Appreciating the Role of Thermodynamics in LCA Improvement

Mar 1, 2011 - Daniel A. Eisenberg, Khara D. Grieger, Danail R. Hristozov, Matthew E. Bates, Igor Linkov. Risk Assessment, Life Cycle Assessment, and ...
0 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/est

Appreciating the Role of Thermodynamics in LCA Improvement Analysis via an Application to Titanium Dioxide Nanoparticles Geoffrey F. Grubb† and Bhavik R. Bakshi*,†,‡ †

William G. Lowrie Department of Chemical and Biomolecular Engineering, The Ohio State University, Columbus, Ohio 43210, United States ‡ Department of Energy and Environment, TERI University, New Delhi, 110 070, India

bS Supporting Information ABSTRACT: Although many regard it as the most important step of life cycle assessment, improvement analysis is given relatively little attention in the literature. Most available improvement approaches are highly subjective, and traditional LCA methods often do not account for resources other than fossil fuels. In this work exergy is evaluated as a thermodynamically rigorous way of identifying process improvement opportunities. As a case study, a novel process for producing titanium dioxide nanoparticles is considered. A traditional impact assessment, a first law energy analysis, and an exergy analysis are done at both the process and life cycle scales. The results indicate that exergy analysis provides insights not available via other methods, especially for identifying unit operations with the greatest potential for improvement. Exergetic resource accounting at the life cycle scale shows that other materials are at least as significant as fossil fuels for the production of TiO2 nanoparticles in this process.

’ INTRODUCTION Traditional life cycle assessment methodologies focus on emissions and impacts over all stages in the product or process supply and use network and often include an accounting of the energy consumption. Here energy usually includes the reliance on fossil fuels. However, as defined by the first law of thermodynamics, energy is never consumed but always conserved.1 It is the quality of the energy source that is degraded by a minimum amount defined by the second law. It is exergy that is consumed, not energy. Exergy represents the maximum amount of useful work available in a stream due to its difference from the reference state. This quantity provides a thermodynamically rigorous way of accounting for losses both locally and overall in a process. These attractive properties of exergy analysis have been used extensively for understanding and improving industrial systems. Several studies and practical applications demonstrate that exergy losses can provide insight not otherwise available.1-4 In particular, exergy places value on heat according to the work it can provide. First law methods lack this ability and can therefore exaggerate the value of low quality heat. Process level exergy analysis is typical in the literature for energy systems such as power plants and district heating systems as well as some materials processing systems such as metallurgical.5 These studies show how exergy can help in identifying opportunities for improving industrial systems. Despite the popularity of exergy in engineering, it has been dismissed in some of the life cycle literature as irrelevant. For example, methods like Eco-indicator 99 consider exergy analysis as too abstract to be helpful in LCA.6 The Eco-indicator literature cites an article by Mueller-Wenk,7 who argues that exergy evaluation is not helpful for quantifying resource depletion because it values the entirety of virgin ores and not just the mineral of interest. It can be argued, however, that the exergy of the whole ore should be accounted for, as it is being used and discarded as waste r 2011 American Chemical Society

in the mining process. The natural state of the ore has been disturbed, and there is an environmental price associated with that disruption. On the other hand, exergy methods are available which allow for the valuation of only the minerals of interest.1 Despite such criticism, the scientific rigor of exergy has attracted many researchers to its ability of accounting for resource use.1,8-10 Because exergy can account for materials and fuels and represents the maximum amount of work available in a resource or the minimum work that must be lost in a process, it is considered to provide a metric closer to the true value of resources to mankind than methods which consider only the difficulty of extraction. A missing component of these efforts is that the ability of exergy analysis to provide insight into opportunities for improving the life cycle has still not been explored. The purpose of the current work is to bridge this gap by comparing directly the insight available from various methods for informing environmental decision making. Unlike most of the previous work, the emphasis here is not just on resource accounting but more on improving the industrial processes and life cycles. The methods used here include traditional life cycle assessment, first law energy analysis, and exergy analysis. Each method is applied to a process for producing titanium dioxide nanoparticles. Identical boundaries are considered to avoid bias in the comparison. Both a process level (gate-to-gate) and a life cycle level (cradle-to-gate) boundary are used. The results and implications for improvement provided by each method are compared and critiqued. In addition to the insight about the ability to identify improvement opportunities via exergy analysis, this paper

Received: July 30, 2010 Accepted: February 7, 2011 Revised: February 6, 2011 Published: March 01, 2011 3054

dx.doi.org/10.1021/es1025855 | Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

ARTICLE

also presents original data and results for titania nanoparticles that extends the results presented in ref 11. The next section of this paper summarizes how exergy analysis was applied in this work. This is followed by a brief description of the titanium dioxide nanomanufacturing case study along with insight about how various methods can be used for identifying improvement opportunites. Ideas for future work are discussed in the final section. The Supporting Information contains the background about other methods relevant to identifying improvements and details of the calculations and data.

’ APPROACH This work explores a thermodynamic approach for identifying improvement opportunities. In a first law energy analysis, energy is conserved, and the balance over a given control volume is the difference between all entering and leaving flows. The sum of all flows in an energy balance must be zero as conveyed by this equation12 0¼

_ in - ∑W _ out ∑ðmHÞ _ in - ∑ðmHÞ _ out þ ∑W þ ∑Q_ in - ∑Q_ out

ð1Þ

where enthalpy, work, and heat are flowing in and out of the control volume. While energy is conserved, some flows are usually considered to be losses. Most often these losses are in the form of waste streams or unusable heat. When the second law of thermodynamics is included, irreversibilities introduce a different kind of loss. Due to entropic losses, exergy is not conserved. Any operation which is not reversible will have some exergy loss, commonly called lost work. In a realistic system, there will always be some positive value of lost work. The exergy balance may be written as follows12









_ in - W _ out LW ¼ ðmBÞ _ in - ðmBÞ _ out þ W       T0 T0 þ >0 Q_ 1 Q_ 1 T in T out





ð2Þ

Comparing these two equations also clarifies some of the differences between energy and exergy analysis. The most obvious difference is the Carnot efficiency factor applied to heat flows in the lost work calculation. For a given process or unit operation, these balance equations can be applied to find the energy losses and exergy losses or lost work. Such calculations are facilitated by the availability of process information such as pressures, temperatures, and composition of each stream. For improvement purposes, these metrics can be very enlightening. Logically, the unit operations with the greatest energy losses and/or lost work also should have the greatest potential for improvement. Design alternatives which move toward the reversible limit will have smaller lost work values. The methods of life cycle assessment and exergy analysis are well established in the open literature and have been applied in many different situations. The techniques presented here are not new in this regard, and established protocol is followed. The novelty of this work arises from applying these methods to the same system at two different scales and comparing the insight for improvement that each provides. The traditional engineering boundary contains only the local process. This scale is important for industry because they have direct control over the equipment and operating conditions in their plants. However, the decisions made at the process level have implications beyond the process

scale boundaries. In addition, information from outside the local boundary is very important when making decisions within it. For these reasons, it is crucial that the relationships between the process and life cycle scales are considered. A change in a process, which is at first perceived to be an improvement, may be just the opposite when the wider implications are accounted for. The goal of improvement analysis is to identify the parts of a process with the greatest potential for improvement. It is assumed that the areas with the most harmful emissions, largest losses, or lowest efficiencies will offer these opportunities. At the other end of the spectrum, a process unit running at a high efficiency with few losses or emissions can scarcely be improved. It is not expected that the results of an environmental impact assessment will necessarily agree with the results of an exergy analysis. Differences, where they exist, should provide insight about the value of each method in given situations and across scales.

’ APPLICATION TO MANUFACTURING PROCESS AND LIFE CYCLE OF TITANIUM DIOXIDE NANOPARTICLES Manufacturing by Hydrochloride Process. In recent years, a market has opened up for anatase phase titanium dioxide (TiO2) nanoparticles. These nanoparticles are pervasive in the cosmetics industry, used especially for cohesion and sun protection in sunscreens and skin care products. Anatase nanoparticle coatings are also photocatalytic when exposed to ultraviolet light. This unique property has led to applications such as antibacterial coatings or so-called self-cleaning surfaces. These coatings have the added advantage of holding up indefinitely over long exposure to sunlight. The Altair hydrochloride process is a novel process for the production of titanium dioxide nanoparticles. This process currently exists at the pilot scale with a larger plant under construction. The following process description is based on patent literature and articles published by Altair in the open literature.13-19 These sources also provide the data used for the calculations and results in the sequel. Ilmenite ore is the main feedstock to the process. The ideal chemical formula for ilmenite is FeTiO3, but impurities and variations are always present. Although ilmenite is common in many places around the world, the largest sources of ilmenite ore from mining are Australia, South Africa, and Canada.20 Ilmenite is first introduced in the digestion unit, along with an excess of concentrated hydrochloric acid (HCl) resulting in the following reaction

FeTiO3 þ 4HCl f FeCl2 þ TiOCl2 þ 2H2 O

ð3Þ

Most of the acid is recovered in subsequent steps and recycled back into the digestion, but a small feed is required to maintain the necessary level. After digestion, the goal is to remove the iron and chloride ions before reacting to form the titanium dioxide. The mixture is reduced with the addition of a small amount of iron powder to ensure that all the iron is in the form of Fe2þ, so that it will crystallize out as iron chloride (FeCl2) when the temperature is lowered to about 4 °C 2Feþ3 þ Fe f 3Feþ2

ð4Þ

The FeCl2 crystals are then filtered from the product stream and sent to the pyrohydrolysis unit that will be described later. After filtration, the product stream goes through a solvent extraction with trialkyl phosphine oxide to remove any residual 3055

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

ARTICLE

iron ions. The last separation step is an ion exchange with NaOH to remove remaining chloride ions. At this point the clean product stream is ready for reaction. The reaction in eq 5 is carried out in a modified spray dryer. A large amount of methane is burned at this stage to dehydrate the product. This is the innovative part of the hydrochloride process that allows for production of uniformly distributed nanoparticles. The spray hydrolysis reaction produces high purity titanium dioxide in the form of hollow spheres a few micrometers in diameter via the following reaction TiOCl2 þ H2 O f TiO2 þ 2HCl

inputs

ð5Þ

The hydrochloric acid produced in this reaction is sent to distillation to be recycled. These spheres are further processed in a wet media mill and other optional coating steps. The end product is 40 nm anatase phase TiO2 nanoparticles. The current pilot scale plant produces approximately 100 kg/h of product from an ilmenite feed of about 210 kg/h. In a parallel recycle loop a pyrohydrolysis reaction is used to regenerate the HCl from the FeCl2 crystals 1 2FeCl2 þ 2H2 O þ O2 f Fe2 O3 þ 4HCl 2

Table 1. Mass, Energy, and Exergy Values of Inputs Needed To Produce 1 Kilogram (kg) of Titanium Dioxide Nanoparticles, from ref 11

ð6Þ

Iron oxide (Fe2O3) is produced as a byproduct, and the regenerated HCl is sent to a swing distillation unit to break the water-HCl azeotrope before it is recycled back to the digestion unit. High-pressure steam is used to run the distillation. The six major inputs to the process are ilmenite, iron powder, hydrochloric acid, methane, high pressure steam, and electricity. These inputs and their commercial production chains form the basis of the life cycle boundary considered in this study. Calculations. A traditional damage assessment, first law energy analysis, and exergy analysis have been done at both the process and life cycle scales. The calculations at the process scale are based on process information available in patents and other articles cited in the previous subsection. Emissions for individual equipment in process level damage assessment had to be estimated in most cases because information was not available in the literature or from Altair. A conservative 1% fugitive emission rate was assumed for the volatile organics. The full amount of carbon dioxide from the combustion of methane was assumed to be released to the atmosphere as the literature did not indicate otherwise. The first law energy content of all streams in the process was calculated as the enthalpy and fuel value where appropriate. The exergy was calculated as the sum of the physical and chemical exergy as calculated by equations provided in the background in the Supporting Information. Energetic and exergetic losses and efficiencies were calculated at each of the eight major processing units. The units with the greatest impacts are identified as those having the largest potential for improvement based on the impact assessment. Likewise, units with the greatest exergetic losses and lowest efficiencies are considered to have the greatest potential from an exergetic point of view. The similarities and differences between these results should provide insight into the value of each method for this specific process. Details are provided in the Supporting Information. A few streams were not included in this study because sufficient information could not be obtained or the chemical exergy was not readily calculable. Omissions include the separation agents for the solvent extraction and ion exchange units. In addition, the impacts of the nanoparticle product, including size-

mass (kg/kg TiO2)

energy (MJ)

exergy (MJ)

ilmenite

2.165

0

1.928

iron powder

0.103

0

0.691

hydrochloric acid

0.065

0

0.151

methane

0.866

44.894

46.690

steam electricity

14.948 -

2.559 5.443

1.979 5.443

dependent property variations, are currently unknown. Studies on the toxicity of titanium dioxide and other nanoparticles are currently in the early stages. For these reasons, the present analysis should be considered a best case scenario for this process. When more is known about the possible effects of the nanoparticles, it will be an easy matter to integrate that information into the current analysis. At the life cycle scale, the impact assessment was done using SimaPro 7.1. The six major inputs to the process were identified as ilmenite, iron powder, hydrochloric acid, methane, steam, and electricity. The amounts of each of these inputs required to produce one kilogram of product are shown in Table 1. These values were entered into the SimaPro software, where the raw materials, energy, and environmental impacts due to the upstream processing of each input was calculated. The impact assessment was taken from this output, and the material and fuel inventory obtained was used as the basis of the energy and exergy analyses. Physical exergy is neglected at this scale because properties such as temperature and pressure are unknown. The exergy analysis at the life cycle scale consists of quantifying the chemical exergy of all the raw materials required for the upstream processing of the major inputs to the nanomanufacturing process. The exergy of basic elements was taken from refs 5 and 21, while exergy factors for more complex compounds were calculated by the authors according to equations in the SI. Details of these calculations can be found in the additional information of ref 11. Process Scale Results. As shown in Figure 1, the impact assessment at the process scale identified global warming potential (GWP) and human toxicity potential (HTP) as the most significant impact categories due to direct emissions from the process. Fugitive emissions of hydrochloric acid contribute to the HTP in the early stages of the process. Release of carbon dioxide from the combustion of methane as well as fugitive emissions of methane make up the GWP in the hydrolysis units. Improvement analysis based on these observations would suggest prevention, collection, or sequestration of the carbon dioxide produced as well as careful scrutiny of fugitive emissions of harmful chemicals such as hydrochloric acid. The energy and exergy results present a more interesting view of the process. First and second law consumption for each major step in the process are displayed in Figure 2. The first law energy demand is highest for the pressure swing distillation stage. These losses represent a large amount of warm water leaving the process. This might suggest that the distillation unit has the highest potential for improvement. However, exergetic losses tell a different story. The rejected warm water has a much lower exergy content because it is no longer very useful. In other words, it is low quality heat. Exergetic losses in the process suggest that improvement efforts should instead be concentrated on the spray hydrolysis stage. 3056

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

ARTICLE

Figure 1. Normalized environmental impacts at the process scale for production of 1 kg of TiO2 nanoparticles. Categories include photochemical ozone creation potential (POCP), global warming potential (GWP), and human toxicity potential (HTP).

Figure 2. Energy and exergy consumption for unit operations at the process scale. Both metrics expressed in MJ required to produce 1 kg of TiO2 nanoparticles.

This is confirmed in Figure 3, where spray hydrolysis has the lowest thermodynamic efficiency of any of the process units, operating around 12%. Pyrohydolysis and distillation also warrant investigation as their exergetic efficiencies are only around 30%. Another very effective way to present the exergy flows in the process is through a Grassmann diagram. Presented in Figure 4, the thicknesses of the arrows between unit operations are directly proportional to the magnitude of the exergy flows. These diagrams express a great deal of information in a single figure. At a glance, one can see the relative size, losses, and efficiency of

all the flows and unit operations in the system. For example, in Figure 4, it is easy to see that the spray hydrolysis and distillation units are the largest exergy users, while the material intensive operations early in the process are relatively efficient. Life Cycle Scale Results. The life cycle results presented here are in addition to those shown in previous work.11 Energetic and exergetic efficiencies are shown in Figure 5. The first and second law efficiencies are nearly identical for the upstream production processes for the inputs required by the Altair hydrochloride process. This result demonstrates one reason why exergy analysis 3057

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

ARTICLE

Figure 3. Energetic and exergetic efficiencies for unit operations at the process scale.

is discounted as irrelevant in some of the literature, as discussed in Section 1. However, looking closer at the life cycle scale in the Grassmann diagram, Figure 6(a), there is valuable information available from exergy analysis. In particular, the diagram clearly shows that while the material inputs represent very small deliveries of exergy to the process, their upstream value chains are just as exergy intensive as those needed for the fuel inputs. This is due to the material intensity of the upstream production processes for these inputs. Material losses cannot be adequately accounted for in a first law analysis. The Ecological Cumulative Exergy Consumption (ECEC) Grassmann diagram in Figure 6(b) shows the cumulative amount of exergy required to produce the six major inputs to the process. ECEC is expressed in solar emergy joule equivalents as discussed in the Supporting Information and is a quality corrected thermodynamic quantity. Comparing this diagram to part (a) of the figure reveals that natural gas, iron, and crude oil have the largest ECEC footprints relative to the other inputs. In addition, ilmenite, methane, and steam are confirmed as the most exergy intensive inputs. From an improvement standpoint, this should prioritize efforts to reduce methane, steam, and ilmenite requirements to the process before any others. Discussion. Looking across scales at the impact assessment and exergy analysis in an improvement context, several insights present themselves. Thermodynamic losses and harmful emissions at the process scale are more intense where the fuels and energy inputs are being used. The combustion of methane for both hydrolysis units presents a large loss of exergy and emits large amounts of greenhouse gases. In addition, high energy steam is used destructively for distillation. For these reasons, improvement at the process scale should focus around the fuel intensive steps. A first effort might include heat integration to reduce fuel requirements. At the broader life cycle scale, fossil fuel losses are concentrated in the production of methane, steam, and electricity. However, exergy analysis reveals that the material inputs see significant losses in their production streams not only from fossil

fuel use but also from the materials and natural resources used to extract them from the environment. This insight implies that newer technologies may be highly materials intensive, as discussed in ref 22. The material inputs are much more diffuse and so require more material and natural resources such as water to produce usable concentrations. At this scale, improvement efforts should be directed just as much toward the material input supply chains, especially ilmenite mining and hydrochloric acid production. Enhancing the material efficiency via recycling and remanufacturing may also help.

’ FUTURE WORK This work demonstrates the benefit of exergy analysis for identifying improvement opportunities at the process and life cycle scales. One possibility this research raises is that exergy analysis can be used to effectively bridge the gap between traditional LCA practices and engineering process design. LCA currently concentrates on the evaluation of processes, quantifying emissions and their impacts. When these methods involve feedback, it is usually after the fact in the form of criticism or with end of the pipe suggestions for reducing impacts. Ideally, impact evaluation would be included during the development stages of a process and be used as a tool to inform the design process. This sounds much easier than it is in practice, because there are issues of scale. Engineering process design is done at the local scale, concerning itself with things like equipment size or operating temperatures and pressures. These things are very important, but the broader environmental implications of these decisions are far from obvious at the design level of scrutiny. It is only after considering many other factors, such as the upstream supply chains, or the interaction with the local and global environment or economy, that the true impacts of design decisions can be realized. In addition, more studies like the one in this article may allow for the development of general environmental heuristics for process design, considering both the process and life cycle 3058

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

ARTICLE

Figure 4. Grassmann diagram of exergy flows at the process level. Exergy flows expressed as MJ/kg of TiO2 produced. Arrows indicate the direction of flow; hash marks indicate lost work.

Figure 5. Energetic and exergetic efficiencies at the life cycle scale. These results represent the overall efficiency of the upstream processing required to produce each of the main inputs to the process. Error bars represent the effect of excluding turbine water from the analysis. 3059

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

ARTICLE

Figure 6. Life cycle scale Grassmann diagrams of (a) upstream exergy flows expressed in MJ/kg of product and (b) cumulative exergy (ECEC) flows expressed in solar emergy joules.

implications of design alternatives. Such work may benefit from the use of thermodynamics-based metrics such as indices for renewability, return on investment, and sustainability. Of course, after identifying the improvement opportunity, the challenge of how to modify the existing system remains and will require creative engineering solutions.

’ ASSOCIATED CONTENT

bS

Supporting Information. Background information, detailed examples of calculations, and data tables. This material is available free of charge via the Internet at http://pubs.acs. org.

3060

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061

Environmental Science & Technology

’ AUTHOR INFORMATION Corresponding Author

*Phone: (614)292-4904. Fax: (614)292-6591. E-mail: bakshi.2@ osu.edu.

’ ACKNOWLEDGMENT The research presented in this article was supported by the U.S. National Science Foundation (Grant No. EEC-0425626) and the U.S. Environmental Protection Agency (Grant No. R832532).

ARTICLE

Pigment Process. Proceedings of the Sohn Symposium on Advanced Processing of Metals and Materials, 2006. (19) Verhulst, D.; Sabacky, B.; Lang, J.; Ellsworth, D. Iron Control in the Altair Hydrochloride Pigment Process. Proceedings of the 3rd International Symposium on Iron Control in Hydrometallurgy, 2006. (20) Survey, U. G. Titanium Mineral Concentrates, 2006. (21) Rivero, R.; Garfias, M. Standard Chemical Exergy of Elements Updated. Energy 2006, 31, 3310–3326. (22) Gutowski, T. G.; Branham, M. S.; Dahmus, J. B.; Jones, A. J.; Thiriez, A.; Sekulic, D. P. Thermodynamic Analysis of Resources Used in Manufacturing Processes. Environ. Sci. Technol. 2009, 43, 1584–1590.

’ REFERENCES (1) Dewulf, J.; Van Langenhove, H.; Muys, B.; Bruers, S.; Bakshi, B. R.; Grubb, G. F.; Paulus, D. M.; Sciubba, E. Exergy: Its potential and limitations in environmental science and technology. Environ. Sci. Technol. 2008, 42, 2221–2232. (2) Meester, B. D.; Dewulf, J.; Verbeke, S.; Janssens, A.; Langenhove, H. V. Exergetic Lifecycle Assessment (ELCA) for Resource Consumption Evaluation in the Built Environment. Build. Environ. 2009, 44, 11–17. (3) Rosen, M. A.; Dincer, I. Thermoeconomic Analysis of Power Plants: An Application to a Coal Fired Electrical Generating Station. Energy Convers. Manage. 2003, 44, 2743–2761. (4) Meyer, L.; Tsatsaronis, G.; Buchgeister, J.; Schebek, L. Exergoenvironmental Analysis for Evaluation of the Environmental Impact of Energy Conversion Systems. Energy 2009, 34, 75–89. (5) Szargut, J. Exergy Method: Technical and Ecological Applications; WIT Press: 2005. (6) Goedkoop, M.; Spriensma, R. The Eco-indicator 99 A Damage Oriented Method for Life Cycle Impact Assessment: A Methodology Report; Technical Report; 2000. (7) Mueller-Wenk, R. Depletion of abiotic resources weighted on base of ”virtual” impacts of lower grade deposits used in future. IWO Diskussionsbeitrag 1998. (8) Ayres, R. U. Life-Cycle Analysis - A Critique. Resour., Conserv., Recycl. 1995, 14, 199–223. (9) Wall, G.; Gong, M. On exergy and sustainable development Part 1: Conditions and concepts. Exergy 2001, 1, 128–145. (10) Thermodynamics and the Destruction of Resources; Bakshi, B. R., Gutowski, T. G., Sekulic, D. P., Eds.; Cambridge University Press: 2011. (11) Grubb, G. F.; Bakshi, B. R. Life Cycle of Titanium Dioxide Nanoparticle Production - Impact of Emissions and Use of Resources. J. Ind. Ecol. 2010. (12) Seider,W. D.; Seader, J. D.; Lewin, D. R. Product and Process Design Principles: Synthesis, Analysis, and Evaulation, 2nd ed.; Wiley: 2003. (13) Duyvesteyn, W. P. C.; Spitler, T. M.; Sabacky, B. J.; Prochazka, J. Processing Titaniferous Ore to Titanium Dioxide Pigment, U.S. Patent 6375923, 2002. (14) Duyvesteyn, W. P. C.; Sabacky, B. J.; Verhulst, D. E. V.; WestSells, P. G.; Spitler, T. M.; Vince, A.; Burkholder, J. R.; Huls, B. J. P. M. Processing Aqueous Titanium Chloride Solutions to Ultrafine Titanium Dioxide, U.S. Patent 6440383, 2002. (15) Duyvesteyn, W. P. C.; Spitler, T. M.; Sabacky, B. J.; Vince, A.; Prochazka, J. Processing Aqueous Titanium Solutions to Titanium Dioxide Pigment, U.S. Patent 6548039, 2003. (16) Verhulst, D.; Sabacky, B.; Spitler, T.; Duyvesteyn, W. The Altair TiO2 Pigment Process and its Extension to the Field of Nanomaterials. CIM Bull. 2002, 95, 89–94. (17) Verhulst, D.; Sabacky, B.; Spitler, T.; Prochazka, J. New Developments in the Altair TiO2 Hydrochloride Pigment Process. Proceedings of the 5th International Conference in Honor of Professor Ian Ritchie 2003, 565–575. (18) Verhulst, D.; Sabacky, B.; Marganski, R.; Wang, B.; Ellsworth, D. Optimization of the Critical Steps of the Altair Hydrochloride 3061

dx.doi.org/10.1021/es1025855 |Environ. Sci. Technol. 2011, 45, 3054–3061