CO2 emissions accounting for the chemical industry: an empirical analysis for China


The chemical industry is one of the most important industry sectors in terms of energy consumption and CO2 emissions in China. However, few studies have undertaken accounting of the CO2 emissions in the chemical industry. In addition, there are some shortcomings in the traditional accounting method as a result of poor data availability, such as the incomplete consideration of emission sources and overestimation of actual emissions. Based on the traditional accounting method and the actual situation of the chemical industry, this study proposes a method called the Emission Accounting Model in the Chemical Industry, which covers fossil energy-related emission, indirect emission generated by electricity and heat, carbonate-related process emission and the reuse of CO2. In particular, fossil energy used as feedstock is included. By applying the Emission Accounting Model in the Chemical Industry in China, the calculated CO2 emissions would be 19–30% less than the result from the traditional method. In addition, it is found that the indirect CO2 emissions generated by electricity and heat account for 67% of the total amount, the fossil energy-related emissions account for approximately 37%, the process-related emissions accounted for 2%, and reuse of CO2 accounts for − 6% in 2016. The production of ammonia, ethylene and calcium carbide generated approximately half of the total CO2 emissions in 2016. In addition, in view of emission sources and carbon source flow, two other bottom-up accounting methods are proposed that can take effect when the chemical plant-level data are available.


The energy consumption of the chemical industryFootnote 1 has generally increased from 143 million ton coal equivalent (Mtce) in 2000 to 480 Mtce in 2016. As shown in Fig. 1, the energy consumption of the chemical industry increased rapidly from 2001 to 2007, with an annual growth rate of 15.0%. After a decline in 2008, the energy consumption increased by 6.5% annually from 2009 until 2015 while dropping slightly in 2016. This makes the chemical industry one of the most important industry sectors, the energy consumption of which accounts for 17.2% among all industry sectors and 11.0% of national consumption in China in 2016.

Fig. 1

Energy consumption of chemical industry in China (2000–2016). Notes: The size of the ball represents the proportion of the chemical industry’s energy consumption in total amount of industry sectors

The large quantities of energy used in the chemical industry are accompanied by a large amount of CO2 emissions. In the context of climate change, CO2 emissions need to be drastically reduced to achieve the goal of temperature control in the future and China in particular, which has the largest CO2 emissions in the world, will hugely influence it (Wei et al. 2018). This calls for numerous research studies on carbon quota allocation (Li and Tang 2016; Raupach et al. 2014; Zhou and Wang 2016) and the adoption of a carbon tax (Martin et al. 2014) to aid in mitigation. In addition, China plans to introduce carbon markets for key industrial emission sectors (e.g., electricity, iron and steel, cement, chemicals and so on). However, there is no doubt that relatively accurate emission accounting is the basis of carbon pricing and policy design for industrial sectors. Currently, in terms of sectoral CO2 emission accounting and emission reduction, most studies focus on electricity (Chen et al. 2016, 2017; Tang et al. 2018), iron and steel (An et al. 2018; Worrell et al. 2001), transport (Selvakkumaran and Limmeechokchai 2015) and cement (Zhang et al. 2018). However, the chemical industry seemingly lacks such related studies (Broeren et al. 2014), mainly due to its numerous products and complex production processes, which make it a complicated system and difficult to conduct comprehensive accounting (Zhu et al. 2010). In addition, the poor data availability in particular has enlarged the difficulties relevant to the chemical industry (Broeren et al. 2014; Griffin et al. 2017).

Because the chemical industry is one of the most important CO2 emitters in industrial sectors, research on its emissions in chemical industry should be given more attention, and first, it is essential to adopt an accurate method to undertake CO2 emission accounting when there is poor data availability. On the one hand, a relatively accurate estimation of the quantity of CO2 produced by the chemical industry can be obtained by emission accounting, which will benefit research based on the chemical industry; for example, with more accurate emissions, the carbon quota allocation would be fairer for carbon market construction. On the other hand, the important emission sources can be identified by CO2 emission accounting, which will aid when taking action to reduce emissions from the perspective of energy use and technological improvement. British Petroleum (BP) and the International Energy Agency (IEA) have accounted for national CO2 emissions based on their energy statistics, including China (BP 2017; IEA 2017). However, they only consider fuel combustion-related emissions, excluding emissions from other sources such as feedstock use, process emissions and so on. In addition, it did not show specific industry sectors or chemical industry.

In the existing studies of CO2 emission accounting for the chemical industry, one of the biggest problems is oversimplification as a consequence of the poor data availability. The CO2 emissions for the chemical industry mainly come from fossil fuel combustion, feedstock-related processes, electricity and heat use, carbonate decomposition-related processes and the reuse of CO2 in fertilizer production. Among them, fossil fuel combustion and feedstock-related emissions can be viewed as fossil energy-related emissions. However, for the existing method, fossil energy-related emissions are usually accounted by energy use multiplying their emission factors, in which the energy consumption used as feedstock is viewed as fuel to be combusted. Zhou et al. (2010) adopted this method for assessing the reduction potential of the chemical industry. However, in fact, the carbon content in fossil feedstock is transformed into carbonaceous chemical products and residues but not in the form of CO2. This will overestimate the actual CO2 emissions as a consequence of the production of large amounts of carbonaceous products. One of the most important reasons is data availability, as the energy statistics do not distinguish energy used as fuel and feedstock. Thus, Saygin et al. (2009) excluded the corresponding CO2 of key polymers in estimating the reduction potential of the chemical industries of key chemical manufacturing countries, only considering direct fuel use-related emissions and ignoring other emission sources. Process-related emissions and the reuse of CO2 are rarely mentioned in existing research. Because fertilizer is a downstream product of ammonia, Zhu et al. (2010) considered the CO2 reused by fertilizer production in the accounting of ammonia emissions. However, only ammonia-related CO2 emissions are accounted for in that research, not the entirety of emissions from the chemical industry.

Thus, it can be found that there is incomplete consideration of emission sources in existing research, which always accounted for part of the emission sources or simplified the emission accounting. Considering the necessity of emissions accounting for the chemical industry and the shortcomings of existing methods, this study tries to propose a more comprehensive method, the Emission Accounting Model in the Chemical Industry (EACI), by improving the existing methods to reflect the actual CO2 emissions of the chemical industry. We take China as the empirical context, and three questions will be answered.

  1. 1.

    With the current data availability, how can more accurate CO2 emission accounting be undertaken for the chemical industry?

  2. 2.

    How much CO2 emissions have been produced by the chemical industry in the past and what is the contribution of different emission sources?

  3. 3.

    How should CO2 emissions be calculated if chemical plant-level data are available?

This paper makes two main contributions to CO2 emissions accounting for the chemical industry when compared with a limited number of former studies.

  1. 1.

    This study comprehensively considers diversified emission sources in chemical production, including fossil fuel combustion, fossil feedstock-related processes, carbonate decomposition processes, the reuse of CO2 and the indirect emissions of heat and electricity.

  2. 2.

    To solve the problem of accounting caused by poor data availability for the chemical industry, especially in cases with industry-level statistical data, this study proposed a CO2 emission accounting method called EACI to approximate the actual emissions. In the empirical analysis, EACI is applied to China, which accounts for most of the world’s chemical market sales, its share reached up to approximately 40% in 2016 (Cefic 2017). However, for cases in which chemical plant-level data are available, this study illustrated two bottom-up methods, one of which is from the perspective of carbon emission sources and the other being from the perspective of carbon source flow.

The remainder is organized as follows: Sect. 2 presents the Emission Accounting Model in the Chemical Industry (EACI) proposed by this study, the data and parameters. The key CO2 emission results obtained by utilizing the traditional method and the EACI model are explained in Sect. 3. Section 4 discusses the advantages and disadvantages of EACI and proposes two methods based on the accounting of the chemical production enterprise. Conclusions and policy implications are drawn in Sect. 5.


The CO2 emissions of the chemical industry mainly come from six parts, which are fuel combustion emissions, fossil feedstock-related emissions, the indirect emissions generated by using electricity and heat, process emissions produced by carbonate and the reuse during the production of some chemicals. Thus, the accounting of CO2 emissions in the chemical industry should reflect these parts directly or indirectly.

Different from other industry sectors, fossil energy always plays two roles in the chemical industry: In one role, fossil energy powers the plants, and in the other role, fossil energy is used as feedstock to produce chemical products such as oil that can be used as steam cracking feedstock to produce olefins and coal that can be used as gasification feedstock to produce methanol. In general, only when the quantities of energy used as feedstock and used as fuel could be obtained separately can the CO2 emission accounting be accurate. In fact, however, the public energy statistics for the chemical industry do not have separated data, which makes it hard to calculate the fuel combustion emissions and energy feedstock-related emissions directly. Generally, when faced with poor data availability, all fossil energy can be viewed as fuel to combust if high accuracy is not essential (Zhou et al. 2010). In addition, the indirect CO2 emissions produced by purchased electricity and heat use should also be considered, as shown in Eq. 1. This method is defined as the traditional method in this research.

$${\text{TT}}_{t}^{{{\text{CO}}_{2} }} = \sum\limits_{k}^{{}} {E_{k,t} \cdot {\text{EF}}_{k,t} } + {\text{EM}}_{t}^{\text{elec}} + {\text{EM}}_{t}^{\text{heat}}$$

However, the traditional method has two main disadvantages. First, energy used as feedstock is ignored and taken as fuel to combust, which will overestimate the results. Second, some other emission sources such as process emissions and the reuse of CO2 are not considered, which makes the accounting incomplete. To avoid the two disadvantages and make the CO2 emission accounting more accurate for the chemical industry, the emission accounting model in the chemical industry (EACI) is proposed and utilized in this study. It should be noted that the EACI model tries to approximate the real emissions.

The key parameters and their definitions in this study are shown in Table 1.

Table 1 Key parameters and definitions

Emission Accounting Model in the Chemical Industry (EACI)

The core of the EACI model is to use an approximate method to calculate the total emissions associated with fuel combustion and feedstock based on the energy statistics of the chemical industry. To perform the CO2 emission accounting, this model considers fossil energy-related emissions, process emissions associated with carbonate use, indirect emissions from purchased electricity and heat, and the CO2 reuse in nitrogen fertilizer production (Eq. 2). Among them, fossil energy-related emissions (\({\text{EM}}_{t}^{\text{fossil}}\)) include emissions from fossil fuel combustion and fossil feedstock-related emissions. For fossil energy-related emission accounting, a two-step strategy is adopted. In the first step, it is assumed that all energy consumption is used as fuel for combustion and produces CO2 emissions (\(\sum\nolimits_{k} {E_{k,t} \cdot {\text{EF}}_{k,t} }\)). In fact, some of the carbon content in fossil feedstock is used to produce chemical products, which means that the corresponding carbon content used is not discharged into the atmosphere in the form of CO2. Thus, on the basis of emissions calculated by step 1, step 2 will count that carbon content and remove its corresponding CO2 (\({\text{EM}}_{t}^{\text{fixat}}\)) (Eq. 3).

$$T_{t}^{{{\text{CO}}_{2} }} = {\text{EM}}_{t}^{\text{fossil}} + {\text{EM}}_{t}^{\text{process}} + {\text{EM}}_{t}^{\text{elec}} + {\text{EM}}_{t}^{\text{heat}} - {\text{EM}}_{t}^{\text{reuse}}$$
$${\text{EM}}_{t}^{\text{fossil}} = \sum\limits_{k}^{{}} {E_{k,t} \cdot {\text{EF}}_{k,t} } - {\text{EM}}_{t}^{\text{fixat}}$$

The emissions from fossil fuel combustion are determined by both the quantity of fossil fuel (\(E_{k,t}\)) and its corresponding CO2 emission factors (\({\text{EF}}_{k,t}\)), as shown in Eq. 3. Among all the fossil fuel categories, the consumption of lubricants, paraffin waxes, white spirit and bitumen asphalt is excluded from the CO2 emission accounting. This is because those kinds of energy are usually neither used as fuel to combust nor as feedstock to produce basic chemicals, which will not produce CO2 emissions during their use. The CO2 emission factors of energy combustion are determined by lower heating value (\(h_{k,t}\)), carbon content (\(C_{k,t}\)), carbon oxidation rate (\(f_{k,t}\)) in combustion and so on (IPCC 2006), as shown in Eq. 4.

$${\text{EF}}_{k,t} = h_{k,t} \cdot C_{k,t} \cdot f_{k,t} \cdot \varphi$$

To calculate fossil energy-related emissions, the CO2 emissions corresponding to carbon content in carbonaceous chemical products (\({\text{EM}}_{t}^{\text{fixat}}\)) should be excluded (Eq. 5). The accounting for \({\text{EM}}_{t}^{\text{fixat}}\) can be divided into two steps. First, the carbon content in carbonaceous chemical products should be calculated, which is determined by their yields (\(P_{i,t}\)) and the carbon content in per unit production (\(\theta_{i,t}\)). Then, the corresponding CO2 can be derived based on the total carbon content and the coefficient of carbon transferring to CO2 (\(\varphi\)). Considering the practical condition of the chemical industry and its data availability, the carbonaceous chemical products are selected, including olefins (ethylene, propylene and butadiene), BTX (benzene, toluene and xylene), carbon black, methanol, calcium carbide and so on, due to their high yields and high carbon contents.

$${\text{EM}}_{t}^{\text{fixat}} = \sum\limits_{i}^{{}} {P_{i,t} \cdot \theta_{i,t} } \cdot \varphi$$

Process emissions (\({\text{EM}}_{t}^{\text{process}}\)) in this study refer to CO2 emissions generated by carbonate decomposition in chemical production, which can be calculated by the process emission factors (\(m_{l,t}\)) multiplied by their quantities (\(Pr_{l,t}\)), as shown in Eq. 6.

$${\text{EM}}_{t}^{\text{process}} = \sum\limits_{l}^{{}} {Pr_{l,t} \cdot m_{l,t} }$$

Although purchased electricity and steam are not produced in chemical plants and the chemical industry does not produce those CO2 emissions directly, in view of energy end use, this study also considers the indirect emissions generated by them (\({\text{EM}}_{t}^{\text{elec}}\) and \({\text{EM}}_{t}^{\text{heat}}\)), which are calculated by their emission factors (\(h_{t}^{\text{elec}}\) and \(h_{t}^{\text{heat}}\)) and quantities (\(E_{t}^{\text{elec}}\) and \(E_{t}^{\text{heat}}\)), as shown in Eqs. 7 and 8.

$${\text{EM}}_{t}^{\text{elec}} = E_{t}^{\text{elec}} \cdot h_{t}^{\text{elec}}$$
$${\text{EM}}_{t}^{\text{heat}} = E_{t}^{\text{heat}} \cdot h_{t}^{\text{heat}}$$

In addition, chemical production is accompanied by the process of CO2 reuse, especially in the production of nitrogen fertilizers such as urea and ammonium bicarbonate, which will adsorb CO2 and reduce the overall CO2 emissions for the chemical industry (Eq. 9).

$${\text{EM}}_{t}^{\text{reuse}} = \sum\limits_{j}^{{}} {P_{j,t} \cdot b_{j,t} }$$

Data and basic parameters

The data of energy consumption for different energy types in the chemical industry come from the China Energy Statistics Yearbook (NBSC 2014, 2015, 2016, 2017a). It should be noted that the energy consumption data of the previous years were amended in 2014; therefore, data for year 2000–2013 are all from the revised data by the NBSC (2014).

As the same kind of energy can have different characteristics in different industry sectors, the parameters used to calculate energy emission factors in this study fully consider the characteristics of the chemical industry, such as lower heating value, carbon content in per GJ of energy and carbon oxidation rate, as shown in Table 2.

Table 2 Basic parameters of different energy types

For the calculation of fossil energy-related emissions, the quantities of carbonaceous chemical products are from the Wind database,Footnote 2 National Bureau of Statistics (NBSC) and other sources. However, their carbon contents refer to the “Accounting method and report guide for greenhouse gas emission in China’s chemical production enterprises (Trial)” policy document (NDRC 2013).

Limited by the data availability, process emissions generated by carbonate decomposition just consider calcium carbide in the application of emission accounting in this study. The process emission factor is from Liu et al. (2011). The other emission sources of calcium carbide, including fuel combustion and feedstock-related emission, are calculated in the category of fossil fuel-related emissions.

For the CO2 reuse of the chemical industry, urea and ammonium bicarbonate are selected because they are two of the most important products in nitrogen fertilizer. In the production process, urea can be synthesized by the chemical reaction of CO2 and ammonia at high temperatures and pressures, and ammonium bicarbonate can be produced by CO2, ammonia and water. The absorption of CO2 in those processes will have a positive effect on emission reductions and is considered reuse by the chemical industry in this study. The reuse quantities of per ton nitrogen fertilizer production used in this study are their theoretical values, which can be calculated based on their chemical reaction equations, shown below. According to the relative weights of different reactants and products, it can be calculated that 0.733 tons of CO2 is reused by producing a ton of urea and 0.557 tons of CO2 reused per ton of ammonium bicarbonate production.

$$2{\text{NH}}_{3} + {\text{CO}}_{2} = {\text{CO}}({\text{NH}}_{2} )_{2} + {\text{H}}_{2} {\text{O}}$$
$${\text{NH}}_{3} + {\text{CO}}_{2} + {\text{H}}_{2} {\text{O}} = {\text{NH}}_{4} {\text{HCO}}_{3}$$

The CO2 emission factor of electricity refers to Baseline Emission Factors for Regional Power Grids in China (NDRC 2011), and the average value for six regions is adopted. In addition, the emission factor of heat is taken as 0.11 t CO2/GJ according to one of the China’s national standards (SAC 2015).

Key results

Traditional research overestimates CO2 emissions

The amount of CO2 emissions is influenced by energy consumption and its structure, and the emission factors of different energy types. Because there have been few changes in the energy consumption structure of China’s chemical industry in recent years, CO2 emissions show a three-stage growth trend, similar to the total energy consumption. Here, we take China as the empirical context to validate the EACI model.

For the convenience of comparison, this study also undertakes CO2 emission accounting according to the traditional accounting method, the results of which are shown in Fig. 2. Although the results according to the traditional method are inaccurate, they can reflect the emission structure of different energy categories. The total CO2 emissions for China’s chemical industry increased, with an average annual growth of 8.7%, and the increase was from 329.9 million tons (Mt) in 2000 to 1260.6 Mt in 2016 according to the traditional method. During that period, the total CO2 emissions showed a significant increasing trend between 2000 and 2007. However, due to the financial crisis in 2008, total energy consumption decreased, which resulted in the total emissions showing a temporary decline. During 2009–2015, due to the increase in total energy consumption, China’s CO2 emissions increased again. Then, the total CO2 emissions showed a downward trend in 2016, mainly due to the slight adjustment of the energy consumption structure and decrease in total volume. Compared with 2015, coalFootnote 3 and natural gas consumption in the chemical industry fell by 14.6% and 4.1%, respectively, in 2016. Meanwhile, consumption of petroleum and its productsFootnote 4 is increased by 8.4%, coking products and gasFootnote 5 consumption increased by 8.5%, and electricity and heat consumption increased by 4.6% in 2016. These trends led to a 0.66% drop in total energy consumption and a 3% drop in CO2 emissions in 2016 when compared with 2015.

Fig. 2

CO2 emission by utilizing traditional method and EACI model. Notes: The different color columns in this figure represent the emissions of different energy types calculated by traditional method. “Real emission” refers to the total amount of CO2 emission calculated by EACI in China’s chemical industry

Among all energy types consumed by China’s chemical industry, electricity is the most dominant and is followed by coal. Their massive consumption also caused the most CO2 emissions, which together contribute more than 60% of the total CO2 emissions. In 2016, electricity accounted for 37.2%, coal emissions accounted for 24.0%, and petroleum and its products, coking products and gas, heat and natural gas accounted for 14.7%, 10.3%, 9.8% and 4%, respectively.

By comparing the results calculated by the traditional method in this research with that of Zhou et al. (2010), an obvious difference is found. This is primarily due to the revision of energy consumption data in 2014 to its previous year (NBSC 2014). Thus, the results of this study are more convincing when utilizing the same traditional method or a similar method.

Actually, the CO2 emission trend found by utilizing EACI model is similar to that of the traditional method and energy consumption. However, in most cases, the emissions are more than 19% lower than the emissions calculated by the traditional method. Especially since 2010, the emissions have been reduced by more than 24% each year, which demonstrates that the EACI model can significantly improve accounting accuracy (Fig. 2).

The indirect emissions produced by electricity and heat contribute most

There are two reasons that the results are lower for the EACI model than those of the traditional method. One of the most important reasons is the removal of CO2 corresponding to the carbon fixed in the carbonaceous products (defined as “carbon in product” in this study), and the other is the reuse of CO2 in the production of nitrogen fertilizer.

Taking 2016 as an example, the CO2 emissions accounted for by EACI are approximately 885.9 Mt, which accounts for approximately 9.7% of the total amount for China.Footnote 6 This result is approximately 30% lower than that by the traditional method, as shown in Fig. 3. If all energy consumption is by combustion, the total amount of “fuel combustion” is 668.3 Mt, of which coal accounts for 48%, oil accounts for 25%, and coke products and gas accounts for 20%, while natural gas only accounts for approximately 8%. However, it is not the real fossil energy-related emissions as the feedstock-related emissions are ignored. Only when “carbon in products” is removed from “fuel combustion” can the results be closer to the actual value. In fact, approximately 93.2 Mt of carbon are not burned in fossil energy consumption by the chemical industry, but is fixed in carbonaceous products. This amount of carbon corresponds to approximately 341.7 Mt CO2, which accounts for 51% of “fuel combustion.” This means that the fossil energy-related CO2 emission is nearly twice overestimated when using the traditional method. The corresponding CO2 of carbon in carbonaceous products is mainly contributed by olefins (ethylene, propylene and butadiene) and BTX (pure benzene, toluene and xylene), which accounted for 42.1% and 26.4%, respectively. However, methanol, calcium carbide and carbon black account for 17.4%, 8.6% and 5.4%, respectively. Considering both of “fuel combustion” and “carbon in products,” the CO2 emissions related to fossil energy are 326.6 Mt for 2016, which accounts for 36.9% of the total emissions.

Fig. 3

CO2 emissions by sectors in 2016. Notes: The “Fuel combustion,” “Electricity & Heat” and “Process emission” columns represent the positive values of emissions, whereas the “Reuse” and “Carbon in product” column represent negative values. “Fuel combustion” refers to the CO2 emissions if all energy consumption is combusted in the chemical industry, and “Carbon in product” represents the corresponding CO2 emission in carbonaceous products, which should be eliminated from “fuel combustion”

The indirect CO2 emissions generated by electricity and heat uses contribute approximately 66.9% to the total amount for the chemical industry. Among those uses, electricity use generated approximately 469.2 Mt CO2 and has become the largest emission sources of the chemical industry, which accounts for approximately 53.0% of the total. Heat use produced 123.1 Mt CO2, accounting for approximately 13.9% of the total.

The process emissions generated by carbonate decomposition are 20.2 Mt CO2, accounting for approximately 2.3%. In addition, reused CO2 causes the total emissions to be reduced by approximately 6.1%. In fact, products that can absorb CO2 failed to be fully counted and taken into consideration, which causes the absolute value of CO2 reuse to be underestimated.

The emissions produced by key chemical products account for half

In view of product level, most CO2 emissions in the chemical industry are generated by several basic chemicals, in particular ammonia, ethylene and calcium carbide. The emissions of these three products and their structures are shown in Fig. 4.

Fig. 4

CO2 emissions by products and their structures in 2016

Ammonia production is among the largest emission sources in the chemical industry, primarily because of the large amount of production and its production structure. Coal to ammonia accounted for 96% in China, and the remaining 4% is produced by natural gas in China. However, it will generate approximately 4.582 t CO2 for producing per ton ammonia with coal but 2.104 t CO2 with natural gas (Zhou et al. 2010). Thus, the 57.08 Mt of ammonia production (NBSC 2017b) produced approximately 256 Mt CO2, which accounted for 28.9% of that produced by the chemical industry in 2016. In addition, almost all of the emissions of ammonia were generated by coal to ammonia, while natural gas to ammonia only accounted for 2%. According to this study and Chen et al. (2018), the ethylene productionFootnote 7 accounted for approximately 6.8% of chemical industry. Among them, steam cracking, coal to olefins (CTO) and methanol to olefins (MTO) contributed 48.8%, 47.8% and 3.4%, respectively. The production of calcium carbide generated approximately 134.3 Mt CO2 in 2016,Footnote 8 accounting for approximately 15.2%. In terms of production process, the manufacturing process produced the most emissions, accounting for approximately 65.2%, and emissions from feedstock preparation and furnace gas treatment accounted for 24.4% and 10.4%, respectively, in 2016.


This section will discuss the advantages and disadvantages of the EACI model and try to answer how to conduct CO2 emission accounting based on the wealth of chemical plant-level data.

Limited by the poor data availability, numerous products and complex production processes, CO2 emissions accounting for the chemical industry are hardly accurate. Although the real emissions could be approximated by utilizing our improved EACI model method, a gap with the actual result remains, primarily due to the fact that not all chemical products could be fully considered and without duplication, especially carbonaceous products and products with process emissions and reuse of CO2. This always leads to overestimations of fossil energy-related CO2 emissions and underestimations of process emissions. In addition, the absolute amount of reuse is also underestimated, causing a biased CO2 emission estimation. However, the EACI model is an improved method that considers more comprehensive emission sources, and it could perform better when the data are detailed and abundant.

The disadvantages discussed are based on the statistics of the entirety of chemical industry. However, it is easier to collect detailed data and account for the CO2 emissions of each of the chemical plants. Thus, a bottom-up mode can be adopted to calculate CO2 emission at the chemical plant level by two primary means. One is to calculate each of the six emission sources and then summarize the six parts from the enterprise level, and the other is from the perspective of carbon source flow, by which the CO2 emissions can be calculated based on the carbon loss. These two methods will be discussed in Sects. 4.1 and 4.2.

Method based on emission sources

In this method, CO2 emissions should be calculated by emission sources and by chemical plants. In addition, fuel combustion emissions (\({\text{EM}}_{e,t}^{\text{combus}}\)), fossil feedstock-related emissions (\({\text{EM}}_{e,t}^{\text{feedstock}}\)), carbonate-related process emissions (\({\text{EM}}_{e,t}^{\text{process}}\)), indirect emissions of electricity and heat (\({\text{EM}}_{e,t}^{\text{elec}}\) and \({\text{EM}}_{e,t}^{\text{heat}}\)) and the reuse of CO2 (\({\text{EM}}_{e,t}^{\text{reuse}}\)) need to be separately calculated (Eq. 10). However, fossil energy used as fuel and feedstock should be well distinguished by the chemical enterprises; emissions produced by them can be calculated with Eqs. 11 and 12. The emission factors of fossil fuel (\({\text{EF}}_{{k_{n} ,t}}\)) are calculated by Eq. 13, which is similar to Eq. 4. Other parameters can be obtained by using Eqs. 69.

Although this method is similar to the EACI model in terms of emission source, their accounting modes and application scopes differ. The method based on emission source is a bottom-up model that accounts for the CO2 emissions at chemical plant level by each source. However, the EACI model is in a top-down mode to perform the accounting from the chemical industry level, in which emissions from fuel combustion and fossil feedstock use need to be combined as fossil fuel-related emissions by utilizing the method shown in Eq. 3. Thus, in cases where chemical plant-level data are available, the accounting will be more accurate by utilizing the method based on emission sources. However, for cases where only industry-level data are available, the EACI model will perform better.

$${\text{AAT}}_{t}^{{{\text{CO}}_{2} }} = \sum\limits_{e}^{{}} {{\text{EM}}_{e,t}^{\text{combus}} + {\text{EM}}_{e,t}^{\text{feedstock}} + {\text{EM}}_{e,t}^{\text{process}} + {\text{EM}}_{e,t}^{\text{elec}} + {\text{EM}}_{e,t}^{\text{heat}} - {\text{EM}}_{e,t}^{\text{reuse}} }$$
$${\text{EM}}_{e,t}^{\text{combus}} = \sum\limits_{{k_{n} }}^{{}} {E_{{e,k_{n} ,t}} \cdot {\text{EF}}_{{e,k_{n} ,t}} }$$
$${\text{EM}}_{e,t}^{\text{feedstock}} = \sum\limits_{{k_{m} }}^{{}} {E_{{e,k_{m} ,t}} \cdot {\text{EF}}_{{k_{m} ,t}} }$$
$${\text{EF}}_{{k_{n} ,t}} = h_{{k_{n} ,t}} \cdot C_{{k_{n} ,t}} \cdot f_{{k_{n} ,t}} \cdot \varphi$$

The chemical plants need to measure the feedstock-related emissions for per ton product production (\({\text{EF}}_{{k_{n} ,t}}\)) professionally. However, considering that there are little changes in the same product production for the same chemical plant, this kind of measurement does not need to be performed every time only if the feedstock structure changes a lot. By this method, the feedstock-related emissions can be easily accounted.

Method based on carbon source flow

In fact, emissions related to fuel combustion, feedstock use and carbonate use can also be accounted in view of carbon source flow (Eq. 14). In this method, both the carbon input into the chemical plants and output from chemical plants should be counted. The calculation boundary is chemical plants. The statistical range of input carbonaceous matter includes fossil fuel, fossil feedstock, carbonates and others, whereas that of output matter includes the chemical products and by-products, carbonaceous dust, residue and others. The amount of carbon can be calculated by the quantity of carbonaceous matter (\(G_{e,g,t}\) and \(Y_{e,y,t}\)) and their carbon contents (\(c_{g,t}\) and \(c_{y,t}\)), as shown in Eqs. 15 and 16. When the amounts of input carbon (\(C_{e,t}^{\text{in}}\)) and output carbon (\(C_{e,t}^{\text{out}}\)) are obtained, the corresponding CO2 emissions can be calculated with parameter \(\varphi\). The accounting methods of electricity use, heat use and CO2 reuse are consistent with that in the EACI model.

$${\text{ABT}}_{t}^{{{\text{CO}}_{2} }} = \sum\limits_{e}^{{}} {\left( {C_{e,t}^{\text{in}} - C_{e,t}^{\text{out}} } \right) \cdot \varphi + {\text{EM}}_{e,t}^{\text{elec}} + {\text{EM}}_{e,t}^{\text{heat}} - {\text{EM}}_{e,t}^{\text{reuse}} }$$
$$C_{e,t}^{\text{in}} = \sum\limits_{g}^{{}} {G_{e,g,t} \cdot c_{g,t} }$$
$$C_{e,t}^{\text{out}} = \sum\limits_{y}^{{}} {Y_{e,y,t} \cdot c_{y,t} }$$

This method can theoretically calculate emissions accurately, but it is tedious to determine the amount of carbon inflows and outflows regularly in its production. A similar method has been regulated in one of the China’s national standards titled “Requirements of greenhouse gas emission accounting and reporting—Part 10: Chemical production enterprise” (SAC 2015). However, different from this study, the carbon source flow is only used in the accounting of feedstock-related emissions in that national standard.

Conclusions and policy implications


This study proposes the Emission Accounting Model in the Chemical Industry (EACI), which can be used to approximate real emissions given poor data availability for the entirety of the chemical industry. In addition, this study also proposes two bottom-up methods that are based on emission sources and carbon source flow, and they can take effect if the chemical plant-level data are available. The EACI model is applied to China’s chemical industry in this study, and several conclusions are drawn.

  1. 1.

    It is found that the EACI model can make the emissions of the chemical industry decrease by 19–30% during 2000 and 2016 when compared with the results calculated by the traditional method. In particular, the EACI model is effective for fossil energy-related emissions that include fuel and feedstock. By comparison, it is found that the fossil energy-related CO2 emissions can be reduced by approximately 51% by utilizing the EACI model, which means that they are doubly overestimated by the traditional method. In addition, this method also considers the reuse of CO2. In particular, the EACI model will be more effective when the data are elaborate.

  2. 2.

    According to the EACI model, this study undertakes an accounting of the chemical industry in China. The amount of CO2 emissions increased overall since 2000. In addition, there was a slight decrease in 2016 when compared with 2015, primarily because of the small change in energy consumption and its structure. The amount is approximately 886 Mt for 2016, accounting for approximately 9.7% of China’s total CO2 emissions. The indirect emissions generated by electricity and heat use are the largest emission source, accounting for 66.9% of the entirety of the emissions of the chemical industry, and are followed by fossil energy-related emissions, which account for approximately 36.9%. Meanwhile, the process emissions account for 2.3% and the reuse of CO2 contributes 6.0% to the reduction in chemical emissions.

  3. 3.

    Among thousands of chemical products, the emissions for ammonia, ethylene and calcium carbide production accounted for 28.9%, 15.2% and 6.8% of total, respectively, together contributing approximately half to the total. Among them, the producing way of coal to ammonia contributed almost all of the emissions in the production of ammonia. Manufacturing process contributed the most emissions in calcium carbide production, accounting for approximately 65.2%.

  4. 4.

    To decrease the accounting error with industry-level statistics, this study proposed two other methods based on chemical plants. They employ a bottom-up mode and can calculate relatively accurate actual emissions. However, they may require more effort in the collection of data.

Policy implications

On the basis of the conclusions drawn above, several policy implications are provided as follows.

  1. 1.

    Data availability should be improved in the chemical industry to promote chemical-related researches. A sound data monitoring and reporting system is required to support accurate emission accounting.

  2. 2.

    The EACI model is suggested when energy data are based on the entirety of the chemical industry. However, if chemical plant-level data are available, the bottom-up methods based on emission sources and carbon source flow are recommended.

  3. 3.

    Considering the large amount of indirect emissions, cleaner electricity and heat production will make important contributions to CO2 emission reduction. In addition, more attention should be paid to the emission reduction of key products like ammonia, ethylene and calcium carbide.


  1. 1.

    The chemical industry in this study refers to the division 26 called Manufacturing of Chemical Materials and Chemicals which is defined by the national standard (SAC 2017). It covers thousands of chemical products and can be divided into eight parts: (1) Manufacturing of Basic Chemical Raw Materials, (2) Manufacturing of Fertilizer, (3) Manufacturing of Pesticide, (4) Manufacturing of Coatings, Inks, Pigments and Similar Products, (5) Manufacturing of Synthetic Materials, (6) Manufacturing of Special Chemical Products, (7) Manufacturing of Explosives, Pyrotechnics and Fireworks Products and (8) Manufacturing of Daily Chemical Products.

  2. 2.

  3. 3.

    In this study, coal includes raw coal, cleaned coal and other washed coal.

  4. 4.

    In this study, petroleum and its products refer to crude oil, gasoline, kerosene, diesel oil, fuel oil, naphtha, petroleum coke, LPG, refinery gas and other petroleum products.

  5. 5.

    Coking products represent coke and other coking products. Gas includes coke oven gas, blast furnace gas, converter gas and other gas.

  6. 6.

    This ratio is calculated on the basis of BP (2017), in which China produced about 9163 Mt CO2.

  7. 7.

    Here, the ethylene production also includes the production of its by-products, such as propylene and butadiene in this process. However, it should be noted that only part of propylene and butadiene is produced in accompaniment with ethylene, not the whole.

  8. 8.

    Here, the emissions of calcium carbide are total emissions. They were calculated by the authors. The production of calcium carbide and its emission factor come from NBSC (2017c) and (Liu et al. 2011) respectively.


  1. An R, Yu B, Li R, Wei Y-M (2018) Potential of energy savings and CO2 emission reduction in China’s iron and steel industry. Appl Energy 226:862–880

    Article  Google Scholar 

  2. BP (British Petroleum) (2017) BP statistical review of world energy June 2017. BP, London

  3. Broeren MLM, Saygin D, Patel MK (2014) Forecasting global developments in the basic chemical industry for environmental policy analysis. Energy Policy 64:273–287.

    Article  Google Scholar 

  4. Cefic (European Chemical Industry Council) (2017) Facts & figures of the European chemical industry. Brussels, Belgium

    Google Scholar 

  5. Chen H, Tang B-J, Liao H, Wei Y-M (2016) A multi-period power generation planning model incorporating the non-carbon external costs: a case study of China. Appl Energy 183:1333–1345.

    Article  Google Scholar 

  6. Chen H, Kang J-N, Liao H, Tang B-J, Wei Y-M (2017) Costs and potentials of energy conservation in China’s coal-fired power industry: a bottom-up approach considering price uncertainties. Energy Policy 104:23–32.

    Article  Google Scholar 

  7. Chen J-M, Yu B, Wei Y-M (2018) Energy technology roadmap for ethylene industry in China. Appl Energy 224:160–174.

    Article  Google Scholar 

  8. Griffin PW, Hammond GP, Norman JB (2017) Industrial energy use and carbon emissions reduction in the chemicals sector: a UK perspective. Appl Energy.

    Article  Google Scholar 

  9. IEA (International Energy Agency) (2017) Key world energy statistics. France

  10. IPCC (Intergovernmental Panel on Climate Change) (2006) IPCC guidelines for national greenhouse gas inventories (2007-04-01). Accessed 10 Mar 2018

  11. Li R, Tang B-J (2016) Initial carbon quota allocation methods of power sectors: a China case study. Nat Hazards 84:1075–1089

    Article  Google Scholar 

  12. Liu X, Bing Z, Zhou W, Hu S, Chen D, Griffy-Brown C (2011) CO2 emissions in calcium carbide industry: an analysis of China’s mitigation potential. Int J Greenh Gas Control 5:1240–1249

    Article  Google Scholar 

  13. Martin R, De Preux LB, Wagner UJ (2014) The impact of a carbon tax on manufacturing: evidence from microdata. J Public Econ 117:1–14

    Article  Google Scholar 

  14. NBSC (National Bureau of Statistics of China) (2014) China energy statistical yearbook. China Statistics Press, Beijing

    Google Scholar 

  15. NBSC (National Bureau of Statistics of China) (2015) China energy statistical yearbook. China Statistics Press, Beijing

    Google Scholar 

  16. NBSC (National Bureau of Statistics of China) (2016) China energy statistical yearbook. China Statistics Press, Beijing

    Google Scholar 

  17. NBSC (National Bureau of Statistics of China) (2017a) China energy statistical yearbook. China Statistics Press, Beijing

    Google Scholar 

  18. NBSC (National Bureau of Statistics of China) (2017b) The production of ammonia. Accessed 10 Apr 2018

  19. NBSC (National Bureau of Statistics of China) (2017c) The production of calcium carbide. Accessed 11 Apr 2018

  20. NDRC (National Development and Reform Commission) (2011) Guidelines for the compilation of provincial greenhouse gas emission inventories (trial). Accessed 25 Mar 2018

  21. NDRC (National Development and Reform Commission) (2013) Accounting method and report guide for greenhouse gas emission in China’s chemical production enterprises (trial) (2013-11-04). Accessed 15 Apr 2018

  22. Raupach MR et al (2014) Sharing a quota on cumulative carbon emissions. Nat Clim Change 4:873

    Article  Google Scholar 

  23. SAC (Standardization Administration of China) (2015) Requirements of greenhouse gas emissions accounting and reporting—part 10: chemical production enterprise vol 13.020.10. Standards Press of China, Beijing

    Google Scholar 

  24. SAC (Standardization Administration of China) (2017) Industrial classification for national economic activities vol 35.040. Standards Press of China, Beijing

    Google Scholar 

  25. Saygin D, Patel MK, Tam C, Gielen DJ (2009) Chemical and petrochemical sector: potential of best practice technology and other measures for improving energy efficiency. IEA information paper

  26. Selvakkumaran S, Limmeechokchai B (2015) Low carbon society scenario analysis of transport sector of an emerging economy—the AIM/Enduse modelling approach. Energy Policy 81:199–214.

    Article  Google Scholar 

  27. Tang B, Li R, Yu B, An R, Wei Y-M (2018) How to peak carbon emissions in China’s power sector: a regional perspective. Energy Policy 120:365–381

    Article  Google Scholar 

  28. Wei Y-M et al (2018) An integrated assessment of INDCs under shared socioeconomic pathways: an implementation of C(3)IAM. Nat Hazards 92:585–618.

    Article  Google Scholar 

  29. Worrell E, Price L, Martin N (2001) Energy efficiency and carbon dioxide emissions reduction opportunities in the US iron and steel sector. Energy 26:513–536.

    Article  Google Scholar 

  30. Zhang C-Y, Han R, Yu B, Wei Y-M (2018) Accounting process-related CO2 emissions from global cement production under shared socioeconomic pathways. J Clean Prod 184:451–465

    Article  Google Scholar 

  31. Zhou P, Wang M (2016) Carbon dioxide emissions allocation: a review. Ecol Econ 125:47–59

    Article  Google Scholar 

  32. Zhou W, Zhu B, Li Q, Ma T, Hu S, Griffy-Brown C (2010) CO2 emissions and mitigation potential in China’s ammonia industry. Energy Policy 38:3701–3709.

    Article  Google Scholar 

  33. Zhu B, Zhou W, Hu S, Li Q, Griffy-Brown C, Jin Y (2010) CO2 emissions and r eduction potential in China’s chemical industry. Energy 35:4663–4670.

    Article  Google Scholar 

Download references


The authors acknowledge the financial support received from the National Key R&D Program of China (2016YFA0602603), the National Natural Science Foundation of China (Nos. 71822401, 71603020, 71521002 and 71642004) and the Joint Development Program of Beijing Municipal Commission of Education. We are also thankful for the support and help provided by CEEP-BIT colleagues.

Author information



Corresponding author

Correspondence to Biying Yu.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, J., Yu, B. & Wei, Y. CO2 emissions accounting for the chemical industry: an empirical analysis for China. Nat Hazards 99, 1327–1343 (2019).

Download citation


  • CO2 emission
  • EACI model
  • Bottom-up accounting method
  • Chemical industry