to naphtha (petroleum product) and finally to dry natural gas. The availability of natural gas for distribution resulted in a residual production in these plants in 2002 and they tend to disappear.
Figure 9: Raw material used in gas production in gasification plants changed from mineral coal to naphtha and from naphtha to natural gas. Since the latter was available, the distribution network began using dry natural gas and the gasification plants are practically deactivated.
Figure 10 shows that the carbon balance is entirely unsatisfactory since deviations greater than 70% and up to 150% can be observed. The energy balance quality also deteriorates along time.
Figure 10: Carbon balance in the gasification plants shows large values along the period, due to the bad quality of the coefficients used. Evolution along years (see previous figure) makes one suppose that the coefficients used for naphtha and natural gas are worse than that of mineral coal.
Energy and carbon balances are shown in Table 7 for 1990
Table 7: Energy and Carbon Balances in Gasification Plants in 1990
Examining the coefficients of Table 7, it seems probable that the produced gas (residence gas) has its composition considerably changed when the raw material is changed. The gas produced from coal could contain carbon monoxide, and this would explain the high value of the coefficient used. When this gas was substituted, in order not to change the characteristics of the distributed gas, it was chosen to deplete the gas produced from naphtha and from dry natural gas. If, for example, the gas produced from natural gas comes from adding inert gas, the C mass/energy ratio would be much closer to this value relative to natural gas (15,3 tC/TJ) than to the value used (29,5 tC/TJ), which would be adequate for a gas with high carbon monoxide content and this coefficient would then be the adequate one.
In 1990, the carbon mass involved in the transformation (input) would be 245 Gg for a gross internal offer of about 120 thousand Gg, corresponding to 0.2% of the fuel mass used. A 50% error in this evaluation would correspond to 0.1% of the total emissions. Since we are dealing with a fuel that is ending, this imprecision in its evaluation, even though uncomfortable from the methodological point of view, would have no impact on the future emissions. The impact on past emissions should not be higher than the value calculated for 1990.
The alcohol sector and hydroelectric generation are the main Brazilian producers of renewable energy. Even though the CO2 produced by biomass is not computed in the greenhouse effect emissions, the same is not true for methane emissions, which are accounted for. Furthermore, when they are considered as carbon dioxide equivalent using the Kyoto Protocol equivalence (GWP – Global Warming Potential) they can significantly reduce the positive impact assigned to using alcohol. The carbon balance in this type of unit is relevant relative to the Brazilian policy demonstration for attenuating the greenhouse effect.
The carbon balance in the alcohol plants for 2002 is shown in Table 8 and presents a negative balance of 27% in carbon mass. The same happens for the selected years, as can be seen in Figure 11 and Table 9.
Table 8: Energy and Carbon Balances in Distilleries in 2002
It is expected that there would be a negative balance in the distilleries, since vinhoto, that contains organic compounds, is not accounted for. Since this waste can contribute to emissions, including methane, it would be convenient to include it in the carbon balance.
On the other hand, the emission coefficients used for sugarcane juice and molasses were based on generic values recommended for biomass sources in the IPCC methodology. The anhydrous and hydrated alcohol coefficients as well – used in the inventory Top-Down approach – were based on the average emission factor of the national fleet calculated from measurements made by CTESB in vehicles.
As we have a compound of known composition, the emission factor – that correlates the carbon mass with energy – can be obtained with good approximation from the alcohol composition .. Using this factor, based on pure ethylic alcohol, it would produce a balance with better approximation.
Figure 11: Energy and carbon balance in distilleries that present a large carbon deficit in the formed product.
Table 9: Carbon Mass Balance in Distilleries in Selected Years
Sugarcane – not considering left-over and leaves – is now the second largest primary energy source according to BEN’s criterion, corresponding to 13% in energy, and it has exceeded firewood and hydraulic energy (calorific correspondence). The carbon mass participation is even larger (18% of the total). The carbon and energy balance treatment should deserve more attention in future approaches.
In other transformation centers (charcoal kilns and electric power plants) the emissions of gases liberated to the atmosphere are accounted for in the units themselves and it is necessary to quantify them in order to determine the carbon balance. This will be shown in the next issue of this periodical together with other consuming centers.
Carbon balance in transformation centers was useful for detecting problems regarding the calculation of carbon gases emissions. The most important problems detected concern the biomass area, specially alcohol production. Furthermore, alcohol presents an apparent error in the carbon content used in the Brazilian inventory that considers only the emitted CO2.
The coefficients used in the first calculation of the national inventory were most of them generic coefficients for solid and liquid biomass. Due to the importance of the alcohol sector in the Brazilian energy matrix, a more precise treatment would be desirable. It should be noted that, in spite of the fact that CO2 emissions are not accounted for, those of methane are and they could, for example, reduce the eventual carbon credit due to the use of alcohol engines. The carbon and energy balances in refineries present good results regardless of the large variation in the composition of petroleum products in Brazil after the 1979 petroleum prices shock.
As energy “is not conserved” in transformation centers accountability due to the non registered inherent losses in the processes, BEN does not present systematic energy balance calculation in these centers. However, for some of them (where consumption is calculated in the energy sector) there are apparent gaps in the coefficients that can be corrected in the future editions. An example is the use of the same energy/volume coefficient for humid natural gas in different years in spite of the fact that the percent value of liquid products obtained in the NGPU are different for various years.
In the case of refineries, because it regards chemical reactions, where heat generation (consumed energy) and the transformation change of hydrocarbons constitution do not have a defined threshold. The specification of an intermediary product (as fuel oil) in order to be accounted for as a refinery product and energy source used in the energy sector is actually an accounting artifice used to organize an energy balance. In fact, the advantage of introducing the energy sector in the (energy) balance is that it makes the transformation centers accounting more elegant by accounting losses as consumption in the energy sector.
 The same thing does not happen, for example, with sugar which can be considered as a “non energy” sugarcane product (at least from the energy balance point of view).
 Concerning natural gas, problems regarding the apparent inconsistency about the difference between the high and low calorific values and the expected carbon content were identified . It should also be noticed the consistency of the calorific values along time. These values should in principle vary according to the origin of the natural gas (contained products that are condensable at room temperature).
 Using the previous years, the year 2002 presents a low quantity of considered raw material and only of one type (natural gas).
 The lower calorific value (LCV) for ethylic alcohol in the handbook is 6621 kcal/kg or 0.027712 TJ/t. This corresponds to 36.085 t/TJ. For C2H5OH we have 24/46 x 36.085 tC/TJ or 18.8 tC/TJ. The LCV considered by BEN for anhydrous alcohol is 6750 kcal/TJ and consequently, 18.5 tC/TJ
Archives in pdf:
Graphic Edition/Edição Gráfica:
Tuesday, 11 November 2008.