Sustainable development, i.e. meeting the present needs without affecting future generations' ability to meet their own requirements, has been a major worldwide issue for several decades, having been addressed at the 1987 Brundtland Commission (United Nations Commission on Environment and Development).¹ 

Two key aspects, arising from sustainable development, that are valid from a chemical, engineering and energy point of view are: (i) the rate at which the current generation can consume fossil fuels and (ii) the quantities of waste that the environment can support on a sustainable basis.

The Earth has a natural capacity to cope with the waste and pollution produced by society and, when this capacity is surpassed, unsustainability results.^1,2  Furthermore, it is becoming increasingly clear that the rate at which non-renewable fossil feedstocks, i.e. coal, oil and natural gas, are being consumed is much higher than the rate at which natural geological processes can replace them.³  This makes their use, in the long run, unsustainable. Using fossil fuels also leads to the production of high rates of carbon dioxide (CO₂), rates that are much higher than what can be assimilated by the environment, something that the scientific community widely accepts as the leading problem of climate change.³ 

So as to tackle this sustainability issue, the United States Environmental Protection Agency (EPA) coined the term Green Chemistry in the early 1990s, defined as follows:

This definition indicates that green chemistry is a philosophy rather than some new type of chemistry, whereby the practice of chemistry and engineering should be done in a sustainable manner. With this in mind, Paul Anastas and John Warner developed the 12 Principles of Green Chemistry in 1998, shown in Figure 1.1.⁵ 

One strategy for meeting the principles of green chemistry is to reduce the use of hazardous organic solvents and to encourage the utilisation of sustainable or more environmentally friendly greener solvents. Several of the twelve principles of green chemistry have a strong connection with the use of supercritical, superheated or pressurised solvents. The most obvious of which are number 5 “the use of auxiliaries (e.g. solvents) should be made unnecessary whenever possible and innocuous when used” and number 3 “synthetic methodologies should be designed to use and generate substances that possess little or no toxicity to human health and the environment”. Common supercritical solvents utilised throughout the research community include carbon dioxide and water, both of which are relatively non-toxic, abundant and can be easily recycled.

Legislation surrounding solvent registration, application, disposal and emission have been introduced and enforced in order to ensure that solvents are safely used in industry.⁶  Legislation has effectively led to the reduction in the use of more toxic solvents via the complete prohibition of their use or by introducing maximum residual levels. A number of solvent guides have been developed in order to recommend to industry suitable alternatives to solvents that are deemed undesirable.^7–9  The CHEM21 (Chemical Manufacturing Methods for the 21st Century Pharmaceutical Industries) solvent guide, which ranks the solvent greenness based on a benchmark of existing solvent selection guides, highlights water as a recommended solvent (Figure 1.2).⁹  While water is regarded as being favourable, Figure 1.2 demonstrates that non-polar solvents, such as the hydrocarbons, are all regarded as hazardous or problematic due to consistently bad safety and environmental scores.⁹ 

Although supercritical fluids such as carbon dioxide and also gas expanded liquids were not included in the guide, the authors did highlight that the project CHEM21 did involve the use of these solvent systems.⁹  Supercritical carbon dioxide can be utilised as a greener alternative to traditional non-polar solvents such as the hydrocarbons, e.g. hexane.¹⁰  Pressurised solvent systems can improve efficiency and reduce solvent usage. However, supercritical, superheated or pressurised solvents are consistently criticised for being energy intensive processes and, as such, this must be considered in any environmental and economic evaluation of such a process.¹¹ 

The standard definition of a supercritical fluid is any substance that is above its critical temperature and pressure (P_c, T_c).¹²  Variations in the temperature and pressure lead to a change in the physical properties of a substance; a phenomenon that can be better explained by referring to a phase diagram.¹³ 

Figure 1.3 represents a typical P–T phase diagram for a substance. All substances that are stable are said to have a triple point and a critical point.¹⁴  The former refers to the pressure and temperature conditions at which the solid, liquid and gas phases coexist as they are in equilibrium with each other. The line that moves from the triple point to the critical point, found lying between the liquid and gaseous regions, is called the gas–liquid (G–L) coexistence curve.¹⁵  When moving towards the critical point along this G–L curve, thermal expansion causes the density of the liquid to decrease while the increase in pressure causes the density of the gas to increase; the densities of the two phases become equivalent at the critical point, making it no longer possible to distinguish between the liquid and gas phases due to them having identical properties. Therefore, the critical point, consisting of the critical temperature and critical pressure, can be described as being the maximum pressure and temperature applied wherein a substance exists as a liquid and a gas in equilibrium with one another. Beyond the critical point, there is no longer a distinction between the liquid and gas phases.^13,15  The critical points for various common supercritical fluids can be found in Table 1.1.

From a macroscopic point of view, the G–L coexistence curve terminates at the critical point. This is the standard explanation and it is assumed that, once the critical point is surpassed, the substance acts as a homogeneous fluid (no longer a biphasic heterogeneous system). However, recent work has shown that, even under supercritical conditions, there are two regions that exhibit distinct liquid-like and gas-like behaviour.^16,17  It is widely reported that there is a maximum in the specific heat (C_p) as the temperature or pressure is changed across an extension line of the G–L coexistence curve, referred to as the ‘Widom line’ shown in Figure 1.3. The Widom line can be viewed as the transition line between gas-like and liquid-like behaviour in supercritical fluids. It has been defined by McMillan et al. as “the locus of maximum correlation length that extends into the single fluid phase beyond the critical point”.¹⁷ 

Studies have shown that the most interesting supercritical fluid (SCF) applications take place in the range 1 < T/T_c < 1.1 and 1 < P/P_c < 2. Within this range, the SCFs are found in a single condensed phase with physical properties intermediate between those of a liquid and a gas.¹²  Varying the pressure and temperature slightly could result in significant changes in these properties.¹⁸  Numerous authors have shown that the “logarithm of solubility is approximately linearly dependent on the solvent density”.^19–21  This was demonstrated in various experiments conducted on hydrocarbons, whereby it was shown that a change in the density of the SCF (supercritical CO₂, ethane and ethylene) led to an exponential variation in the solubility of the hydrocarbons.^22–24  This work demonstrated that slight modifications to the pressure and temperature result in large variations in the density which, in turn, leads to changes in the solvent properties that are density-dependent, including the solubility parameter, partition coefficient and dielectric constant.²⁵  Furthermore, besides the solubility, the solvent power of SCFs are also strongly influenced by other factors including the diffusivity and viscosity.²⁶  The density coefficient, along with the viscosity and diffusion coefficients, are summarised in Table 1.2. It can be noted that the diffusivity of a supercritical fluid is typically an order of magnitude higher than that of a liquid, while the viscosity is an order of magnitude lower.^26,27  In addition, there are SCFs, including supercritical CO₂ (scCO₂), that have a negligible vapour pressure, which leads to enhanced heat and mass transfer.²⁸ 

Another interesting phenomenon, introduced by Dobbs and Johnston, is the entrainer effect, in which certain solutes present in the supercritical phase act as co-solvents leading to greater solubility enhancements.²⁹  Dobbs and Johnston carried out solubility measurements for different systems (binary, ternary and quaternary), which consisted of scCO₂, a co-solvent and combinations of solid phases. In the ternary systems composed of two solutes and the supercritical phase, a proportional increase in one solute's solubility was observed relative to the second solute's solubility. This causes an entraining effect, wherein the more soluble solute leads to an enhancement of the less soluble one. As an example, considering naphthalene and phenanthrene, the former has a much higher solubility in CO₂ than the latter. However, in a system comprising scCO₂, naphthalene and phenantrene, the solubility of phenanthrene is raised by naphthalene by 75%. Therefore, the additions of certain molecules to SCFs leads to a variation in the solvent properties.²⁹ 

Dobbs et al. found that adding small amounts of numerous co-solvents leads to a noticeable improvement in the non-polar SCF selectiveness for polar vs non-polar solid compounds.^29,30  An enhancement in the solubility of certain solids (of higher polarity) in scCO₂ was observed with the addition of several mol% of different co-solvents. Upon addition of around 3.5 mol% of methanol, there was an increase in the solubility of 2-aminobenzoic acid in scCO₂ by 620%.^29,30 

By far, the most popular SCF is scCO₂ as it is an ideal solvent for a variety of different applications, ranging from extraction and separation (chromatography) to reactions, processing of materials and power generation.³¹  One can exploit the advantages of near-critical operation at low temperatures (<35 °C) due to the CO₂ relatively low T_c. CO₂ has favourable health, safety and environmental characteristics, being non-flammable and demonstrating very low toxicity. Furthermore, it is cheap, widely available, unregulated and easily recyclable. The relatively high P_c of CO₂ (73.8 bar) may be seen as a disadvantage. However, it has become fairly routine to operate at such pressures in industrial processes like industrial scCO₂ extraction (such as in the decaffeination of coffee and extraction of hops).^31,32 

scCO₂ has a polarity that is quite similar to the polarities of hexane and toluene.^33,34  The addition of probe molecules, for example Reichardt's dye [2,6-diphenyl-4-(2,4,6-triphenylpyridinio) phenolate], is a practical and easy way to measure the polarity of a solvent.^35,36  Since Reichardt's dye is a zwitterionic molecule, it demonstrates solvatochromic effects as a result of the ground state of the dye interacting with the solvent.³⁵  The E_T(30) and E_T^N scales are the empirical scales of solvent polarity associated with Reichardt's dye:

Kamlet and Taft proposed that the interaction between the phenoxide oxygen and the solvent in question is around 2/3 of the shift of the maximum absorption wavelength of Reichardt's dye. From E_T(30), values of 30.7 kcal mol⁻¹ and 63.1 kcal mol⁻¹ are obtained for tetramethylsilane (TMS) (highly non-polar solvent) and water (highly polar solvent), respectively. These values can then be used to normalise the scale such that a dimensionless figure is obtained, whereby E_T^N = 0 for tetramethylsilane and E_T^N = 1 for water.³⁵ 

As can be seen from Table 1.3, the E_T^N value for scCO₂ lies between those of hexane and toluene. It can therefore be taken as a general rule of thumb that compounds of low molar mass, which have appreciable vapour pressures, that are soluble in hexane should also dissolve in scCO₂ (this is, however, not always the case).

With chemists and industrialists continuously attempting to develop greener and more environmentally benign chemical processes, there is ever-growing attention towards using water close to or above the critical point (374 °C, 220.5 bar) as a medium in chemical processes.³⁷  This is due to the fact that the replacement of conventional organic solvents with subcritical (near-critical) or supercritical water in chemical processes can have environmental advantages as well as prevent pollution. Significant research has been dedicated to utilising sub- and supercritical water in a variety of applications, including organic chemistry, biomass processing, waste treatment, synthetic fuel production, synthesis of materials and geochemistry.^37,38 

There are substantial differences between the physical properties of ambient liquid water and the physical properties of water near the critical point.³⁸  Near-critical water (NCW) has a much lower polarity (dielectric constant, α and E_T^N) and there is a significant decrease in the number and persistence of hydrogen bonds. This leads to much higher solubilities of organic compounds in NCW (it behaves more like typical organic solvents) compared to water at room temperature and there is complete miscibility of organic compounds with supercritical water (SCW). As a general rule of thumb, the E_T^N of NCW is similar to that of acetone, while at higher temperatures it has complete miscibility with toluene.³⁹  NCW also has lower density, surface tension and viscosity compared to ambient liquid water; however, with the changing temperature, the diffusivity and specific heat capacity of NCW increases while the hydrogen-bond acceptor ability (β) remains constant.³⁸ 

Since the dissociation constant (K_w) increases with the increasing temperature, the concentration of H₃O⁺ (hydronium) and OH⁻ (hydroxide) ions increases and, as water approaches the critical point, its K_w is three orders of magnitude higher than that of water at room temperature.³⁹  Therefore, dense high-temperature water can be used effectively for acid- and base-catalysed organic reactions. In fact, in certain acid-catalysed reactions, the H⁺ concentration of NCW is high enough that no additional acid needs to be introduced. However, as the critical point of water is surpassed (becomes SCW), there is a dramatic decrease in K_w – as an example, at conditions of 600 °C and 253.3 bar, K_w is approximately nine orders of magnitude lower than that of water at room temperature, making SCW a poor solvent for ionic chemistry in this low-density, high-temperature medium.^38,39 

The dielectric constant of water increases when increasing the density and decreases when increasing the temperature. The typical high ε value of 80 only takes place at low temperatures in a small region. ε values of 10–25 occur in a large supercritical region at high density; which are similar to the ε values of dipolar solvents such as acetone or acetonitrile at ambient conditions (shown in Figure 1.4).⁴⁰  ε values of 10–25 are adequately large enough for dissolving and ionising electrolytes while also allowing for miscibility with non-polar molecules. The dielectric constant rapidly decreases at low densities resulting in a decrease in the ability to dissolve and ionise electrolytes. The dielectric constant has a value of six at the critical point.³⁹ 

A further difference between SCW and water at room temperature is that the continuous variation of the properties (including K_w, dielectric constant, viscosity, etc.) occurs over much larger ranges under supercritical conditions, which enables the possibility of fine-tuning the properties of the reaction medium for achieving optimal results by varying the pressure and temperature conditions.²⁸ 

Pressurised liquids, including gas expanded liquids, have been utilised as solvent systems for extractions and as reaction media.^41,42 

Pressurised liquid systems involve the use of liquid solvents at elevated temperature and pressure, to enhance the performance as compared to processes carried out at room temperature and atmospheric pressure.⁴³  Utilising solvents at temperatures above their atmospheric boiling point enhances the solubility and mass transfer properties, thus resulting in higher diffusion rates within the system. The elevated pressure maintains the solvent below its boiling point and in its liquid state. Pressurised liquid extraction (PLE) has been used to reduce extraction times and improve extraction yields, whilst reducing the solvent consumption.⁴⁴  A key advantage of PLE is that solvent mixtures can be tuned to enhance both the efficiency and selectivity of the extraction. Furthermore, automated pressurised solvent systems may also protect light or oxygen sensitive compounds.⁴²  PLE technologies have been demonstrated as effective extraction systems for a range of ‘bioactive compounds’ from foods and herbal plants.⁴² 

A gas-expanded liquid (GXL) is a mixture of a compressible gas dissolved in an organic solvent. Significant work has been carried out to develop generic approaches to assess and design gas-expanded liquids.⁴¹  The properties of carbon dioxide-expanded liquids (CXLs) span the range from those of the pure organic solvent to those of carbon dioxide.⁴⁵  Variation of the ratio between the solvent and carbon dioxide can lead to tuneable solvent properties. These solvent systems have demonstrated promise in applications such as separations, extractions, reactions, fine particle precipitation, polymer processing and other applications.⁴⁶  Carbon dioxide has a number of advantages for use in such processes, including being inert, widely available, low cost, non-flammable and easily removed at the end of the process. CXLs demonstrate high gas miscibility, enhanced transport rates compared to conventional solvent systems and they require lower pressures compared to supercritical carbon dioxide.⁴⁵  The use of such systems also reduces the volume of organic solvent required, thereby helping to prevent waste in chemical processes (a key principle of green chemistry).

Pressurised-solvent systems including supercritical fluids, sub-critical fluids and gas-expanded liquids are becoming ever-more important, especially with the growing legislation restricting the use of conventional organic solvents in industrial chemical processes, as well as the strict guidelines for minimal solvent residues in consumer products. The use of high-pressure systems enables the possibility to design clean, sustainable and environmentally friendly processes, as well as generate novel products with unique properties. The use of supercritical fluids and also pressurised solvents systems is of global interest for use in extraction, reaction and materials processing. In the following chapters, examples of how these pressurised-solvent systems can be utilised in a variety of different processes as effective alternative solvents for obtaining different products will be presented, including the latest innovative research on the use of supercritical, subcritical fluids and gas-expanded liquids in extractions, reactions, processing of materials and power generation.