2

Separation Science Today

This chapter provides a snapshot of the current state of separation science. The context for this analysis is the research themes identified as high priorities in the 1987 report Separation and Purification: Critical Needs and Opportunities (NRC, 1987) and listed in Chapter 1 of the present report (see Box 1-2). The 1987 committee envisioned that pursuit of the research themes would lead to clearer insights into fundamental principles and serve as the basis of a national program on separation research. Although many fundamental challenges that were identified in 1987 persist, there have been a number of dramatic successes in separation science over the last few decades, and the emergence of new materials and experimental and theoretical techniques has set the stage for important advances in separation science in the coming years. The current committee describes some of today’s research efforts and advances in the context of the six 1987 research themes but emphasizes that this chapter is not an exhaustive review of all the advances that have taken place since the 1987 report.

GENERATING IMPROVED SELECTIVITY AMONG SOLUTES IN SEPARATIONS

Improving the selectivity among solutes for all kinds of separations remains one of the important challenges of the field. Many exciting fundamental developments in material synthesis and separation processes have been made in recent decades to improve selectivity. Especially important has been the introduction of several new classes of materials—for example, new types of polymers, mixed-matrix membranes, new solid adsorbents, and ionic liquids—that have greatly extended the types of separations that can be envisioned. Several new materials are highlighted below.

Polymeric separation membranes first became available commercially in the 1970s and early 1980s. Since then, major developments have improved selectivity of polymeric membranes and contributed to their much wider use for separations. For example, researchers are exploring polymer cross-linking, pyrolysis of polymer membranes into carbon materials, and the addition of stiffer structures, such as zeolites and carbon-based structures (mixed-matrix membranes), to the polymers to address the limitations on selectivity that result from wide pore-size distribution and unwanted polymer or pore flexibility in unmodified polymer membranes. Other major developments in the use of polymers for membranes include morphological tuning, such as gradation of pore size in the axial direction of increasing or decreasing flow; the development of coatings, such as regenerated cellulose, to add structural integrity to nonpolar polymers; the incorporation of additives to create polar surfaces (Shannon et al., 2008); and the formation of isoporous membrane structures by using block co-polymers (Karunakaran et al., 2014; Nunes et al., 2010; Qiu et al., 2013). Innovative polymers that incorporate permanent porosity—such as polymers of intrinsic microporosity (PIMs)—have shown the ability to move beyond the fundamental tradeoffs between selectivity and throughput in membrane applications (Budd et al., 2008). Recently, new materials based on cross-linked brush polymers with or without additives that could be scalable and cost-effective and exhibit excellent performance have been proposed (Keating et al., 2016; Grimaldi et al., 2015). Improving selectivity with polymers will continue to be valuable in the new era of separations.

The diversity of zeolites that can be synthesized continues to expand (Guo et al., 2015), and rapid advances have occurred in making hierarchical and two-dimensional zeolites (Kim et al., 2018). The emergence of the mesoporous material most commonly known as MCM-41 (Mobil Composition of Matter No. 41) (Beck et al., 1992; Kresge et al., 1992; Xiu et al., 1996; Yin et al., 2012), metal-organic frameworks (MOFs) (Yaghi and Li, 1995; Yaghi et al., 1995), and the many variations of these mate-

rials now provide opportunities for highly selective separations. Databases catalog thousands of these materials (Chung et al., 2014).

Similar progress has taken place in the creation of tunable microporous and mesoporous materials that are disordered. Important progress has been made in controlling the properties of carbon molecular-sieve materials and has produced practically feasible devices from these highly robust materials (Koh et al., 2016). Carbon-based materials have also been produced in highly porous aerogel forms (Sun et al., 2013), and carbide-derived carbons have shown promise as robust porous materials (Presser et al. 2011). Work on graphene-based materials has hinted that these materials might open avenues for highly selective separations (Li et al., 2013). Important new developments in the dimensions and structure of silica particles for high-performance liquid chromatography include particles that are less than 2 micrometers (de Villiers et al., 2006), superficially porous particles with submicrometer shells (Wei et al., 2017), and wider-pore silica extending to pore diameters of 100 nm (Wagner et al., 2017).

Magnetic nanoparticles with an array of specific coatings are now commonly available for selective removal and concentration of compounds of interest. Selective coatings that use affinity ligands, molecularly imprinted polymers, amines, graphite, or titanium dioxide can be used to extract biomarkers from biological fluids and trace compounds from food and environmental samples. The combination of high surface area and selective sorption of the nanoparticles and their magnetic properties often provides high extraction efficiency of the compounds.

A complementary approach to selective coatings is to use the bioinspired molecular imprinting of polymer surfaces to emulate biological binding pockets with exquisite steric design and with multiple intermolecular interactions, such as electrostatic, hydrophobic, and van der Waals (Han et al., 2003). However, such approaches are often limited by their low capacity, although multimodal ligands (ones that use multiple interactions) are now being used in protein fractionation with chromatography (Robinson et al., 2018).

Ionic liquids (ILs) also offer opportunities to increase selectivity. ILs are salts that are composed of cations and anions that have melting points below 100°C (Wasserscheid and Welton, 2007; Ge et al., 2009). Unlike conventional organic solvents, ILs generally have negligible vapor pressure and can dissolve many organic and inorganic compounds. They are being studied for a variety of separation systems, including gas–liquid, gas–solid, liquid–liquid, and liquid–solid applications (Brennecke and Maginn, 2001; Cadena et al., 2004; Plechkova and Seddon, 2012, 2014, 2015; De Los Rios and Fernandez, 2014; Shiflett and Maginn, 2017). ILs can also be polymerized into membranes and used to coat porous and nonporous materials (Geurrero-Sanchez et al., 2005; Fehrmann et al., 2014). The field has developed rapidly over the last 2 decades, and commercial applications in separations are beginning to be announced. The National Institute of Standards and Technology (NIST) maintains a physical-property database (ILThermo) that contains several hundred ILs and thousands of data values¹ (Dong et al., 2007). It has been estimated that 10⁹ ionic liquids can be synthesized; and work on task-specific ILs and mixtures can provide new systems for highly selective separations (Shiflett and Scurto, 2017).

Incorporating principles of molecular-recognition chemistry and supramolecular chemistry into the design of separation materials has increased the effectiveness of separation processes already in use and allowed new ones to be developed (Schneider, 2012). New molecular platforms—such as calixarenes, cyclodextrins, and related macrocycles—can be tailored with multiple functionalities that enable the binding of almost any target ion (Lumetta et al., 2000). Made sufficiently hydrophobic, these materials have facilitated highly selective liquid–liquid extractions in solvent extraction or liquid-membrane configurations (Moyer, 2013). They can also be fixed to polymeric or inorganic supports to create new sorbent materials that have many applications (Izatt et al., 2015). Beginning with Nobel Prize-winning contributions of Lehn (1988), Cram (1988), and Pedersen (1978), many of the early examples of molecular recognition focused on cation binding, naturally drawing from classical coordination chemistry. The generality to anions and neutral molecules was soon apparent in the greater context of supramolecular chemistry (Lehn, 1995). Anion-coordination chemistry (Bowman-James and Werner, 2005) in particular appeared as a contrast to classical coordination concepts and have led to the ability to recognize and separate anions by using receptors that contain hydrogen-bond-donor and Lewis-acid groups (Moyer and Singh, 2004).

RECOVERING SUBSTANCES FROM DILUTE SOLUTIONS

Concentrating solutes from dilute solutions poses special challenges. Here, three examples in which substantial progress has been made—the recovery of actinides from dilute solutions, the use of preconcentration, and the removal of dilute salts from water—are highlighted.

The recovery of actinides from dilute solutions in complex and sometimes extreme environments has seen many successes (Clark et al., 2017). For example, amidoxime ligands grafted onto polymer backbones have

___________________

¹ Kazakov, A., Magee, J. W., Chirico, R. D., Paulechka, E., Diky, V., Muzny, C. D., Kroenlein, K., and Frenkel, M., NIST Standard Reference Database 147: NIST Ionic Liquids Database - (ILThermo), Version 2.0, National Institute of Standards and Technology, Gaithersburg, MD 20899. http://ilthermo.boulder.nist.gov.

been used in the selective uptake of uranium from seawater in which the ambient concentration of uranium is only 3 ppb (Abney et al., 2017). Effective new chelating extractants have been developed for the recovery of uranium at parts-per-million concentrations in phosphoric acid during the manufacture of fertilizer (Beltrami et al., 2014). The separation of minor actinides (especially the heat-emitter Am-241)—considered to be the key to full nuclear-fuel recycling—has become possible through intense effort in multiple countries. The need to ensure that the lanthanides, which are abundant fission products in used nuclear fuel and potent neutron poisons in nuclear reactors, are excluded from recycled actinides led to the fundamental question of how to separate trivalent lanthanides and actinides that have similar ionic radii and solution properties (Hill, 2010). Relying heavily on solvent extraction, which is amenable to remote handling in shielded facilities, researchers have developed new generations of extractants, such as the bis(triazinyl) pyridines and phenanthrolines, which exhibit strong selectivity for Am(III) over trivalent lanthanides. Drawing heavily from progress in developing new extractants for nuclear fuel-cycle separations, radioanalytical chemists have successfully tackled such problems as the determination of trace actinides in biological and environmental media and have developed separation techniques, such as extraction chromatography (Kahn, 2007).

In analytical chemistry settings, preconcentration can be a powerful general strategy for the detection and characterization of trace components. Among analytical scale preconcentration methods, an important advance has been the commercialization and wide use of solid-phase microextraction (SPME). Advantages of SPME include short extraction times, small sample requirements, ability to use it in portable devices, and compatibility with most chromatographic instruments. The technology has diverse applications, including trace analysis of herbicides or pesticides in natural waters, air analysis, and preconcentration of biological fluids for disease detection and diagnosis (Reyes-Garcés et al., 2018). SPME is not perfect: the coatings used in SPME can be stripped from the supporting fiber, and the coatings will swell in some solvents. Thus, there is much room for enhancements of this technology.

Removal of salts from brackish water is more difficult than desalination of seawater, despite the lower salt concentration (0.2–0.5 wt% vs 3–5 wt%), which means that membrane selectivity of 95–98% NaCl rejection is adequate, and transmembrane pressures of no more than about 30 bar are required. The reasons for the greater difficulty are that the feedwater-recovery requirements are almost always higher (that is, there is not an unlimited supply of inland brackish water, as is the case for seawater), brine disposal poses a problem, and membrane fouling is a serious issue. Fouling is more problematic because of the greater complexity of the ions present (for example, high percentages of Ca²⁺, Mg²⁺, HCO₃^-, and SO₄²); the higher feedwater-recovery requirements, which result in precipitation of salts; and the presence of organics. However, there has been substantial progress in addressing those challenges. In 2015–2017, feedwater was desalinated at 9,500,000 m^3/d, enough for more than 25,000,000 people.² About 22% of that was desalinated brackish or other inland waters.³ The development of thin-film composite membranes means that drinking water can be produced from groundwater aquifers that have salt concentrations of 0.05 wt% or lower. Nonetheless, traditional cellulose acetate membranes continue to be used for some brackish water desalination, even though composite membranes might have better performance (Baker, 2012). Challenges that still need to be addressed include increased permeation fluxes, disposal of concentrated brine, and better prediction of selectivity for dissolved organics in brackish water.

UNDERSTANDING AND CONTROLLING INTERFACIAL PHENOMENA

Effective implementation of many separation techniques requires fine control of interfacial phenomena. As defined in Chapter 1, the committee uses the term interface to mean the area of contact between a bulk phase and the interfacial region. The interfacial region is the domain between the two bulk phases; the structure and extent of the interfacial region are directed by the molecular interaction between the bulk phases (see Figure 2-1). Interfaces are present between any two bulk phases; common examples encountered in separations are solid–liquid, solid–vapor or solid–gas, and liquid–liquid interfaces.

In general, interfaces and interfacial regions in separation systems are poorly understood and poorly characterized. Although it is well known that interfaces can affect mass transfer and partitioning, there are no design criteria for modifying them to provide the desired outcome. However, there have been advances in techniques to mitigate the negative impacts of interfaces (such as fouling of membranes) or adverse effects of nonhomogeneous interfaces (such as carbon stationary phases). Some of the advances are described below.

Several advances in materials synthesis and in interfacial characterization have contributed to the current status of this topic. One advance has been the development of molecular criteria for designing low protein-adhering surfaces. Some of these are hydrophilic, some have hydrogen-bond acceptors, some do not have hydrogen-bond donors, and some are net electrically neutral (Holmlin et al., 2001; Ostuni et al., 2001). Examples of interfaces that

___________________

² See https://water.usgs.gov/edu/qa-home-percapita.html.

³ See https://www.desalination.com/publications/catalogue/idahandbook.

meet the criteria include self-assembled monolayers of zwitterions, of oligo(ethylene glycol), of tertiary amines, and of mixtures of positively charged and negatively charged moieties (Mi et al., 2010; Gu et al., 2013).

A dramatic example of the use of interfacial control to improve separation processes is the development of fouling-resistant membranes for water purification and protein-resistant membranes for bioseparations. Use of highly controlled polymerization, such as controlled radical polymerizations (Lligadas et al., 2017), has allowed superb control in surface modification of separation materials and polymeric membranes (Husson and Bhut, 2015) and thus has enabled that development. Polymerization methods include atom transfer radical polymerization (Matyjaszewski and Xia, 2001; Matyjaszewski et al., 2007), stable free radical polymerization (such as activators regenerated by electron transfer), and reversible addition-fragmentation transfer.

A second striking example of how fundamental advances in interfacial phenomena have improved separation performance is the use of carbon materials as chromatographic media. One commercially available carbon-stationary phase provides a flat homogeneous surface with primarily basal-plane carbons presented on the surface with few defects. The high level of homogeneity results in reproducible selectivity, and the flat surface allows ready separation of isomers (West et al., 2010).

An understanding of interfacial phenomena is critical for creating composite materials, materials that provide a powerful general strategy for overcoming limitations that are inherent in individual classes of materials. An example of such composite materials is mixed-matrix membranes, which incorporate additives in polymer films to enhance membrane performance. Making effective composites requires precise control of the interfaces between materials. Specifically, a critical aspect of inserting additives (for example, zeolites, MCM-41, MOFs, and PIMS) into polymers to improve performance involves detailed control of the interactions between the polymers and the added particles.

Interfacial properties are also critical in unconventional settings. Formation of reverse micelles often occurs in solvent extraction because the amphiphilic ligands that are used as extraction reagents almost invariably aggregate as their concentration increases or as aggregation-inducing species are extracted into the solvent (Nilsson et al., 2013). Because the reverse micelles contain a core of water molecules, they introduce an interface in solvent extraction. Application of such techniques as dynamic light scattering and small-angle x-ray and neutron scattering has accelerated understanding of the solution structure and dynamics. One of the most important developments relates the properties of the hydrophobic rim to micelle–micelle interactions that culminate in third-phase formation and thereby provide fresh insight into the problem of third-phase formation, which has remained a vexing mystery. A lingering question concerns the inner structure of reverse micelles and how to control and exploit it.

A key challenge in understanding interfacial phenomena lies in characterizing interfaces. Although solids that have high surface area are explicitly designed to have large amounts of interfacial region, the effective volume associated with interfaces in many situations is quite small relative to the bulk phases. However, sever-

al interface-sensitive techniques have become available in recent decades, including sum-frequency generation, atomic-force microscopy, surface-plasmon resonance, quartz-crystal microbalance with dissipation, and x-ray and neutron scattering techniques (see Chapter 3 for more detail on these techniques). Those and related techniques can allow superb insight into interfacial properties, but characterization under physical conditions that are truly representative of separation processes remains challenging. That challenge has led to the growth of modeling and simulation of complex liquid interfaces, and this growth has allowed features of interfacial structure and dynamics to be correlated with extraction and separation efficacy.

INCREASING RATE AND CAPACITY IN SEPARATIONS

The throughput of a separation is the rate at which a given flow or input stream is processed, and it is a critical measure that determines whether a separation concept can advance to practical applications. For example, membranes run continuously, and the flux of the species (the rate at which it passes) through the membrane mainly determines the net rate of separation and thus the overall productivity of the system. In batch operations, such as chromatography, the capacity of the column for the analytes might be the main determinant of throughput and therefore a more useful measure of throughput. Nonetheless, the overall cycle processing time (the time it takes to process all the species) is the focus, so both throughput and capacity are important.

Throughput has been a driver of the development of new materials for membranes. For example, there has been considerable effort in developing inorganic membranes made from such materials as thin zeolite or carbon films partly because of the possibility of high throughput per unit area relative to polymeric membranes (Lai et al., 2003). The same motivation has guided much work in the use of MOFs and similar materials in membranes. However, efforts to increase membrane flux point to an important caveat: research efforts might be misplaced if they focus on throughput in an idealized setting without considering the overall set of factors that can control throughput in an operational setting. For example, in the use of membranes for water desalination, concentration polarization is likely to influence throughput in operation strongly, so increasing the intrinsic flux possible through the active membrane material might not be the optimal strategy (Werber et al., 2016).

In separation processes that use membranes, fouling has often decreased throughput. For example, early uses of synthetic-membrane filtration for biotechnology, food and beverage production (such as cheese, beer, and wine production), and especially protein production and purification were challenging because of fouling. Membrane fouling leads to a rapid decline in performance, such as decreased permeation rates and adverse changes in selectivity. Moving from nonpolar to polar membrane surfaces and improvements in mass transfer and fluid mechanics began to address that problem directly in some applications (Rana and Matsuura, 2010). Membrane modules were redesigned with short path lengths and improved transport characteristics. Many membranes today are

coated with regenerated cellulose to reduce fouling and prevent target proteins from denaturing (Castilho et al., 2002). Addressing the problem of fouling and the associated decrease in throughput has allowed membrane technology to become widely used in a variety of applications, such as in the production of monoclonal antibodies (major protein products in the biotechnology industry), in the use of filters for buffer, and in conjunction with chromatography for product purification and concentration (see Figure 2 2).

In addition to new or improved separation materials, innovations in separation processes can increase throughput. In cyclic processes, such as pressure swing adsorption, structured contactors can allow cycle times that are far shorter or driving forces that are much smaller than are possible with more traditional packed beds (Lively et al., 2009).

Increases in rate or capacity can also enable effective handling of more complex samples. Ultraperformance liquid chromatography, which was commercialized in 2004, uses particles that are typically less than 2 µm in diameter and applied column inlet pressures as high as 15,000 psi. The technique provides ultrafast separations—in minutes and sometimes in less than a second—with high efficiency. When the method is combined with superficially porous packing materials, further improvement in separation speed and efficiency is observed, typically by a factor of 8 to 10 compared with 5-µm particles (Hayes et al. 2014). Although the superficially porous particles were introduced briefly in the 1960s for analytical separations, later reductions in particle size led to dramatically improved efficiency, which enables the separation of more complex materials.

Other examples of high-throughput systems that can separate complex mixtures are multidimensional analytical separation systems, such as comprehensive gas chromatography (GC×GC) and comprehensive liquid chromatography (LC×LC). GC×GC includes an initial separation in one column and then fast transfer to a second column with different selectivity. It is used in petroleum characterization (Pollo et al., 2018), odorant separations (Cordero et al., 2018), and metabolomics (Prebihalo et al., 2018). Hundreds or even thousands of compounds can be separated with this method, and better separation can be achieved with a multidimensional approach for very complex volatile or semivolatile mixtures. Comprehensive LC×LC is also showing increasing value, especially in the genomics, proteomics and metabolomics fields in which complex mixtures are the norm. It can separate and help to identify hundreds or thousands of compounds and has substantially advanced proteomic and metabolomic analysis and pharmaceutical analysis (Cacciola et al., 2017). However, there is considerable room for further advances in column selectivity and instrumentation for comprehensive chromatographic methods.

DEVELOPING IMPROVED PROCESS CONFIGURATIONS FOR SEPARATION EQUIPMENT

Although process development is outside the scope of the present study, critical links between specific separation technologies and the design and development of process equipment that enables the use of new technologies are necessary, especially for large-scale processes (Rajagopalan et al., 2016; Estupinan Perez et al., 2016). Some of the major advances have been in a field now called process intensification, which focuses on smaller, cleaner, safer, and more energy-efficient technology (Stankiewicz and Moulijn, 2000).

One example is reactive distillation, in which chemical reactions take place in the distillation column. Reactive distillation is now common in fuel-ether and chemical-ester production in which the reactions depend on equilibrium conditions (Orjuela et al., 2016). Catalytic distillation, in which the heterogeneous catalyst is packed in the distillation column, is being used for some hydrogenation processes to recover products efficiently and selectively. A second example is divided-wall columns, which effectively combine multiple distillation columns into a single column. This can result in the isolation of multiple product streams from one unit (Weinfeld et al., 2018).

A third example is the adoption of hybrid separation systems. As an alternative to complex distillation schemes, a combination of older and newer technologies can be hybridized to lead to cost-effective upgrades to improve manufacturing process rates and capacities. An example is the combination of novel membrane separation with conventional distillation for the separation of olefins and paraffin (Xu et al. 2012). Membrane separations are low-energy processes but generally do not produce high purity products, whereas distillation is energy intensive but gives high purity products. The hybrid process can produce high purity products but with lower overall energy intensity.

Conventional chemical separation processes are being replaced by newer nonthermal methods slowly, largely because of the enormous capital costs associated with world-scale chemical facilities. An analysis of recently announced projects in the chemical industry associated with the feedstock shift in the United States from petroleum to shale-gas liquids demonstrates the cost and scale of changing a process (Zhang and El-Halwagi, 2017). For a new process with a capacity of 1 million tons/year, the average estimated capital investment (2016 basis) is $125–150 million for a single functional unit, such as a separation process. Implementing a new separation technology could enable a separation that is not achievable today or could substantially improve a separation relative to conventional choices. However, if the new technology

is not extremely well understood, the risk is too great because the tangible risk associated with knowledge gaps can far outweigh the potential benefits for decision-makers in a risk–benefit comparison.

The requirements that can hinder decisions to implement new processes—whether they are entirely new technologies or hybrid systems—have important implications for the fundamental research that is the primary topic of the present report. For a material to be used in a new separation process, it must have robust and reproducible methods of synthesis and characteristics that can be measured to ensure its quality. For a separation process to be adopted, data must clearly demonstrate its benefit under industrial conditions, and the long-term behavior or degradation of the separation material must be understood. High-quality information that allows reliable process design and optimization also needs to be available. Although it is not necessarily the role of fundamental research to satisfy all those requirements, if any requirement is not met, the adoption of a newly discovered material or separation technology will be slow at best.

Despite the many requirements outlined above, there are examples of the commercial implementation of new nonthermal separations. Whey processing is an excellent example of how new equipment dramatically changed a separation process and transformed a waste byproduct into a highly valued product. During cheese manufacturing, milk proteins are precipitated and sent to a cheese production plant; the supernatant liquor (whey), which contains high concentrations of salt and lactose, was traditionally disposed of in rivers, fields, and sewage systems or sold as animal feed (Smithers, 2008). However, those disposal practices are now prohibited in the United States, so manufacturers have fewer free or money-making options. Recently, the separation of proteins (α-lactalbumin and β-lactoglobulin), lactose, and salts from whey has become possible because of advances in membrane fractionation. The advances have allowed the transformation and fractionation of the whey byproduct into useful and valuable products (Kumar et al., 2013). For example, nanofiltration has been used to demineralize whey to produce powders used in baby and clinical formulas (Smithers, 2008); microfiltration and ultrafiltration membranes have been used to produce high-protein, low-fat whey protein concentrates, which contain 35–80% protein, and isolates, which contain 85–90% protein; and reverse osmosis has been used to recover total solids and water. More recently, ion-exchange membranes have been used to isolate and fractionate whey proteins. Given the strict disposal regulations, the dairy industry was able to profit from transforming or purifying whey into useful and economically valuable products instead of simply disposing of it.

IMPROVING ENERGY EFFICIENCY

Many separations are still being performed by using distillation, which is energy-intensive. New facilities can be built to use newer technology; but at a time when energy is viewed as inexpensive, older facilities with older technology cannot justify spending the money to replace old systems. However, older facilities can integrate new separation technologies with the old, both to gain economic benefits and to become more sustainable. These ideas are explored below.

Industry uses vast amounts of energy to achieve chemical separations. It is convenient to think of energy consumption in the United States in terms of quadrillions of BTUs given that 97.7 quadrillion BTUs of primary energy were consumed in the United States in 2017.⁴ That is equivalent to 28,633 terawatt-hours (TWh) and a consumption rate of 3.27 TW. U.S. manufacturing currently uses about 19 quadrillion BTUs (of the 97.7 quadrillion), of which 15 quadrillion are used to convert feedstocks into high-value products (EIA, 2014, October 2017 revision).

The chemical industry accounts for 23.5% (3.53 quadrillion BTUs) of U.S. manufacturing fuel consumption. It is difficult to determine the fraction of the 3.53 quadrillion BTUs that is used specifically for chemical separations because chemical manufacturing facilities are highly integrated, including good heat integration. However, the predominance of distillation suggests that a substantial amount of the energy used in chemical manufacturing is associated with separation technologies. For example, a recent study by the U.S. Department of Energy (DOE, 2015) highlighted 24 chemicals that are manufactured in the United States, separations related to which consume 40% of the manufacturing energy. Thus, one can safely conclude that industrial chemical separations in the United States consume well over a quadrillion BTUs of energy each year.

The large energy consumption of separations in U.S. chemical manufacturing makes clear that better energy use in chemical separations offers an opportunity for tremendous energy savings. An analysis of that opportunity can be found in the aforementioned study conducted by the DOE Advanced Manufacturing Office (DOE, 2015). It concluded that the development and industrial implementation of an array of new technologies (including those chemical separations) could lead to an energy savings of 0.950 quadrillion BTUs per year by the U.S. chemical industry.

Energy efficiency in U.S. chemical manufacturing has improved recently. For example, the energy conversion efficiency (process energy consumption divided by

___________________

⁴ See https://www.eia.gov/energyexplained/?page=us_energy_home.

the inherent energy content of the feedstock) decreased from about 1.6 (EIA, 2010, February 2014 revision) to about 1.4 (EIA, 2014, revised October 2017). Much of the improvement has two sources. First, capital expenditures to implement state-of-the-art technologies in U.S. manufacturing facilities have improved the energy intensity of some chemicals, particularly light olefins.⁵ Second, considerable maintenance expenditure (for example, for installation of insulation to reduce operating costs) during a time when energy was relatively expensive has substantially reduced energy losses to the environment. Those energy-efficiency improvements have been accomplished without implementing advanced separation techniques.

The widespread adoption of reverse-osmosis membranes for desalination, which was highlighted in Chapter 1, is a good example of how large reductions in energy use can drive changes in the separation technology that is installed in new facilities. Technology advances resulted in the reduction in the power consumption of the reverse-osmosis stage in industrial seawater desalination plants from over 15 kWh/m³ of clean water in 1970 to less than 2 kWh/m³ in 2008 (see Figure 2-3), making reverse osmosis a clear choice over evaporation.

In general, reduction in energy use of separation units is not sufficient to drive industrial adoption of new separation technology. That is because, unlike seawater desalination plants, most separation units are not standalone units but rather parts of larger chemical manufacturing facilities. Heat integration and similar strategies have been carefully optimized in operating facilities, so new separation strategies can be evaluated in the context of highly integrated plants and in an environment that involves massive sunk costs—expenditures that have been incurred and cannot be recovered.

In addition, the revolution associated with shale gas and related hydrocarbon resources in the United States has profoundly shaped developments in the commodity-chemical industry. The new resources have led to a renaissance in chemical manufacturing, including the construction of many new world-scale chemical facilities in the United States. For ethylene manufacturing alone, the availability of lower-cost natural-gas liquids and liquefied petroleum gas has led to additional capacities of 11 million tons/year (2017–2022) in the United States.⁶ A less-appreciated aspect of this change is a perception that energy is “cheap”, which reduces the impetus to use energy more efficiently than at other times in the last 30 years.

Sustainability in chemical manufacturing is a more compelling reason to implement advanced separation techniques than is the cost reduction due to lower energy use, given the complexities outlined above. Because a large amount (77.6% in 2017) of primary energy in the United States still comes from fossil-fuel resources,⁷ any reduction in energy use corresponds to a reduction in greenhouse-gas emissions. Furthermore, advanced separation technologies can be used in ways beyond simply replacing existing distillation operations in the chemical industry to enhance both economic and environmental sustainability (see Figure 2-4).

___________________

⁵ See https://www.hydrocarbonprocessing.com/magazine/2010.

⁶ See http://analysis.petchem-update.com/engineering-and-construction/insights-first-wave-us-ethylene-projects-drive-secondwave-decisions.

⁷ See https://www.eia.gov/energyexplained/?page=us_energy_home.

Another way that improved separation technologies contribute to sustainability is in the downsizing of technologies for analytical separations, which has resulted in large reductions in solvent use. High-performance liquid chromatography (HPLC) has become ubiquitous as an analytical laboratory technique in many industries (with chemical, pharmaceutical, food, and medical applications). Many tens of thousands of HPLC systems are in operation globally, and they produce tens of millions of liters of hazardous solvents each year. Although those amounts from the analytical laboratories in support of manufacturing facilities are small in comparison with the throughput of the large commercial chemical manufacturing facilities, hazardous waste is a large burden for analytical laboratories in waste-disposal costs and potential exposure of personnel. If HPLC instruments were replaced by ultra HPLC separation technology, the same high-quality separations could be achieved in a similar timeframe with reductions in solvent use and waste generation of more than 60%.

ADVANCES NOT ANTICIPATED BY THE 1987 REPORT

Scientific and technological advances not directly anticipated by the 1987 report have occurred. The committee highlights here several key advances poised to influence the continued development of chemical separations.

Nanofiltration membranes, often called loose reverse-osmosis membranes, retain molecules that have a molecular mass of 200–1,500 Daltons. They are used to remove divalent ions, organics, colors, sugars, amino acids, bacteria, and viruses. They are about 2–5 times as permeable and operate at much lower transmembrane pressures (2–20 bar) than seawater reverse-osmosis membranes (60 bar). They are widely used in the western United States to soften potable water (removing such divalent ions as Ca²⁺ and Mg²⁺), to desalinate water, and to remove trace amounts of agricultural contaminants (Baker, 2012). Fixed positive and negative charges on nanofiltration membranes allow Donnan exclusion to influence the separation of charged species, and neutral membranes reject solutes according to molecular size and nonelectrostatic thermodynamic interactions, such as dielectric exclusion. Further research is needed to understand the complex connection between size and charge selectivity.

The widespread implementation of additive manufacturing methods, such as three-dimensional printing, has created many opportunities for chemical separations (Wang et al., 2014; Thakkar et al., 2016; Low et al., 2017). The types of materials that can be processed with additive techniques are expanding rapidly, and this provides opportunities to develop structured devices for separations (Luelf et al., 2018; Zhang et al., 2018).

An important change in the research landscape since 1987 has been the proliferation of powerful modeling tools at various scales. Quantum-chemistry methods are now widely available and capable of giving detailed insight on atomic scales. Molecular modeling has similarly made enormous advances. It is now possible to use these methods to simulate libraries of thousands of materials or to simulate complex phases and interfaces that involve millions of atoms (Holian et al., 1998; Patil et al., 2018). The field of multiscale modeling and continuum modeling is now powerful and broad and can tackle challenging phenomena, such as membrane fouling, that cannot be ad-

dressed with atomically detailed methods. Developments in data-science methods have also had a major influence on the kinds of computational models that can be considered for physical systems (Ferguson, 2018). As in many fields of the physical sciences, data-science techniques are likely to complement many aspects of modeling and design approaches used in chemical separations in the future. This topic is revisited in Chapter 3.

Many major user facilities that can determine material structure and properties have been opened and made available to researchers in the last 40 years. The isolated opportunities to use synchrotron-generated, variable-energy x-rays for the characterization of material structure and properties available at the time of the 1987 report rapidly gave way to a proliferation of dedicated user facilities (see Table 2-1). The powerful instrumentation available through these user facilities provide opportunities—particularly in the characterization of solutions, liquids, amorphous and hierarchical materials—that are still underused by the separations community.

Following a similar trend, there have been monumental advances in computational capabilities. Computations that once required large computers can now be accomplished on hand-held or desktop devices. A wide array of chemistry and chemical-engineering programs are available for use on small portable machines. The growth in computational power, developed through advances in parallel processing and in theory and modeling, now provides predictive powers that can begin to address some of the longstanding recalcitrant problems imposed by complex separation scenarios. As it does with large-scale experimental facilities, DOE provides computational resources for general users who wish to have access to these powerful capabilities (see Table 2-2). The National Science Foundation (NSF) also provides computational resources as noted in Table 2-2.

Managed primarily by DOE, the synchrotron, neutron, and computing facilities are available to researchers through general user programs; time is awarded through proposal-based systems. Each facility uses beamline or computational scientists expert in machine capabilities, whose jobs include helping general users with all aspects of their proposed projects.

Underpinning the experimental and computational advances have been the invention and proliferation of the Internet, which allows data-sharing and information access not even imagined at the time of the 1987 report. The facilities and other important experimental techniques developed since 1987 are described in Chapter 3.

The idea of determining all the nucleotide base pairs that make up the human genome was hatched in the middle 1980s, and the Human Genome Project was launched in 1990. The sequencing of the human genome was accomplished in an amazingly short time: results were published in 2001 and finalized in 2003. That is an example of how separation technology was critical for the achievement of an ambitious national scientific goal. Multicapillary electrophoresis was commercialized during that period, and it dramatically decreased the time needed for analysis and sequencing of the human genome.

EDUCATIONAL, WORKFORCE, AND INDUSTRY NEEDS

As noted in Chapter 1, the 1987 report provided several recommendations on the development of separation curricula in chemical engineering, chemistry, and biochemistry. The 1987 committee recommended that basic

TABLE 2-1 Major Synchrotron and Neutron DOE Scientific User Facilities in Operation in the United States

Abbreviations: ANL, Argonne National Laboratory; BNL, Brookhaven National Laboraotry; LBNL, Lawrence Berkeley National Laboratory; NIST, National Institute of Standards and Technology; ORNL, Oak Ridge National Laboratory; and SLAC, Stanford Linear Accelerator Center.

concepts taught in those curricula (and in specific courses) be tied directly to concepts in separation. Interdisciplinary approaches to education and training were proposed to broaden education and training in separation science and technology. In response to the need to develop a trained workforce, the 1987 report stated that courses in separations should be included in all undergraduate training in chemical engineering and chemistry and that federal research funding should be distributed among many graduate programs rather than concentrated in only a few.

University education in separations continues to pose challenges. An American Institute of Chemical Engineers (AIChE) survey of chemical engineering and review of course syllabi in top analytical chemistry departments in the United States show little progress toward meeting the recommendations of the 1987 report. Furthermore, the use of continuing-education techniques, such as workshops and conferences, has not been supported or implemented. Chapter 6 provides further discussion of this important topic.

CONCLUSION

This chapter has examined the progress in the field of chemical separations since the 1987 report. The methodological and technical advances that have occurred have considerably expanded the complexity of chemical separations that can be imagined, and this positive trajectory will continue. Further progress can be made if advances from intersecting disciplines, as described in Chapter 3, are applied. Scientific advances in computational modeling, synthesis methods, and materials characterization have strongly influenced the field since 1987, but many of the overarching challenges articulated in 1987 remain. The remaining challenges are explored in Chapter 4.

TABLE 2-2 DOE and NSF Computing Research Facilities

¹DOE Facility.

²NSF Facility.

Abbreviations: ANL, Argonne National Laboratory; LBNL, Lawrence Berkeley National Laboratory; ORNL, Oak Ridge National Laboratory; and SLAC, Stanford Linear Accelerator Center.

REFERENCES

Cacciola, F., P. Donato, D. Sciarrone, P. Dugo, and L. Mondello. 2017. Comprehensive liquid chromatography and other liquid-based comprehensive techniques coupled to mass spectrometry in food analysis. Analytical Chemistry 89(1):414–429. doi: 10.1021/acs. analchem.6b04370.

Cadena, C., J. L. Anthony, J. K. Shah, T. I. Morrow, J. F. Brennecke, and E. J. Maginn. 2004. Why is CO₂ so soluble in imidazolium-based ionic liquids? Journal of the American Chemical Society 126(16):5300–5308. doi: 10.1021/ja039615x.

Castilho, L. R., F. B. Anspach, and W. D. Deckwer. 2002. Comparison of affinity membranes for the purification of immunoglobulins. Journal of Membrane Science 207(2):253–264. doi: 10.1016/S0376-7388(02)00257-0.

Chung, Y. G., J. Camp, M. Haranczyk, B. J. Sikora, W. Bury, V. Krungleviciute, T. Yildirim, O. K. Farha, D. S. Sholl, and R. Q. Snurr. 2014. Computation-ready, experimental metal-organic frameworks: A tool to enable high-throughput screening of nanoporous crystals. Chemistry of Materials 26(21):6185–6192. doi: 10.1021/cm502594j.

Clark, A. E., P. Yang, and J. C. Shafer. 2017. Coordination of actinides and the chemistry behind solvent extraction. Experimental and Theoretical Approaches to Actinide Chemistry: From Fundamental Systems to Practical Applications. Hoboken, NJ: John Wiley & Sons, 237–282.

Cordero, C., J. Kiefl, S. E. Reichenbach, and C. Bicchi. 2018. Characterization of odorant patterns by comprehensive two-dimensional gas chromatography: A challenge in omic studies. TrAC Trends in Analytical Chemistry (in press). doi: 10.1016/j.trac.2018.06.005.

Cram, D. J. 1988. The Design of Molecular Hosts, Guests, and Their Complexes. Journal of Inclusion Phenomena 6:397-413. doi: 10.1007/BF00658982

De Los Ríos, A. P., and F. J. H. Fernández. 2014. Ionic Liquids in Separation Technology. Amsterdam, The Netherlands: Elsevier.

de Villiers, A., F. Lestremau, R. Szucs, S. Gelebart, F. David, and P. Sandra. 2006. Evaluation of ultra performance liquid chromatography - Part I. Possibilities and limitations. J. Chromatogr. A 1127(1-2):60-69. https://doi.org/10.1016/j.chroma.2006.05.071.

DOE (U.S. Department of Energy). 2015. Bandwidth Study on Energy Use and Potential Energy Use Saving Opportunities in U.S. Chemical Manufacturing. Washington, DC: U.S. Department of Energy.

Dong, Q., C. D. Muzny, A. Kazakov, V. Diky, J. W. Magee, J. A. Widegren, R. D. Chirico, K. N. Marsh, and M. Frenkel. 2007. ILThermo: A free-access web database for thermodynamic properties of ionic liquids. Journal of Chemical and Engineering Data 52(4):1151-1159. doi: 10.1021/je700171f.

EIA (U.S. Energy Information Administration). 2010. 2010 Manufacturing Energy Consumption Survey (MECS) Data. https://www.eia.gov/consumption/manufacturing/ data/2010/.

EIA (U.S. Energy Information Administration). 2014. 2014 Manufacturing Energy Consumption Survey (MECS) Data. https://www.eia.gov/consumption/manufacturing/data/2014/index.php?view=data.

Elimelech, M., and W. A. Phillip. 2011. The future of seawater desalination: Energy, technology, and the environment. Science 333(6043):712–717. doi: 10.1126/sci-ence.1200488.

Estupinan-Perez, L., A. Rajendran, A. Avila, J. Sawada, S. Kuznicki. 2016. Process optimization-based adsorbent selection for ethane recovery from residue gas. Separation and Purification Technology 168:19-31. doi: 10.1016/j.seppur.2016.05.010.

Fehrmann, R., A. Riisager, and M. Haumann. 2014. Supported Ionic Liquids: Fundamentals and Applications. Hoboken, NJ: John Wiley & Sons.

Ferguson, A. L. 2018. Machine learning and data science in soft materials engineering. Journal of Physics Condensed Matter 30(4). doi: 10.1088/1361-648X/aa98bd.

Ge, R., C. Hardacre, J. Jacquemin, and D. W. Rooney. 2009. Thermophysical properties of ionic liquids. Ionic Liquids: From Knowledge to Application. Washington, DC: ACS Publications, 43–60.

Grimaldi, J., J. Imbrogno, J. Kilduff, and G. Belfort. 2015. New class of synthetic membranes: Organophilic pervaporation brushes for organics recovery. Chemistry of Materials 27(11):4142–4148. doi: 10.1021/acs. chemmater.5b01326.

Gu, M., A. J. Vegas, D. G. Anderson, R. S. Langer, J. E. Kilduff, and G. Belfort. 2013. Combinatorial synthesis with high throughput discovery of protein-resistant membrane surfaces. Biomaterials 34(26):6133–6138. doi: 10.1016/j.biomaterials.2013.04.051.

Guerrero-Sanchez, C., F. Wiesbrock, and U. S. Schubert. 2005. Polymer synthesis in ionic liquids: Free radical polymerization in water-soluble systems. Ionic Liquids in Polymer Systems. Washington, DC: ACS Publications, 37–49.

Guo, P., J. Shin, A. G. Greenaway, J. G. Min, J. Su, H. J. Choi, L. Liu, P. A. Cox, S. B. Hong, P. A. Wright, and X. Zou. 2015. A zeolite family with expanding structural complexity and embedded isoreticular structures. Nature 524(7563):74–78. doi: 10.1038/nature14575.

Han, M., R. Kane, M. Goto, and G. Belfort. 2003. Discriminate surface molecular recognition sites on a microporous substrate: A new approach. Macromolecules 36(12):4472–4477. doi: 10.1021/ma021736e.

Hayes, R., A. Ahmed, T. Edge, and H. Zhang. 2014. Core-shell particles: Preparation, fundamentals and applications in high performance liquid chromatography. Journal of Chromatography A 1357:36–52. doi: 10.1016/j.chroma.2014.05.010.

Hill, C. 2010. Overview of recent advances in An(III)/Ln(III) separation by solvent extraction. In Ion Exchange Solvent Extraction. Boca Raton, FL: CRC Press.

Holian, B. L., and P. S. Lomdahl. 1998. Plasticity induced by shock waves in nonequilibrium molecular-dynamics simulations. Science 280(5372):2085–2088. doi: 10.1126/science.280.5372.2085.

Holmlin, R. E., X. Chen, R. G. Chapman, S. Takayama, and G. M. Whitesides. 2001. Zwitterionic SAMs that resist nonspecific adsorption of protein from aqueous buffer. Langmuir 17(9):2841–2850.

Husson, S. M., and B. V. Bhut. 2015. Membrane surface modification. Clemson Patents 495. https://tigerprints.clemson.edu/clemson_patents/495.

Izatt, R. M., S. R. Izatt, N. E. Izatt, K. E. Krakowiak, R. L. Bruening, and L. Navarro. 2015. Industrial applications of molecular recognition technology to separations of platinum group metals and selective removal of metal impurities from process streams. Green Chemistry 17(4):2236–2245. doi: 10.1039/c4gc02188f.

Kahn, B. 2007. Radioanalytical Chemistry. New York: Springer-Verlag.

Karunakaran, M., S. P. Nunes, X. Qiu, H. Yu, and K. V. Peinemann. 2014. Isoporous PS-b-PEO ultrafiltration membranes via self-assembly and water-induced phase separation. Journal of Membrane Science 453:471–477. doi: 10.1016/j.memsci.2013.11.015.

Keating, J. J., J. Imbrogno, and G. Belfort. 2016. Polymer brushes for membrane separations: A review. ACS Applied Materials and Interfaces 8(42):28383–28399. doi: 10.1021/acsami.6b09068.

Kim, D., M. Shete, and M. Tsapatsis. 2018. Large-grain, oriented, and thin zeolite MFI films from directly synthesized nanosheet coatings. Chemistry of Materials 30(10):3545–3551. doi: 10.1021/acs.chemmater.8b01346.

Koh, D. Y., B. A. McCool, H. W. Deckman, and R. P. Lively. 2016. Reverse osmosis molecular differentiation of organic liquids using carbon molecular sieve membranes. Science 353(6301):804–807. doi: 10.1126/science.aaf1343.

Kresge, C. T., M. E. Leonowicz, W. J. Roth, J. C. Vartuli, and J. S. Beck. 1992. Ordered mesoporous molecular sieves synthesized by a liquid-crystal template mechanism. Nature 359(6397):710–712. doi: 10.1038/359710a0.

Kumar, P., N. Sharma, R. Ranjan, S. Kumar, Z. F. Bhat, and D. K. Jeong. 2013. Perspective of membrane technology in dairy industry: A review. Asian-Australasian Journal of Animal Sciences 26(9):1347–1358. doi: 10.5713/ajas.2013.13082.

Lai, Z., G. Bonilla, I. Diaz, J. G. Nery, K. Sujaoti, M. A. Amat, E. Kokkoli, O. Terasaki, R. W. Thompson, M. Tsapatsis, and D. G. Vlachos. 2003. Microstructural optimization of a zeolite membrane for organic vapor separation. Science 300(5618):456–460. doi: 10.1126/science.1082169.

Lehn, J. M. 1988. Supramolecular Chemistry-Scope and Perspectives. Journal of Inclusion Phenomena 6:351-396.

Lehn, J. M. 1995. Supramolecular Chemistry: Concepts and Perspectives. VCH: Weinheim.

Li, H., Z. Song, X. Zhang, Y. Huang, S. Li, Y. Mao, H. J. Ploehn, Y. Bao, and M. Yu. 2013. Ultrathin, molecular-sieving graphene oxide membranes for selective hydrogen separation. Science 342(6154):95–98. doi: 10.1126/science.1236686.

Lively, R. P., R. R. Chance, B. T. Kelley, H. W. Deckman, J. H. Drese, C. W. Jones, and W. J. Koros. 2009. Hollow fiber adsorbents for CO₂ removal from flue gas. Industrial & Engineering Chemistry Research 48(15):7314–7324. doi: 10.1021/ie9005244.

Lligadas, G., S. Grama, and V. Percec. 2017. Recent developments in the synthesis of biomacromolecules and their conjugates by single electron transfer–living radical polymerization. Biomacromolecules 18(4):1039–1063. doi: 10.1021/acs.biomac.7b00197.

Low, Z. X., Y. T. Chua, B. M. Ray, D. Mattia, I. S. Metcalfe, and D. A. Patterson. 2017. Perspective on 3D printing of separation membranes and comparison to related unconventional fabrication techniques. Journal of Membrane Science 523:596–613. doi: 10.1016/j.mem-sci.2016.10.006.

Luelf, T., D. Rall, D. Wypysek, M. Wiese, T. Femmer, C. Bremer, J. U. Michaelis, and M. Wessling. 2018. 3D-printed rotating spinnerets create membranes with a twist. Journal of Membrane Science 555:7–19. doi: 10.1016/j. memsci.2018.03.026.

Lumetta, G. J., R. D. Rogers, and A. S. Gopalan (editors). 2000. Calixarenes for Separations. ACS Symposium Series 757.

Matyjaszewski, K., D. Hongchen, W. Jakubowski, J. Pietrasik, and A. Kusumo. 2007. Grafting from surfaces for “everyone”: ARGET ATRP in the presence of air. Langmuir 23(8):4528–4531. doi: 10.1021/la063402e.

Matyjaszewski, K., and J. Xia. 2001. Atom transfer radical polymerization. Chemical Reviews 101(9):2921–2990. doi: 10.1021/cr940534g.

Mi, L., M. T. Bernards, G. Cheng, Q. Yu, and S. Jiang. 2010. pH responsive properties of non-fouling mixed-charge polymer brushes based on quaternary amine and carboxylic acid monomers. Biomaterials 31(10):2919–2925. doi: 10.1016/j.biomaterials.2009.12.038.

Moyer, B. A. (editor). 2013. Ion Exchange and Solvent Extraction: Volume 21, Supramolecular Aspects of Solvent Extraction. Boca Raton, Florida: CRC Press.

Moyer, B. A., and R. P. Singh (editors). 2004. Fundamentals and Applications of Anion Separations. Boston, MA: Springer.

Nilsson, M., P. R. Zalupski, and M. R. Antonio. 2013. Supramolecular aspects of stable water-in-oil microemulsions in chemical separations. Ion Exchange and Solvent Ex

traction: Volume 21, Supramolecular Aspects of Solvent Extraction. Boca Raton, Florida: CRC Press, 197–268.

NRC (National Research Council). 1987. Separation Purification: Critical Needs and Opportunities. Washington, DC: The National Academies Press.

Nunes, S. P., R. Sougrat, B. Hooghan, D. H. Anjum, A. R. Behzad, L. Zhao, N. Pradeep, I. Pinnau, U. Vainio, and K. V. Peinemann. 2010. Ultraporous films with uniform nanochannels by block copolymer micelles assembly. Macromolecules 43(19):8079–8085. doi: 10.1021/ma101531k.

Orjuela, Alvaro, Miguel A. Santaella, and Paola A. Molano. 2016. Process Intensification by Reactive Distillation. Process Intensification in Chemical Engineering: Design Optimization and Control 131-181. Cham: Springer International Publishing.

Ostuni, E., R. G. Chapman, R. E. Holmlin, S. Takayama, and G. M. Whitesides. 2001. A survey of structure-property relationships of surfaces that resist the adsorption of protein. Langmuir 17(18):5605–5620. doi: 10.1021/la010384m.

Patil, S. P., A. Rege, M. Itskov, and B. Markert. 2018. Fracture of silica aerogels: An all-atom simulation study. Journal of Non-Crystalline Solids 498:125–129. doi: 10.1016/j.jnoncrysol.2018.06.005.

Pedersen, C. J. 1978. Synthetic Multidentate Macrocyclic Compounds. Izatt, R. M. and Christensen, J. J., Eds.; Academic Press: New York, p 1-52.

Plechkova, N. V., and K. R. Seddon. 2014. Ionic Liquids Further UnCOILed: Critical Expert Overviews. Hoboken, NJ: John Wiley & Sons.

Pollo, B. J., G. L. Alexandrino, F. Augusto, and L. W. Hantao. 2018. The impact of comprehensive two-dimensional gas chromatography on oil & gas analysis: Recent advances and applications in petroleum industry. Trends in Analytical Chemistry 105:202-217. doi: https://doi.org/10.1016/j.trac.2018.05.007.

Prebihalo, S. E., K. L. Berrier, C. E. Freye, H. D. Bahaghighat, N. R. Moore, D. K. Pinkerton, and R. E. Synovec. 2018. Multidimensional gas chromatography: Advances in instrumentation, chemometrics, and applications. Analytical Chemistry 90(1):505–532. doi: 10.1021/acs.analchem.7b04226.

Presser, V., J. McDonough, S.-H. Yeon, and Y. Gogotsi. 2011. Effect of pore size on carbon dioxide sorption by carbide derived carbon. Energy & Environmental Science 4:3059-3066. doi: 10.1039/C1EE01176F.

Qiu, X., H. Yu, M. Karunakaran, N. Pradeep, S. P. Nunes, and K. V. Peinemann. 2013. Selective separation of similarly sized proteins with tunable nanoporous block copolymer membranes. ACS Nano 7(1):768–776. doi: 10.1021/nn305073e.

Rajagopalan, A. K., A. M. Avila, and A. Rajendran. 2016. Do adsorbent screening metrics predict process performance? A process optimisation based study for postcombustion capture of CO₂. International Journal of Greenhouse Gas Control 46:76–85. doi: 10.1016/j.ijg-gc.2015.12.033.

Rana, D., and T. Matsuura. 2010. Surface modifications for antifouling membranes. Chemical Reviews 110(4):2448–2471. doi: 10.1021/cr800208y.

Reyes-Garcés, N., E. Gionfriddo, G. A. Gómez-Ríos, M. N. Alam, E. Boyacl, B. Bojko, V. Singh, J. Grandy, and J. Pawliszyn. 2018. Advances in solid phase microextraction and perspective on future directions. Analytical Chemistry 90(1):302–360. doi: 10.1021/acs.analchem.7b04502.

Robinson, J., D. Roush, and S. Cramer. 2018. Domain contributions to antibody retention in multimodal chromatography systems. Journal of Chromatography A 1563:89–98. doi: 10.1016/j.chroma.2018.05.058.

Schneider, H. J. 2012. Applications of Supramolecular Chemistry. Boca Raton, Florida: CRC Press.

Shannon, M. A., P. W. Bohn, M. Elimelech, J. G. Georgiadis, B. J. Marĩas, and A. M. Mayes. 2008. Science and technology for water purification in the coming decades. Nature 452(7185):301–310. doi: 10.1038/nature06599.

Shiflett, M. B., and E. J. Maginn. 2017. The solubility of gases in ionic liquids. AIChE Journal 63(11):4722–4737. doi: 10.1002/aic.15957.

Shiflett, M. B., and A. M. Scurto. 2017. Ionic liquids: Current state and future directions. ACS Symposium Series 1250:1–13.

Smithers, G. W. 2008. Whey and whey proteins—From “gutter-to-gold.” International Dairy Journal 18(7):695–704. doi: 10.1016/j.idairyj.2008.03.008.

Stankiewicz, A. J. and J. A. Moulijn. 2000. Process Intensification: Transforming Chemical Engineering. Chemical Engineering Progress 96:22-34.

Sun, H., Z. Xu, and C. Gao. 2013. Multifunctional, ultra-flyweight, synergistically assembled carbon aerogels. Advanced Materials 25(18):2554–2560. doi: 10.1002/adma.201204576.

Thakkar, H., S. Eastman, A. Hajari, A. A. Rownaghi, J. C. Knox, and F. Rezaei. 2016. 3D-printed zeolite monoliths for CO₂ removal from enclosed environments. ACS Applied Materials and Interfaces 8(41):27753–27761. doi: 10.1021/acsami.6b09647.

Tkacik, G. 2015. Advances in membrane morphology control for optimum performance in biotechnology separations. Presented at the North American Membrane Society 25th Annual Meeting, Boston, MA, May 30-June 3, 2015.

Wagner, B. M., S.A. Schuster, B.E. Boyes, T. J. Shields, W.L. Miles, M. J. Haynes, R.E. Moran, J.J. Kirkland, and M.R. Schure. 2017. Superficially porous particles with 1000 angstrom pores for large biomolecule high performance liquid chromatography and polymer size exclusion chromatography. J. Chromatogr. A 1489:75-85. doi: 10.1016/j.chroma.2017.01.082.

Wang, Z., J. Wang, M. Li, K. Sun, and C. J. Liu. 2014. Three-dimensional printed acrylonitrile butadiene styrene framework coated with Cu-BTC metal-organic frameworks for the removal of methylene blue. Scientific Reports 4. doi: 10.1038/srep05939.

Wasserscheid, P., and T. Welton. 2007. Ionic Liquids in Synthesis. 2nd edition, Vol. 1. Hoboken, NJ: John Wiley & Sons.

Wei, B. C., B. Zhang, B. Boyes, and Y.T. Zhang. 2017. Reversed-phase chromatography with large pore superficially porous particles for high throughput immunoglobulin G(2) disulfide isoform separation. J. Chromatogr. A, 1526:104-111. doi: 10.1016/j.chroma.2017.10.040.

Weinfeld, J. A., S. A. Owens, and R. B. Eldridge. 2018. Reactive dividing wall columns: A comprehensive review. Chemical Engineering and Processing—Process Intensification 123:20–33. doi: 10.1016/j.cep.2017.10.019.

Werber, J. R., C. O. Osuji, and M. Elimelech. 2016. Materials for next-generation desalination and water purification membranes. Nature Reviews Materials 1. doi: 10.1038/natrevmats.2016.18.

West, C., C. Elfakir, and M. Lafosse. 2010. Porous graphitic carbon: A versatile stationary phase for liquid chromatography. Journal of Chromatography A 1217(19):3201–3216. doi: 10.1016/j.chroma.2009.09.052.

Xiu S. Z., G. Q. (Max) Lu, and G. J. Millar. 1996. Advances in Mesoporous Molecular Sieve MCM-41. Industrial and Engineering Chemistry Research 35(7):2075-2090. doi: 10.1021/ie950702a.

Xu, L., M. Rungta, M. K. Brayden, M. V. Martinez, B. A. Stears, G. A. Barbay, and W. J. Koros. 2012. Olefinsselective asymmetric carbon molecular sieve hollow fiber membranes for hybrid membrane-distillation processes for olefin/paraffin separations. Journal of Membrane Science 423–424:314–323. doi: 10.1016/j.mem-sci.2012.08.028.

Yaghi, O.M., G. Li, and H. Li. 1995. Structural studies and conductivity of [Fe(O3C4)(COO)]·H2O based H4btec (H4btec = 1,2,4,5-benzenetetracarboxylic acid). Nature 378:703-706. doi: 10.1038/378703a0.

Yaghi, O.M., and H. Li. 1995. Hydrothermal synthesis of a metal-organic framework containing large rectangular channels. Journal of American Chemical Society 117(41):10401-10402. doi: 10.1021/ja00146a033.

Yin, J., E. S. Kim, J. Yang, and B. Deng. 2012. Fabrication of a novel thin-film nanocomposite (TFN) membrane containing MCM-41 silica nanoparticles (NPs) for water purification. Journal of Membrane Science 423–424:238–246. doi: 10.1016/j.memsci.2012.08.020.

Zhang, C., and M. M. El-Halwagi. 2017. Estimate the capital cost of shale-gas monetization projects. Chemical Engineering Progress 113(12).

Zhang, F., Y. Ma, J. Liao, V. Breedveld, and R. P. Lively. 2018. Solution-based 3D printing of polymers of intrinsic microporosity. Macromolecular Rapid Communications 39(13). doi: 10.1002/marc.201800274.