SEARCH

SEARCH BY CITATION

Keywords:

  • industrial ecology;
  • inorganic chemicals;
  • chemical manufacturing;
  • life cycle assessment (LCA);
  • life cycle impact assessment (LCIA);
  • petrochemicals

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

In many cases, policy makers and laymen perceive harmful emissions from chemical plants as the most important source of environmental impacts in chemical production. As a result, regulations and environmental efforts have tended to focus on this area. Concerns about energy use and greenhouse gas emissions, however, are increasing in all industrial sectors. Using a life cycle assessment (LCA) approach, we analyzed the full environmental impacts of producing 99 chemical products in Western Europe from cradle to factory gate. We applied several life cycle impact assessment (LCIA) methods to cover various impact areas. Our analysis shows that for both organic and inorganic chemical production in industrial countries, energy-related impacts often represent more than half and sometimes up to 80% of the total impacts, according to a range of LCIA methods. Resource use for material feedstock is also important, whereas direct emissions from chemical plants may make up only 5% to 10% of the total environmental impacts. Additionally, the energy-related impacts of organic chemical production increase with the complexity of the chemicals. The results of this study offer important information for policy makers and sustainability experts in the chemical industry striving to reduce environmental impacts. We identify more sustainable energy production and use as an important option for improvements in the environmental profile of chemical production in industrial countries, especially for the production of advanced organic and fine chemicals.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

Due to growing societal concerns and increasingly strict governmental regulations, the chemical industry has in recent decades striven to reduce their emissions to the environment. Enhanced end-of-pipe methods, environmental considerations in design stages, and improved efficiency have diminished the environmental impacts of chemical plants significantly (Responsible Care 2009). The importance of sustainable production is generally recognized (Mihelcic et al. 2003), and several companies, such as BASF (Saling et al. 2002; Shonnard et al. 2003), GlaxoSmithKline (Constable et al. 2001; Curzons et al. 2007), and Ciba (Bretz and Frankhauser 1997), have published information on their efforts. Technologies such as waste gas incineration, solvent recycling, and advanced wastewater treatment, however, often lead to an increased use of energy. Although direct emission to air, water, and soil have long been seen as the most damaging aspects of a chemical plant, energy use and indirect emissions are now also recognized as important factors in the environmental performance of chemical production. The Responsible Care initiative of the chemical industry began tracking energy efficiency and greenhouse gas emissions in 2005, although not all countries take into account indirect emissions (Responsible Care 2009). Already, with regard to some technologies, such as waste treatment and solvent distillation, researchers argue that efforts to minimize emissions may, in some cases, exceed the environmental benefits (Romero-Hernandez et al. 1998; Jankowitsch et al. 2001; Romero-Hernandez 2004). Life cycle assessment (LCA) is an ideal tool to assess this matter, because it allows accounting for direct emissions and environmental impacts as well as the indirect emissions caused elsewhere in the supply chain (“cradle-to-gate” assessment). LCA is thus increasingly used to assess the environmental performance of chemical production (Bretz and Frankhauser 1997; Constable et al. 2001; Saling et al. 2002; Shonnard et al. 2003; Hellweg et al. 2004; Geisler et al., 2005a, 2005b; Kralisch et al. 2005; Capello et al. 2007b; Curzons et al. 2007). It has also been shown as a useful tool for comparing waste treatment options (Capello et al. 2007a, 2008).

Existing life cycle inventory (LCI) data (emissions and resource use) are often aggregated and not separated according to individual unit processes. A detailed environmental assessment of the production of a large number of chemical products, including an attribution of impacts to different phases of the production process (e.g., energy production, provision of feedstocks, or wastewater treatment) can therefore be difficult to perform. Recently, a previous version of the LCI database ecoinvent served as the basis of an analysis of material production (Huijbregts et al. 2006), including chemicals. The release of Version 2 of the ecoinvent database in early 2008 (ecoinvent centre 2008) allows for a new perspective on chemical production, as this version contains many more detailed data sets on chemicals. An additional source of high-quality process data was provided by partners from the chemical industry. In this work, we therefore perform a refined analysis of chemical production, using extensive data from public and industry sources. The result is an in-depth analysis of the main drivers of the environmental impact of today's chemical industry, with a cradle-to-factory-gate perspective, which offers valuable information to policy makers and experts in the industry striving to reduce environmental impacts in chemical production. In addition, the results may be helpful for LCA practitioners unable to assess production of a specific chemical product on a process level, a common scenario given that LCIs exist for only a very small fraction of the more than 100,000 chemicals in production today (Wernet et al. 2008, 2009).

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

To assess the environmental impacts of chemical production in general and the relative importance of energy use in particular, we used an LCA approach. The boundaries for this study were chosen to be the full production life cycle of the chemicals, including all resources used and emissions caused by the production process, provision of auxiliaries and energy, and recycling and waste treatment procedures required during these steps. This represents a “cradle-to-factory-gate” approach. We did not consider the use phase and eventual disposal of the chemical because many chemicals, especially basic chemicals, such as those from the ecoinvent data sets, may have different uses and therefore various end-of-life scenarios.

The main data source was a set of LCI data on 78 chemicals from the ecoinvent database. More than half the data (50 data sets) covered organic solvents, which are used in many industries (Capello et al. 2009). This set included both very basic chemicals, such as hexane and methanol, and costlier solvents, such as isopropyl acetate and tetrahydrofuran. To assess the production of other types of chemicals, we also selected more data sets from other chemical classes. Data on the chemicals with the highest worldwide production (McKay et al. 2007) were included in the analysis if production systems were described in ecoinvent. These included ten common monomers used in plastics manufacture; eight other basic organic chemicals (e.g., aniline and ethylene oxide); and ten basic inorganic compounds, such as ammonia, chlorine, and sulfuric acid. As the solvents, monomers, and other basic organic chemicals were different in use but not inherently different in structure, they were grouped into a single class labeled “organic chemicals.” The majority of these compounds are simple organic chemicals; few require more than three or four reaction steps to synthesize. All 78 chemicals are produced in large quantities, in ranges from many thousands to millions of metric tons per year globally. Some chemicals from the ecoinvent database were considered for this work but rejected, as data quality was judged to be too low. For example, in some cases the data relied on estimations for important parameters, such as reaction yield or energy consumption over several process steps.

In addition, we analyzed data on 21 organic chemicals generated in cooperation with a chemical company. The industrial data sets were advanced chemicals of higher complexity than the chemicals covered by the ecoinvent data set. The data were based on production documentation, direct measurements, and a detailed bottom-up modeling of energy use in cooperation with the producer (Szijjarto et al. 2008) and were therefore judged to be of high quality. These data sets were available on a process level and represented chemicals of varying complexity, ranging from simple organic chemicals to advanced and fine chemicals, such as dyes and pigments. Even though the production sites were located in Switzerland and Germany, production was modeled on continental European standards (e.g., use of a Union for the Co-ordination of Transmission of Electricity [UCTE; a continental European group of electricity providers] electricity mix and a generic steam production process) to ensure comparability with the ecoinvent data, most of which were created for production in Europe, including average European supply chains. Production of auxiliaries and background processes were modeled with ecoinvent data if no other data were available. Each chemical was assessed individually, and results were averaged for the different groups of chemicals: one group each for organic and inorganic chemicals from ecoinvent data set, and one group for the organic chemicals in the industrial data sets.

Using a process-based approach, we analyzed the inventories to assess the contributions of energy production and use to the total impacts. To achieve this, we separated all processes in the life cycle into different source categories in the Brightway LCA tool (Mutel and Kestenholz 2008). The two energy-related categories selected for this study were transport (including all forms of transporting goods occurring in the production chain) and heat and electricity use (including all forms of generating energy in the form of heat, steam, or electricity). All emissions and resources due to transport, electricity production, and heat production, including all necessary upstream processes, were classified as “energy-related.” All other emissions and resources were classified as “remaining causes.” Such remaining causes included, for example, emissions during chemical reactions or from waste treatment and resource uses to provide material feedstocks. Special care was taken to separate uses of oil and natural gas for energy production from feedstock requirements, as these two substances are both major energy sources and the most important feedstock resources in chemical production. For example, resource uses and emissions from the use of natural gas (and all inventory flows due to the production and transport of this gas) were classified as “energy-related” if the gas was burned to generate electricity or heat but were classified under “remaining causes” if the gas was a material feedstock in a chemical reaction. Note that all flows from the transport of the gas were still classified as “energy-related,” as that class also included transport.

We calculated category impacts by comparing the results of a regular LCI with the results of a modified calculation. During the calculation of the modified LCI results, the algorithm treated all processes from the category under analysis as having zero inputs and outputs (besides the reference flow). This “removed” the part of the inventory stemming from the processes under analysis, including all upstream processes. A comparison of the results allowed an identification of the impacts for the given classification. Some aggregated or inconsistent processes exist in the ecoinvent database; in these cases, the standard approach failed. We manually adapted these processes using the ecoinvent reports and source data and extended the algorithm to use the appropriate data in the calculations (see the Supporting Information on the Web). Practically speaking, the modified calculation selectively pruned all processes related directly or indirectly to the production of energy in the form of heat or electricity and transport from the overall production process tree. By assessing the difference, we could then determine their contribution.

In a following step, we applied several life cycle impact assessment (LCIA) methods to the LCI data to evaluate the environmental impacts. LCIA methods can either address specific impacts in separate impact categories or aggregate “impact scores” across impact categories. In the latter case, subjective steps may be necessary, such as weighting by social preferences. Several LCIA methods were used in the present study: Cumulative energy demand (CED; VDI 1997) is a single-score method assessing the amount of primary energy equivalents required for the finished product, including direct, indirect, and feedstock energies. CED is therefore based purely on resource extraction and was chosen as a representative of resource-based methods, although others exist (Bosch et al. 2007; DeWulf et al. 2007). Global warming potential (GWP; IPCC 2007) is an assessment method focusing on the effects of emissions of greenhouse gases on the climate and temperature levels of the planet. Ecoindicator 99 (EI99; Goedkoop and Spriensma 2000) is a damage-oriented assessment method providing results for the damage categories of human health, ecosystem quality, and resources. These can be aggregated into single-score results on the basis of various weighting schemes determined by a panel of experts. In this work, we used the hierarchist weighting scheme (H/A). The Ecological Scarcity (ES2006) method (Frischknecht et al. 2006) also provides a single-score result. Impacts are determined through the relation of current and politically set maximum flows. Because it is a Swiss method, the targets are usually based on Swiss regulations or administrative goals. Finally, the Tool for the Reduction and Assessment of Chemical and Other Environmental Impacts (TRACI) method (Bare et al. 2003; Bare et al. 2006) was applied. TRACI provides results for several impact categories, without weighting or aggregation. This combination of methods was chosen to provide an extensive assessment of chemical production from different viewpoints, although other aggregating methods may offer additional perspectives (e.g., Jolliet et al. 2003; Rosenbaum et al. 2008; Goedkoop et al. 2009).

Results and Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

All LCIA methods consistently confirmed that transport is, in general, of only minor environmental concern in the production of chemicals. Average transport-related impacts were between 2% and 4% of total impacts for the CED, GWP, EI99, and ES2006 aggregated scores for both organic chemical data sets and between 4% and 5% for the inorganic chemicals. This was true even for the data sets with a global scope. Impacts of transport are therefore not separately shown in the graphs below but are integrated into a general, energy-related source group together with the impacts of heat and electricity use. These energy-related impacts thus consist of all impacts due to transport and the use of energy in the form of heat or electricity, including all upstream processes, indirect emissions, and resource uses. They are contrasted here with the remaining impacts, caused mostly by the provision of feedstock materials, emissions at the chemical plant itself, and waste treatment.

Figure 1 shows an overview of the environmental profile of chemical production, whereas

image

Figure 1. Environmental impacts and energy-related fractions for the 68 organic chemicals from ecoinvent (OC-Eco), 21 industrial data sets (OC-Ind), and 10 inorganic chemicals from ecoinvent (IC-Eco). Averaged results are shown for the cumulative energy demand (CED) in megajoule-equivalents per kilogram (MJ-eq/kg), global warming potential (GWP) in kilograms of carbon dioxide equivalent per kilogram (kg CO2-eq/kg), the aggregated Ecoindicator 99 (EI99) in points per kilogram (points/kg), and ecological scarcity (ES2006) scores in Umweltbelastungspunkte (environmental load points) per kilogram (UBP/kg). The error bars indicate the standard deviations of the energy-related fractions.

Download figure to PowerPoint

Table 1 shows the full results of all applied methods, including subcategories. The industrial data sets on advanced chemicals always have higher impacts than both the basic organic and inorganic chemicals from ecoinvent, which was expected due to their more complex production requirements. Deviations from the mean values are also higher due to the more heterogeneous makeup of this data set. The quantification and analysis of the environmental impacts of chemical production and the attribution of the impacts to their sources in this work allows a detailed understanding of the current status of the chemical industry in industrial countries and highlights areas where improvements would be most effective. The results consistently show that a large fraction of the environmental impacts of chemical production in industrial countries—often the majority of all impacts, depending on the LCIA method applied—are caused by energy production and use. Energy-related impacts constitute between 40% and 80% of the overall impacts for most LCIA methods and are particularly high with regard to human health impacts and global warming. Among the three classes of chemicals, the industry-data chemicals commonly have the highest fraction of energy-related impacts, except for the inorganic chemicals with regard to the CED method. This can be explained by the fact that the latter chemicals do not require fossil fuels as feedstocks.

Table 1.  Full results for all methods applied
MethodOrganic chemicals (ecoinvent)Organic chemicals (industry data)Inorganic chemicals
  1. Note: For each life cycle impact assessment (LCIA) method, the energy-related fractions of the total impacts are given in boldface, followed by the standard deviation of these fractions. Below these are the mean value of the total impacts and the standard deviation in regular print. MJ-eq/kg = megajoule equivalents per kilogram; kg CO2-eq/kg = kilograms of carbon dioxide equivalents per kilogram; (UBP)/kg = Umweltbelastungspunkte per kilogram, where Umweltbelastungspunkte translates roughly as “ecological scarcity points”; H/A = hierarchist weighting scheme; points/kg = points per kilogram; kg 2,4-D-eq/kg = kilograms of 2,4-dichlorophenoxyacetic acid equivalents per kilogram; kg benzene-eq/kg = kilograms of benzene equivalents per kilogram; kg toluene-eq/kg = kilograms of toluene equivalents per kilogram; kg PM2.5-eq/kg = kilograms of particulate matter no larger than 2.5 microns in diameter per kilogram; H+ moles-eq/kg = moles of hydrogen ion equivalents per kilogram; kg N-eq/kg = kilograms of nitrogen equivalents per kilogram; kg CFC 11-eq/kg = kilograms of chlorofluorocarbon 11 equivalents per kilogram, where “chlorofluorocarbon 11” refers to trichlorofluoromethane (CCl3F); and kg NOx-eq/kg = kilograms of nitrogen oxides equivalents per kilogram.

Cumulative energy demand [MJ-eq/kg]45%± 14%69%± 19%83%± 23%
76.9 ± 22.4200.4 ± 194.122.1 ± 19.6
Global warming potential [kg CO2-eq/kg]81%± 1%82%± 15%78%± 30%
2.5 ± 1.2 9.5 ± 10.21.5 ± 1.2
Ecological Scarcity 200657%± 17%65%± 26%54%± 28%
Total score (Umweltbelastungspunkte [UBP]/kg)2,570 ± 2,77010,460 ± 14,5201,520 ± 1,270
Ecological Scarcity 200679%± 18%87%± 14%79%± 15%
Deposited waste (UBP/kg) 392 ± 1,820768 ± 992214 ± 360
Ecological Scarcity 200666%± 14%72%± 17%58%± 29%
Emission into air (UBP/kg)1,250 ± 719 4,780 ± 4,630795 ± 519
Ecological Scarcity 200641%± 18%37%± 35%9%± 14%
Emission into groundwater (UBP/kg)0.19 ± 0.562.83 ± 5.871.13 ± 0.98
Ecological Scarcity 200640%± 24%58%± 34%44%± 23%
Emission into surface water (UBP/kg) 666 ± 1,870 4,160 ± 11,520 971 ± 2,490
Ecological Scarcity 200688%± 10%85%± 12%67%± 19%
Emission into topsoil (UBP/kg)2.44 ± 2.2715.3 ± 25.93.68 ± 3.63
Ecological Scarcity 200645%± 14%69%± 19%84%± 24%
Energy resources (UBP/kg)257 ± 75 666 ± 63872 ± 63
Ecological Scarcity 200627%± 15%39%± 26%36%± 17%
Natural resources (UBP/kg)22.0 ± 18.077.7 ± 91.220.2 ± 27.9
Ecoindicator 99 H/A43%± 12%63%± 20%59%± 28%
Total [points/kg]0.265 ± 0.0780.737 ± 0.6360.083 ± 0.061
Ecoindicator 99 H/A74%± 13%71%± 21%57%± 32%
Human health (points/kg)0.044 ± 0.0230.210 ± 0.1980.040 ± 0.026
Ecoindicator 99 H/A61%± 20%67%± 20%38%± 20%
Ecosystem quality (points/kg)0.008 ± 0.0040.037 ± 0.0530.009 ± 0.009
Ecoindicator 99 H/A38%± 13%63%± 23%76%± 23%
Resources (points/kg)0.213 ± 0.0620.489 ± 0.4270.035 ± 0.037
Tool for the Reduction and Assessment of Chemical and Other Environmental Impacts (TRACI)55%± 20%58%± 18%46%± 18%
0.52 ± 0.472.32 ± 3.510.37 ± 0.22
Ecotoxicity (kg 2,4-D-eq/kg)
TRACI34%± 24%48%± 25%26%± 12%
Carcinogenics (kg benzene-eq/kg)5.36E-3 ± 1.15E-22.23E-2 ± 3.34E-22.88E-3 ± 1.23E-2
TRACI49%± 20%59%± 22%38%± 20%
Noncarcinogenics (kg toluene-eq/kg)6.98 ± 7.9938.56 ± 47.8912.96 ± 16.91
TRACI77%± 16%70%± 29%62%± 34%
Respiratory effects (kg PM2.5-eq/kg)1.92E-3 ± 1.06E-31.26E-2 ± 1.25E-22.31E-3 ± 2.09E-3
TRACI80%± 17%71%± 28%59%± 36%
Acidifaction (H+ moles-eq/kg)0.47 ± 0.262.84 ± 2.750.52 ± 0.41
TRACI28%± 27%54%± 37%47%± 23%
Eutrophication (kg N-eq/kg)8.30E-3 ± 2.81E-31.44E-2 ± 2.56E-24.08E-4 ± 2.72E-4
TRACI66%± 32%69%± 30%60%± 38%
Ozone depletion (kg CFC 11-eq/kg)1.27E-6 ± 8.10E-61.34E-6 ± 2.11E-63.35E-7 ± 5.43E-7
TRACI64%± 22%80%± 15%72%± 30%
Photochemical oxidation (kg NOx-eq/kg)6.45E-3 ± 5.41E-31.55E-2 ± 1.46E-22.60E-3 ± 1.85E-3

The energy-related contributions to GWP are at a consistently high level (about 80%) for all three groups of chemicals. Remaining GWP impacts are mostly due to organic wastewater emissions and the conversion of these into carbon dioxide (CO2) during waste treatment. These emissions can result in significant contributions to CO2 emissions in cases of low-yield reactions. The conclusion of these results is that energy production and use are important contributors to the environmental impacts of chemical production and are, according to many methods, responsible for the majority of impacts. Thus, increasing the energy efficiency of chemical processes and switching to efficient and, if possible, renewable heat and electricity sources is a very promising way of reducing the impacts of chemical production.

An analysis of the composition of the energy-related impacts for the CED, GWP, and EI99 and ES2006 total scores revealed that roughly half of the impacts are due to use of steam and other forms of heat; electricity is responsible for about 40% to 50% of the energy-related impacts, depending on the method; and transport causes the remaining impacts. All data are based on the use of a UCTE (i.e., a continental European) electricity mix. The electricity contributions would therefore increase somewhat when applied to locations with especially problematic power generation (e.g., Poland, the United States, China). Conversely, ensuring an environmentally benign source of electricity can be an efficient way to reduce overall environmental burdens. The methods of heat generation for the plant are also important, and improvements can be made through maximized use of environmentally efficient heat sources, such as natural gas, renewable energy, or thermal recycling of wastes.

In Ecoindicator 99, the remaining impacts of organic chemical production are mostly due to the resources used. Eighty-eight percent (ecoinvent data) and 67% (industry data) of these nonenergy-related impacts are resource impacts. As the inorganic chemicals are based not on fossil fuels but rather on more common minerals, their resource depletion impacts for material feedstock are low. An analysis of the processes for the organic chemicals revealed that resource uses are almost entirely due to the use of oil and natural gas as material feedstocks in the chemical industry. Optimizing mass efficiency and feedstock use is, hence, another obvious course of action, although efforts are already strong in this field, and the possibilities for improvements may be smaller.

Most striking, the combined human health and ecosystem quality impacts from nonenergy-related emissions in this analysis constitute only 5% (ecoinvent data) and 10% (industry data) of the overall impacts. This is a strong contrast to the still common perception of the chemical industry as polluting mostly due to harmful chemical emissions from the plant itself, and it shows that modernization efforts and stricter regulations have led to changes in the environmental profile of chemical production. Nonetheless, waterborne organic emissions are only poorly assessed in LCA. This is due to missing knowledge about the exact composition of chemical wastewaters and missing impact factors (Köhler et al. 2006, 2007). As a consequence, many organic emissions to water, unless separately measured as a single substance, were omitted from the current analysis. Additional efforts to reduce harmful emissions in chemical plants are therefore still valuable in industrial countries and especially valuable in developing countries, where emission standards may be less strict. Furthermore, researchers should still assess chemical production with traditional risk assessment techniques as well as LCA to ensure its sustainability. In general, the situation in emerging and developing economies, such as Brazil, India, and China, may vastly differ from the data analyzed in this study, both for energy production and for process emissions. As production data for these countries are even scarcer than for Europe, a thorough analysis is not possible at this moment, and the findings of this study should only be applied to industrialized countries.

Another important observation is that for the organic chemicals, the fraction of energy-related impacts increased as total impacts increased. Figure 2 shows that CED displays a trend towards higher fractions with increasing total scores. As the complexity of the production process and the overall impacts on the environment increase, energy-related impacts become more dominant in organic synthesis. This is also confirmed by other studies (Jiménez-Gonzalez et al. 2004; Wernet et al. 2010). Many advanced and fine chemicals require special purification steps, which are often energy-intensive; this further increases the importance of energy. This trend is also observable for the EI99 and the ES2006 aggregated scores, whereas for the GWP fractions are generally high no matter the absolute impacts. An analysis of the different subcategories of TRACI, ES2006, and EI99 showed that such a pattern is present for several but not all individual impact categories (see the Supporting Information on the Web). For the inorganic chemicals, no such trend was observable.

image

Figure 2. Fraction of energy-related impacts and aggregated cumulative energy demand (CED) impact scores for the organic chemicals from ecoinvent (white circles), the industry-data chemicals (black diamonds), and the inorganic chemicals (plus signs). MJ-eq = megajoule equivalents.

Download figure to PowerPoint

The results of this study show that energy use and production are significant factors in chemical production. The high-quality data provided directly by the chemical industry show similar results to the ecoinvent data, albeit with an even higher relevance of energy use. Given that consistent tracking of direct and especially indirect greenhouse gas emissions in the chemical industry is a recent development and is still not carried out by all members of the chemical sector's environmental initiative (Responsible Care 2009), the results for GWP in particular should be noted. Global warming is becoming one of the defining issues of sustainable production, and, as this study shows, it should be a key concern for the chemical sector, joining traditional focus points, such as pollution from chemical emissions. The results may be used as a decision-making aid for producers and policy makers striving to reduce environmental impacts in industrial countries by broadening the scope of environmental hazards of chemical production beyond the traditional focus on harmful air and water emissions. For production of an average chemical product, energy and resource use is a defining factor of the overall environmental profile. Midpoint indicators, such as the CED, can therefore be useful indicators of the overall impact. Tools to estimate the CED of a chemical's production, even in cases of severe data scarcity, exist (Wernet et al. 2008, 2009), and our results show that they can be useful for streamlined and screening LCAs if a process-based analysis of a chemical production is not feasible. This is extremely helpful for closing an important data gap in LCA, as LCI data on chemicals are often confidential and therefore still scarce. The results also confirm that an LCA perspective is vital in assessments of the overall environmental impacts of chemical production, as many relevant resource uses and emissions do not occur within the plant itself. Moreover, the findings indicate that LCA can identify environmental problem areas even in industries where sources of pollution were previously expected elsewhere. Industry initiatives, such as Responsible Care, are starting to focus more on indirect emissions, but not all participants report these yet. A stronger focus on energy use and indirect emissions within the chemical industry could help to improve the overall environmental performance of the chemical sector in industrial countries.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

We thank Ciba AG, Basel (now a part of BASF AG), for supplying us with production data for use in this project. We also thank the Swiss Federal Office for the Environment (Project No. 810.3189.004) and the Swiss Federal Office for Energy (Project No. 101711) for their support in this work.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information
  • Bare, J., G. Norris, D. Pennington, and T. McKone. 2003. TRACI: The tool for the reduction and assessment of chemical and other environmental impacts. Journal of Industrial Ecology 6(3–4): 4978.
  • Bare, J., T. Gloria, and G. Norris. 2006. Development of the method and U.S. normalization database for life cycle impact assessment and sustainability metrics. Environmental Science & Technology 40(16): 51085115.
  • Bosch, M. E., S. Hellweg, M. A. J. Huijbregts, and R. Frischknecht. 2007. Applying cumulative exergy demand (CExD) indicators to the ecoinvent database. International Journal of Life Cycle Assessment 12(3): 181190.
  • Bretz, R. and P. Frankhauser. 1997. Life-cycle assessment of chemical production processes: A tool for ecological optimization. Chimia 51(5): 213217.
  • Capello, C., S. Hellweg, B. Badertscher, H. Betschart, and K. Hungerbühler. 2007a. Environmental assessment of waste-solvent treatment options—Part 1: The ecosolvent tool—Environmental assessment of waste-solvent treatment options. Journal of Industrial Ecology 11(4): 2638.
  • Capello, C., U. Fischer, and K. Hungerbühler. 2007b. What is a green solvent? A comprehensive framework for the environmental assessment of solvents. Green Chemistry 9(9): 927934.
  • Capello, C., S. Hellweg, and K. Hungerbühler. 2008. Environmental assessment of waste-solvent treatment options—Part II: General rules of thumb and specific recommendations. Journal of Industrial Ecology 12(1): 111127.
  • Capello, C., G. Wernet, J. Sutter, S. Hellweg, and K. Hungerbühler. 2009. A comprehensive environmental assessment of the petrochemical solvent production. International Journal of Life Cycle Assessment 14(5): 467479.
  • Constable, D. J. C., A. D. Curzons, L. M. F. dos Santos, G. R. Geen, R. E. Hannah, J. D. Hayler, J. Kitteringham, et al. 2001. Green chemistry measures for process research and development. Green Chemistry 3(1): 79.
  • Curzons, A. D., C. Jiménez-Gonzalez, A. L. Duncan, D. J. C. Constable, and V. L. Cunningham. 2007. Fast Life Cycle Assessment of Synthetic Chemistry (FLASC) Tool. International Journal of Life Cycle Assessment 12(4): 272280.
  • DeWulf, J., M. E. Bosch, B. De Meester, G. van der Vorst, H. van Langenhove, S. Hellweg, and M. A. J. Huijbregts. 2007. Cumulative exergy extraction from the natural environment (CEENE): A comprehensive life cycle impact assessment method for resource accounting. Environmental Science & Technology 41(24): 84778483.
  • ecoinvent centre. 2008. Ecoinvent 2.01 database. http://www.ecoinvent.ch. Accessed 2009.
  • Frischknecht, R., R. Steiner, A. Braunschweig, N. Egli, and G. Hildesheimer. 2006. Swiss ecological scarcity method: The new version 2006. http://www.esu-services.ch/cms/fileadmin/download/Frischknecht-2006-EcologicalScarcity-Paper.pdf. Accessed 2009.
  • Geisler, G., S. Hellweg, T. B. Hofstetter, and K. Hungerbühler. 2005a. Life-cycle assessment in pesticide product development: Methods and case study on two plant-growth regulators from different product generations. Environmental Science & Technology 39(7): 24062413.
  • Geisler, G., S. Hellweg, and K. Hungerbühler. 2005b. Uncertainty analysis in life cycle assessment (LCA): Case study on plant-protection products and implications for decision making. International Journal of Life Cycle Assessment 10(3): 184192.
  • Goedkoop, M. and R. Spriensma. 2000. The Eco-Indicator 99: A damage orientated method for life-cycle impact assessment. http://www.pre.nl/eco-indicator99. Accessed 2009.
  • Goedkoop, M., R. Heijungs, M. A. J. Huijbregts, A. De Shruyver, J. Struijs, and R. van Zelm. 2009. ReCiPe 2008—A life cycle impact assessment method which comprises harmonised category indicators at the midpoint and the endpoint level. The Hague : Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer [Ministry of Housing, Spatial Planning, and the Environment.
  • Hellweg, S., U. Fischer, M. Scheringer, and K. Hungerbühler. 2004. Environmental assessment of chemicals: Methods and application to a case study of organic solvents. Green Chemistry 6(8): 418427
  • Huijbregts, M. A. J., L. J. A. Rombouts, S. Hellweg, R. Frischknecht, J. Hendriks, D. Van de Meent, A. M. J. Ragas, L. Reijnders, and J. Struijs. 2006. Is cumulative fossil energy demand a useful indicator for the environmental performance of products? Environmental Science & Technology 40(3): 641648.
  • IPCC (Intergovernmental Panel on Climate Change). 2007. Fourth report: Climate change 2007. http://www.ipcc.ch/ipccreports/ar4-wg1.htm. Accessed 2007.
  • Jankowitsch, O., L. Cavin, U. Fischer, and K. Hungerbühler. 2001. Environmental and economic assessment of waste treatment alternatives under uncertainty. Process Safety and Environmental Protection 79(B5): 304314.
  • Jiménez-Gonzalez, C., A. D. Curzons, D. J. C. Constable, and V. L. Cunningham. 2004. Cradle-to-gate life cycle inventory and assessment of pharmaceutical compounds. International Journal of Life Cycle Assessment 9(2): 114121.
  • Jolliet, O., M. Margni, R. Charles, S. Humbert, J. Payet, G. Rebitzer, and R. Rosenbaum. 2003. IMPACT 2002+: A new life cycle impact assessment methodology. International Journal of Life Cycle Assessment 8(6): 324330.
  • Köhler, A., S. Hellweg, B. I. Escher, and K. Hungerbühler. 2006. Organic pollutant removal versus toxicity reduction in industrial wastewater treatment: The example of wastewater from fluorescent whitening agent production. Environmental Science & Technology 40(10): 33953401.
  • Köhler, A., S. Hellweg, E. Recan, and K. Hungerbühler. 2007. Input-dependent life-cycle inventory model of industrial wastewater-treatment processes in the chemical sector. Environmental Science & Technology 41(15): 55155522.
  • Kralisch, D., A. Stark, S. Korsten, G. Kreisel, and B. Ondruschka. 2005. Energetic, environmental and economic balances: Spice up your ionic liquid research efficiency. Green Chemistry 7(5): 301309.
  • McKay, M., M. S. Reisch, A. H. Tullo, P. L. Short, J. F. Tremblay, and W. J. Storck. 2007. Facts and figures of the chemical industry. Chemical & Engineering News 85(27): 2968.
  • Mihelcic, J. R., J. C. Crittenden, M. J. Small, D. R. Shonnard, D. R. Hokanson, Q. Zhang, H. Chen, S. A. Sorby, V. U. James, J. W. Sutherland, and J. L. Schnoor. 2003. Sustainability science and engineering: The emergence of a new metadiscipline. Environmental Science & Technology 37(23): 53145324.
  • Mutel, C. and M. Kestenholz. 2008. The Brightway LCA model. http://brightwaylca.org. Accessed 2009.
  • Responsible Care. 2009. Responsible Care status report 2009. http://www.responsiblecare.org/filebank/Status%20Report%2001_05.pdf. Accessed 2009.
  • Romero-Hernandez, O. 2004. To treat or not to treat? Applying chemical engineering tools and a life cycle approach to assessing the level of sustainability of a clean-up technology. Green Chemistry 6(8): 395400.
  • Romero-Hernandez, O., E. N. Pistikopoulos, and A. G. Livingston. 1998. Waste treatment and optimal degree of pollution abatement. Environmental Progress 17(4): 270277.
  • Rosenbaum, R. K., T. M. Bachmann, L. S. Gold, M. A. J. Huijbregts, O. Jolliet, R. Juraske, A. Koehler, et al. 2008. USEtox—The UNEP-SETAC toxicity model: Recommended characterisation factors for human toxicity and freshwater ecotoxicity in life cycle impact assessment. International Journal of Life Cycle Assessment 13(7): 532546.
  • Saling, P., A. Kicherer, B. Dittrich-Krämer, R. Wittlinger, W. Zombik, I. Schmidt, W. Schrott, and S. Schmidt. 2002. Eco-efficiency analysis by BASF: The method. International Journal of Life Cycle Assessment 7(4): 203218.
  • Shonnard, D. R., A. Kicherer, and P. Saling. 2003. Industrial applications using BASF eco-efficiency analysis: Perspectives on green engineering principles. Environmental Science & Technology 37(23): 53405348.
  • Szijjarto, A., S. Papadokonstadakis, U. Fischer, and K. Hungerbühler. 2008. Bottom-up modeling of the steam consumption in multipurpose chemical batch plants focusing on identification of the optimization potential. Industrial and Engineering Chemistry Research 47(19): 73237334.
  • VDI (Verein Deutscher Ingenieure). 1997. VDI-richtlinie 4600: Cumulative energy demand, terms, definitions, methods of calculation. Düsseldorf , Germany : Verein Deutscher Ingenieure.
  • Wernet, G., S. Hellweg, U. Fischer, S. Papadokonstantakis, and K. Hungerbühler. 2008. Molecular-structure-based models of chemical inventories using neural networks. Environmental Science & Technology 42(17): 67176722.
  • Wernet, G., S. Papadokonstadakis, S. Hellweg, and K. Hungerbühler. 2009. Bridging data gaps in environmental assessments: Modeling impacts of fine and basic chemical production. Green Chemistry 11(11): 18261831.
  • Wernet, G., S. Conradt, H. P. Isenring, C. Jiménez-Gonzalez, and K. Hungerbühler. 2010. Life cycle assessment of fine chemical production: A case study of pharmaceutical synthesis. International Journal of Life Cycle Assessment 15(3): 294303.

About the Authors

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

Gregor Wernet was a postdoctoral scientist at the Institute for Chemical and Bioengineering at ETH Zurich in Zurich, Switzerland, at the time the article was written. He is now project manager at the ecoinvent centre, Switzerland. Christopher Mutel is a graduate student and Stefanie Hellweg is a professor at the Institute for Environmental Engineering at ETH Zurich in Zurich, Switzerland. Konrad Hungerbühler is a professor at the Institute for Chemical and Bioengineering, also at ETH Zurich in Zurich, Switzerland.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. About the Authors
  9. Supporting Information

Supplement S1: The supporting information for this article includes a list of all ecoinvent data sets used for the study, a general description of the composition of the industry data sets, a full table of all results for all impact categories for all methods used, additional graphs demonstrating the increase in the fraction of energy-related impacts for other methods, additional information on the method used to identify energy-related impacts, and information on data quality and treatment.

FilenameFormatSizeDescription
JIEC_294_sm_SuppMat.pdf140KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.