Dry Cleaning with Liquid Carbon Dioxide

The process of cleaning one item invariably involves making something else dirty. Whether that something else is an organic or halogenated solvent, soapy water, or a rag, we seldom address the dirtying that accompanies any cleaning process. If we are to achieve environmentally benign cleaning, we must look at the life cycle of solvents employed for cleaning, including the potential for recycling, reuse, or release into the environment. Truly “green” cleaning processes not only minimize the amount of waste generated; but also they prevent the dispersal of that waste into large amounts of solvent, water, soil, or air. Dense-phase carbon dioxide is a great cleaning solvent from a pollution-prevention viewpoint. By-product CO2 generated by other industrial processes can be captured, so it is not necessary to generate CO2 specifically for cleaning. Spills of CO2 will not contaminate groundwater or create a need for soil remediation. Carbon dioxide even has advantages for the work environment, since no chronic, harmful effects are known from repeated inhalation of low concentrations of CO2. The barriers to using CO2 as a cleaning solvent have centered around two issues: the expense of high-pressure equipment and the poor solubility of many contaminants in CO2. Micell Technologies, Inc., based in Raleigh, NC, has addressed the equipment issue by using liquid CO2 just below ambient temperature (∼18–22 °C) and vapor pressure (∼50 bar). The equipment needed to contain this pressure is considerably less expensive than that needed for supercritical CO2 processes. As for the second barrier, Micell has surfactant packages that enhance the ability of CO2 to dissolve many contaminants commonly found on clothes or on metal parts. Micell is in the process of designing and bringing to market integrated CO2 solutions, including equipment and appropriate chemistries, to replace the organic solvents or water traditionally used in garment dry cleaning, metal degreasing, and textile processing. Dry cleaning is a bit of a misnomer, in that clothes are cleaned in a liquid solvent. “Dry” simply means that exposure of a garment, such as a wool suit or silk blouse, to water is minimized to prevent damage to hydrophilic fibers.

Coatings from Liquid and Supercritical Carbon Dioxide

This chapter describes several aspects of the use of carbon dioxide as a solvent or cosolvent in coating applications. The primary impetus for using carbon dioxide for this purpose has been the alleviation of volatile emissions and liquid solvent wastes. However, the special physical properties of liquid and supercritical carbon dioxide may offer some processing advantages over conventional organic or aqueous solvents. Liquid carbon dioxide is quite compressible, and a reduction in temperature results not only in a reduction in the operating pressure, but also in a significant increase in the liquid density to values of approximately 0.9 g/cm3. At these high liquid densities, carbon dioxide exhibits improved solvent performance, but with much lower viscosities and interfacial tensions than aqueous or organic liquid solvents. Under supercritical conditions, carbon dioxide also exhibits high densities, low viscosities, and improved solvent power. Low viscosities and interfacial tensions tend to facilitate the transport of the solvents into any crevices or imperfections on the surface to be covered, and this might prove advantageous in the coating of patterned or etched surfaces. Since carbon dioxide dissolves and diffuses easily into many different polymers and organic liquids, it can also be used to reduce the viscosity of coating solutions. Whether in the liquid or the supercritical state, the temperature and pressure of the mixture can be used to control its physical properties in ways that are impossible to achieve with traditional solvents. These distinguishing features have raised the level of industrial interest in carbon dioxide as a solvent for coating applications, beyond those based solely on environmental concerns. In this chapter, we will discuss current applications and research on the use of CO2 as a solvent for coatings. The first section deals with spray coating from supercritical CO2. Subsequent sections deal with the use of liquid coatings, such as spin and free meniscus coatings, and impregnation coatings. Since the start of the 20th century (ca. 1907), atomization has been the basis for conventional spray coating applications (Muirhead, 1974). Typically, atomization is caused by high shear of the coating fluid in air, leading to droplet or particle formation.

Interfacial Phenomena with Carbon Dioxide Soluble Surfactants

A fundamental understanding of colloid and interface science for surfactant design in CO2-based systems is emerging on the basis of studies of interfacial tension and surfactant adsorption (da Rocha et al., 1999) along with complementary studies of colloid structure (Chillura-Martino et al., 1996; Meredith and Johnston, 1999; Wignall, 1999) and stability (Meredith and Johnston, 1999; O’Neill, 1997; Yates et al., 1997). The interfacial tension, γ, between a supercritical fluid (SCF) phase and a hydrophilic or lipophilic liquid or solid, along with surfactant adsorption, play a key role in a variety of processes including nucleation, coalescense and growth of dispersed phases, formation of microemulsions and emulsions (Johnston et al., 1999), particle and fiber formation, atomization, foaming (Goel and Beckman, 1995), wetting, adhesion, lubrication, and the morphology of blends and composites (Watkins et al., 1999). The first generation of research involving surfactants in SCFs addressed water/oil (w/o) microemulsions (Fulton and Smith, 1988; Johnston et al., 1989) and polymer latexes (Everett and Stageman, 1978) in ethane and propane (Bartscherer et al., 1995; Fulton, 1999; McFann and Johnston, 1999). This work provided a foundation for studies in CO2, which has modestly weaker van der Waals forces (polarizability per volume) than ethane. Consequently, polymers with low cohesive energy densities and thus low surface tensions are the most soluble in CO2: for example, fluoroacrylates (DeSimone et al., 1992), fluorocarbons, fluoroethers (Singley et al., 1997), siloxanes, and to a lesser extent propylene oxide. Since CO2 is nonpolar (unlike water) and has weak van der Waals forces (unlike lipophilic phases), it may be considered to be a third type of condensed phase. Surfactants with the above types of “CO2-philic” segments and a “CO2-phobic” segment have been used to form microemulsions (Harrison et al., 1994; Johnston et al., 1996), emulsions (da Rocha et al., 1999; Jacobson et al., 1999a; Lee et al., 1999b), and organic polymer latexes (DeSimone et al., 1994) in CO2. Microemulsion droplets are typically 2–10 nm in diameter, making them optically transparent and thermodynamically stable, whereas kinetically stable emulsion droplets and latexes in the range of 200 nm to 10 mm are opaque and thermodynamically unstable.

Synthesis and Characterization of Polymers: From Polymeric Micelles to Step-Growth Polymerizations

The benefits of using CO2 in polymer synthesis are numerous, ranging from environmental responsibility to improved materials properties. Carbon dioxide is an inert, nontoxic, nonflammable, and inexpensive reaction and processing medium that is an environmentally benign alternative to the organic solvents or water typically used today. Although the often toxic, carcinogenic, and environmentally hazardous organic solvents are recycled, some release to the environment is inevitable. Replacement of organic solvents with water still requires the costly purification of the wastewater prior to disposal and/or an energy-intensive drying process to remove the water. On the other hand, CO2 can be easily separated from other chemical components and recycled through depressurization and recompression. Although CO2 is a greenhouse gas, the CO2 used as a solvent does not contribute to the greenhouse gases since it is acquired from natural reservoirs or recovered as a by-product from other industrial chemical processes. The more specific environmental benefits of using liquid or supercritical carbon dioxide as a solvent vary depending on the polymerization being considered. The synthesis of fluoropolymers in CO2 is of particular interest since these polymers have historically been prepared in chlorofluorocarbons (CFCs) and other fluorinated solvents, as well as in water. Due to the association of CFCs with ozone-layer depletion, these solvents have been banned and replacement solvents must be found. Alternative fluorinated solvents are expensive and also have environmental concerns. In heterogeneous polymerizations, many polymer latexes produced by emulsion or dispersion polymerization in water or organic solvents can be produced in CO2. To eliminate volatile organic compound (VOC) emissions, more polymer latexes are being synthesized in water. However, for dry polymer applications, the latexes must undergo energy-intensive drying by vacuum or heat to remove the water. For polymer latexes produced in CO2, there are no VOC emissions and the energy-intensive drying step can be significantly reduced since the CO2 has a much lower heat of vaporization in the liquid state and, in fact, has a zero heat of vaporization in the supercritical state. Additionally, the polymer can be shipped dry at 100% solids, thus saving energy and money in shipping the heavy water latex.

Solubility of Polymers in Supercritical Carbon Dioxide

A great deal of information is known about the solvent character of CO2 with a wide range of polymers and copolymers based on well-characterized and systematic solubility studies that are available in the literature (Kirby and McHugh, 1999). Nevertheless, the prediction of polymer solubility in CO2, or any solvent for that matter, presents a formidable challenge since contemporary equations of state are still not facile enough to describe the unique characteristics of a long-chain polymer in solution. The difficulty resides in accounting for the intra- and intersegmental interactions of the many segments of the polymer connected to a single backbone relative to the small number of segments in a solvent molecule. An additional challenge exists to describe the density dependence of the intermolecular potential functions used in the calculations since SCF–polymer solutions (SCF, supercritical fluid) can be highly compressible mixtures. In this brief review, the solvent character of CO2 is described using the principles of molecular thermodynamics and also using a select number of phase behavior studies to reveal the impact of polymer architecture on solubility. To form a stable polymer–SCF solvent solution at a given temperature and pressure, the Gibbs energy, shown in eq. 7.1, must be negative and at a minimum. . . . ΔGmix = ΔHmix − T ΔSmix (7:1) . . . where ΔHmix and ΔSmix are the change of enthalpy and entropy, respectively, on mixing (Prausnitz et al., 1986). Enthalpic interactions depend predominantly on solution density and on polymer segment–segment, solvent–solvent, and polymer segment–solvent interaction energies. The value of ΔSmix depends on both the combinatorial entropy of mixing and the noncombinatorial contribution associated with the volume change on mixing, a so-called equation-of-state effect (Patterson, 1982). The combinatorial entropy always promotes the mixing of a polymer with a solvent. However, the noncombinatorial contribution can have a negative impact on mixing as a result of monomer–monomer interactions that arise due to the connectivity of the segments in the backbone of the polymer chain.

Fluorous Phases and Compressed Carbon Dioxide as Alternative Solvents for Chemical Synthesis: A Comparison

The principal goal of basic research in chemical synthesis is the development of efficient tools for functional group transformations and for the assembly of building blocks during the construction of molecules with increasing complexity. Traditionally, new approaches in this area have focused on the quest for new reaction pathways, reagents, or catalysts. Comparably less effort has been devoted to utilize the reaction medium as a strategic parameter, although the use of solvents is often crucial in synthetically useful transformations. The first choice for a solvent during the development of a synthetic procedure is usually an organic liquid, which is selected on the basis of its protic or aprotic nature, its polarity, and the temperature range in which the reaction is expected to proceed. Once the desired transformation is achieved, yield and selectivity are further optimized in the given medium by variation of temperature, concentration, and related process parameters. At the end of the reaction, the solvent must be removed quantitatively from the product using conventional workup techniques like aqueous extraction, distillation, or chromatography. If the synthetic procedure becomes part of a large-scale application, the solvent can sometimes be recycled, but at least parts of it will ultimately end up in the waste stream of the process. Increasing efforts to develop chemical processes with minimized ecological impact and to reduce the emission of potentially hazardous or toxic organic chemicals have stimulated a rapidly growing interest to provide alternatives to this classical approach of synthesis in solution. At the same time, researchers have started to realize that the design and utilization of multifunctional reaction media can add a new dimension to the development of synthetic chemistry. In particular, efficient protocols for phase separations and recovery of reagents and catalysts are urgently required to provide innovative flow schemes for environmentally benign processes or for high-throughput screening procedures. Fluorous liquid phases and supercritical carbon dioxide (sc CO2) have received particular attention among the various reaction media that are discussed as alternatives to classical organic solvents. The aim of this chapter is to compare these two media directly and to critically evaluate their potential for synthetic organic chemistry.

Free-Radical Chemistry in Supercritical Carbon Dioxide

During the 1990s, the chemical industry has focused on ways to reduce and prevent pollution caused by chemical synthesis and manufacturing. The goal of this approach is to modify existing reaction conditions and/or to develop new chemistries that do not require the use of toxic reagents or solvents, or that do not produce toxic by-products. The terms “environmentally benign synthesis and processing” and “green chemistry” have been coined to describe this approach where the environmental impact of a process is as important an issue as reaction yield, efficiency, or cost. Most chemical reactions require the use of a solvent that may serve several functions in a reaction: for example, ensuring homogeneity of the reactants, facilitating heat transfer, extraction of a product (or by-product), or product purification via chromatography. However, because the solvent is only indirectly involved in a reaction (i.e., it is not consumed), its disposal becomes an important issue. Thus, one obvious approach to “green chemistry” is to identify alternative solvents that are nontoxic and/or environmentally benign. Supercritical carbon dioxide (sc CO2) has been identified as a solvent that may be a viable alternative to solvents such as CCl4, benzene, and chloroflurocarbons (CFCs), which are either toxic or damaging to the environment. The critical state is achieved when a substance is taken above its critical temperature and pressure (Tc, Pc). Above this point on a phase diagram, the gas and liquid phases become indistinguishable. The physical properties of the supercritical state (e.g., density, viscosity, solubility parameter, etc.) are intermediate between those of a gas and a liquid, and vary considerably as a function of temperature and pressure. The interest in sc CO2 specifically is related to the fact that CO2 is nontoxic and naturally occurring. The critical parameters of CO2 are moderate (Tc = 31 °C, Pc = 74 bar), which means that the supercritical state can be achieved without a disproportionate expenditure of energy. For these two reasons, there is a great deal of interest in sc CO2 as a solvent for chemical reactions. This chapter reviews the literature pertaining to free-radical reactions in sc CO2 solvent.

Phase Behavior and Its Effects on Reactions in Liquid and Supercritical Carbon Dioxide

Carbon dioxide, either as an expanded liquid or as a supercritical fluid, may be a viable replacement for a variety of conventional organic solvents in reaction systems. Numerous studies have shown that many reactions can be conducted in liquid or supercritical CO2 (sc CO2) and, in some cases, rates and selectivities can be achieved that are greater than those possible in normal liquid- or gas-phase reactions (other chapters in this book; Noyori, 1999; Savage et al., 1995). Nonetheless, commercial exploitation of this technology has been limited. One factor that contributes to this reluctance is the extremely complex phase behavior that can be encountered with high-pressure multicomponent systems. Even for simple binary systems, one can observe multiple fluid phases, as shown in Figure 1.1. The figure shows the pressure–temperature (PT) projection of the phase diagram of a binary system, where the vapor pressure curve of the light component (e.g., CO2) is the solid line shown at temperatures below TB. It is terminated by its critical point, which is shown as a solid circle. The sublimation curve, melting curve, and vapor pressure curve of the pure component 2 (say, a reactant that is a solid at ambient conditions) are the solid lines shown at higher temperatures on the right side of the diagram; that is, the triple point of this compound is above TE. The solid might experience a significant melting point depression when exposed to CO2 pressure [the dashed–dotted solid/liquid/vapor (SLV) line, which terminates in an upper critical end point (UCEP)]. For instance, naphthalene melts at 60.1 °C under CO2 pressure (i.e., one might observe a three-phase solid/liquid/vapor system), even though the normal melting point is 80.5 °C (McHugh and Yogan, 1984). To complicate things even further, there will be a region close to the critical point of pure CO2 where one will observe three phases as well, as indicated by the dashed–dotted SLV line that terminates at the lower critical end point (LCEP). The dotted line connecting the critical point of the light component and the LCEP is a vapor/liquid critical point locus.

Enhancing the Properties of Portland Cements Using Supercritical Carbon Dioxide

Supercritical CO2 (sc CO2) is being used to accelerate the natural aging reactions (i.e., carbonation) of Portland cement. This treatment method alters the bulk properties of cement, producing profound changes in both structure and chemical composition. As a result of these changes, the mechanical and transport properties of these cements are also dramatically affected, and they display reduced porosity, permeability and pH, as well as increased density and compressive strength. Two areas of application for the sc CO2 treatment of portland cement have been undergoing investigation. Because the calcium carbonate (CaCO3) formed during the accelerated carbonation reaction is found to have excellent cementing properties, it is possible to replace a large fraction of the relatively expensive Portland cement with industrial waste products, such as fly ash and kiln dusts, which have inherently inferior cementing properties. These modified Portland cements, incorporating significant volume fractions of industrial wastes, can be used as low-cost building materials. The second area of application deals with the enhancement of Portland cements used to encapsulate waste products. Portland cement is used as an immobilization matrix for low- and intermediate-level radioactive waste by both the U.S. federal government (Huang et al., 1994) and civilian nuclear power companies in the United States (Wilk, 1997) and abroad (Wilding, 1992). Transportation issues relating to water content, radiolysis, and radionuclide content often preclude the ultimate disposal of these cemented wasteforms (U.S. DOE, 1996). However, the structural and chemical changes produced by accelerated carbonation have been shown to address these problems satisfactorily (Hartmann et al., 1999).

Rheological Properties of Polymers Modified with Carbon Dioxide

Use of supercritical fluids (SCFs), particularly supercritical carbon dioxide, as alternative solvents in polymer synthesis and processing is a rapidly growing research area with successful industrial applications (McCoy, 1999). In some cases, the need for alternative solvents is based on environmental concerns, with regulations mandating replacement solvents. An environmentally mandated example is the 1995 ban of the use of chlorofluorocarbons (CFCs) as physical blowing agents in the manufacture of polymeric foams after CFCs were classified as class-I-ozone-depleting substances (ODPs). Among the alternative blowing agents are gases like CO2 and N2 and refrigerants such as 1,1-difluoroethane (R152a) and 1,1,1,2-tetrafluoroethane (R134a). Under the foaming conditions, at temperatures above the glass transition temperature of a polymer, and at pressures required for flow of highly viscous polymer melts, these alternative blowing agents are frequently supercritical. When polymers are formed into final products by various melt-processing techniques, such as extrusion, injection molding, blow molding, foaming, and spin-coating, extremely high melt viscosity presents a major difficulty. A common method to moderate the processing conditions is to add a liquid solvent or plasticizer to the melt. Solvents and plasticizers lower the glass transition temperature, Tg, of the polymer so that the polymer can be made to flow at lower pressures and temperatures. Replacing liquid solvents with SCFs presents unique processing advantages. Higher diffusivity and lower viscosity of SCFs, compared with liquid solvents, increase rates of dissolution and mixing. The properties of polymer–SCF solutions are tunable via pressure or temperature changes, thus allowing efficient downstream separations. Most importantly, dissolution of an SCF produces very large reductions in melt viscosity compared with a liquid solvent dissolved in the melt. Whether the interest in using SCFs in polymer synthesis and processing is driven by environmental concerns or processing advantages, it is important to understand the rheological behavior of polymer–SCF mixtures. In this chapter, we describe rheological measurements of polymer melts containing dissolved gases for two polymers, polydimethylsiloxane (PDMS) swollen with CO2 at 50 °C and 80 °C and polystyrene (PS) swollen with CO2, R152a, and R134a at 150 °C and 175 °C.