9	Synthesis of Higher Diamondoids by Pulsed Laser Ablation Plasmas in Supercritical Fluids Sven Stauss, Sho Nakahara, Toru Kato, Takehiko Sasaki, and Kazuo Terashima

CONTENTS

9.1    Introduction

9.1.1    Structure and Physical Properties of Diamondoids

9.1.2    Applications of Diamondoids and Diamondoid Derivatives

9.1.3    Natural Occurrence of Diamondoids and Their Isolation

9.1.4    Conventional Synthesis of Diamondoids and Its Limitations

9.2    Generation of Pulsed Laser Plasmas in Supercritical Fluids

9.2.1    Basic Characteristics and Current Applications of Supercritical Fluids

9.2.2    Generation of Pulsed Laser Plasmas in Supercritical Fluids

9.2.3    Application of Pulsed Laser Plasmas in Supercritical Fluids to Nanomaterials Synthesis

9.3    Synthesis of Diamondoids by Pulsed Laser Plasmas

9.3.1    Experimental Procedure

9.3.2    Micro-Raman Spectroscopy

9.3.3    Gas Chromatography–Mass Spectrometry

9.3.3.1    Synthesis of Diamantane

9.3.3.2    Synthesis of Diamondoids with n ≥ 3

9.3.4    Effects of Pyrolysis on the Synthesized Products

9.3.5    Comparison between PLA in scCO₂ and SCXE

9.4    Conclusions and Perspectives

Acknowledgments

References

9.1    INTRODUCTION

Metallic, semiconducting, and carbon nanomaterials and nanoparticles are increasingly finding applications in a variety of fields, ranging from biotechnology and medicine to opto- and nanoelectronics. One of the main difficulties in the synthesis of nanoparticles and their application is to limit the size of particles to the nanometer range and to realize very narrow size distributions. However, this is often difficult to achieve by conventional materials processing techniques, making it necessary to develop new approaches to overcome these obstacles.

Supercritical fluids (SCFs) are increasingly finding applications as high-pressure media in analytic chemistry [1] or for chemical synthesis [2]. SCFs allow their properties, e.g., thermal capacity, thermal conductivity, density fluctuation, dielectric constant, diffusivity, etc., to be tuned continuously by adjusting either the temperature, pressure, or both. On the other hand, plasmas represent highly reactive media containing electrons, excited neutral radicals, and positively and negatively charged atomic and molecular ions, allowing us to realize reactions that are far from thermodynamic equilibrium [3,4]. Plasmas generated in supercritical fluids combine the advantageous transport properties of SCFs with the high reactivity of plasmas. This allows the synthesis of novel nanomaterials or facilitates the tailoring of existing nanomaterials that could be used in a variety of applications, such as medicine, pharmaceutics, biotechnology, nanooptics, and nanoelectronics.

Here we review the current state of research on the synthesis of a special class of carbon nanomaterials that recently has gained (renewed) interest in the scientific community, diamondoids [5,6], by pulsed laser ablation (PLA) plasmas generated in SCFs. We briefly describe the structure and physical properties of diamondoids and their possible applications before introducing the techniques used for realizing chemical reactions by PLA in high-pressure and SCF media. Finally, we present the latest results of the synthesis of diamondoids by PLA in supercritical Xe (scXe) and supercritical CO₂ (scCO₂). It is anticipated that this overview will further increase the interest in this fascinating class of carbon nanomaterials and also further fuel the application of pulsed laser plasmas generated in SCFs for the synthesis of nanomaterials.

9.1.1    STRUCTURE AND PHYSICAL PROPERTIES OF DIAMONDOIDS

Macroscopically, carbon can exist under different forms or allotropes: amorphous carbon, graphite, hexagonal diamond (Lonsdaleite), and cubic diamond. In the last 25 years, other members of carbon allotropes at the nanoscale have been discovered [7]: carbyne and polyyne [8], fullerenes [9], carbon nanotubes [10], graphene [11], nanodiamonds, and diamondoids [5].

The ability of carbon to form different chemical bonds that result in this large variety of nanomaterials is due to the electronic structure of carbon.

The 2s and 2p atomic orbitals of carbon can be mixed in different combinations that can lead to the formation of either sp¹-, sp²-, or sp³-hybridized orbitals. Figure 9.1 illustrates the carbon nanostructures formed as a result of the three types of hybridization. For sp¹ hybridizations, diacetylene (or butadiene) can be considered the basic block for the assembly of linear chains, cumulene, or polyyne. The linking of fragments of ovalene (sp²) conceptually leads to the formation of fullerenes, whereas corannulene can be considered to be a basic unit for the creation of carbon nanotubes (CNTs) and graphenes.

FIGURE 9.1 Overview of carbon nanostructures formed as a consequence of the different hybridizations of C, sp¹, sp², and sp³. sp¹ hybridization leads to the formation of linear chains, cumulene, or polyyne. For sp²-hybridized carbon nanostructures, corannulene can be considered a building block for fullerene, whereas ovalene can be regarded as a precursor for carbon nanotubes or graphene. In contrast to fullerenes, carbon nanotubes, and graphene, the C-C bonds in adamantane are sp³ hybridized and the fusion of a number of these building blocks conceptually leads to the formation of diamondoids and, as the number of cages increases further, nanodiamonds. The molecular structures in this and following figures have been rendered by visual molecular dynamics (VMD). (From W. Humphrey et al., J. Mol. Graph., 14(1), 33, 1996. With permission.)

Finally, sp³ hybridization leads to the formation of a rigid cage structure consisting of carbon atoms.

Therefore the main difference between the other carbon nanomaterials, fullerenes, CNTs, and graphene that are comprised of sp²-hybridized bonds between carbon atoms and diamondoids is that for the latter, their structure consists of C(sp³)-C(sp³) bonds. As illustrated in Figure 9.2, the carbon cages comprising the diamondoids can be superimposed on a diamond lattice. The terminations of the carbon cage structure are well defined and usually consist of hydrogen atoms. The first member of the diamondoid series is admantane (C₁₀H₁₆, number of cages n = 1), and larger diamondoids can conceptually be formed by fusing increasingly larger numbers of these building blocks. The next members in the diamondoid series are diamantane (n = 2) and triamantane (n = 3). From tetramantane (n = 4), the possibilities for fusing additional diamondoid cages grow and the number of isomers increases. This leads to different diamondoid series with the molecular formulae C_4n+6H_4n+12, C_4n+5H_4n+10, C_4n+2H_4n+6, C_4n+4H_4n+8, C_4n+1H_4n+4, C_4n−2H_4n, C_4n−5H_4n−4, and C_4n−3H_4n−2 [5].

FIGURE 9.2 Relation of the diamondoid structure and diamondoid lattice and molecular structures of lower and higher diamondoids. (a) The first member of diamondoids, adamantane, can be regarded as the basic unit of diamond, and diamondoids can be superimposed on a diamond lattice. (b) Structures of lower diamondoids adamantane, diamantane, triamantane, and their relation to a 279-carbon tetrahedral diamond lattice. For higher clarity, the H-terminations are not shown except for one adamantane. For the diamondoids from n = 1 to n = 4 the dual graphs for the systematic classification of diamondoids introduced by Balaban and Schleyer [13] are also indicated.

In order to facilitate the classification of diamondoids, a nomenclature based on graph theory was developed [13]. For this, the centers of the diamondoid cages are connected by lines, the shortest possible path of such a dual graph giving the name of the diamondoid, together with the Greek expression for the number of fused adamantane cages and the suffix -mantane. Figure 9.2(b) shows the structure of the lower diamondoids adamantane, diamantine, and triamantane, and the first higher-order diamondoid, tetramantane, with their corresponding dual graphs and notation. As illustrated for n = 4, the cages can be fused in different combinations, leading to the formation of the [1(2)3](anti), [121](iso), [123](M)(skew), and [123](P)(skew) structural isomers of tetramantane.

While the fusion of increasingly higher numbers of adamantane cages approaches the size of nanodiamonds and ultimately diamond, it is important to make a clear distinction between nanodiamonds and diamondoids. Usually the term nanodiamond is used in a broader context for a variety of structures that contain diamond nanocrystals, present, for instance, in interstellar dust and meteorites, diamond particles nucleated in the gas phase or on substrates, and nanocrystalline films [7]. Consequently, nanodiamonds may have a wide range of size scales, size distributions, and also consist of mixtures containing other elements (i.e., doped nanodiamonds). In contrast, diamondoids “are chemically well defined, of high purity, and structurally well characterized” [6]. Moreover, while the interior of nanodiamonds is composed of sp³-bonded networks, their surface can be terminated by a variety of functional groups or consist of reconstructed surfaces formed by sp²-hybridized graphitic networks [14]. In contrast with nanodiamonds, diamondoids have a well-defined structure and the sp³-bonded carbon cages are terminated by H-atoms or other functional groups. For this reason, diamondoids are sometimes also called diamond molecules.

Therefore despite advances in the purification and functionalization, the major disadvantages of nanodiamonds in comparison to diamondoids are their large size and usually broad size distribution [6] that might limit their application in nanodevices.

In the following sections, we will give a short overview of the possible applications of diamondoids and their derivatives and describe the methods that have been used for obtaining them, either by isolation from natural gas and oil reservoirs or by synthesis using pulsed laser plasmas generated in supercritical xenon and CO₂.

9.1.2    APPLICATIONS OF DIAMONDOIDS AND DIAMONDOID DERIVATIVES

In recent years, diamondoids have elicited increasing interest as possible future nanomaterials for use in a variety of applications.

Work by several groups suggests that the special properties of diamondoids, i.e., high thermal stability, well-defined structure, and nontoxicity, make them interesting for applications in various fields, pharmaceutics, medicine, biotechnology, and nano- and optoelectronics [15].

Because of their nontoxicity and high thermal stability, diamondoids could be used in medical applications, e.g., as biomarkers or as scaffolds for new pharmaceutical drugs. For instance, two derivatives of adamantane, memantine and amantadine, shown in Figure 9.3(a), have been used to treat Alzheimer’s [17] and Parkinson’s [18] diseases, respectively, while another adamantane derivative, rimantadine, has shown antiviral activity [19].

Concerning the use of diamondoids in optoelectronics, it has been reported that large-area self-assembled monolayers (SAMs) of a functionalized diamondoid, [121] tetramantane-6-thiol, could be used for highly monochromatic electron photoemission [20]. The structure of such SAMs is illustrated in Figure 9.4(a). Recent work has shown that the diamondoid monolayer structure and thiol substitution can be controlled, permitting us to change the electronic structure of the SAMs [21].

SAMs of [121]tetramantane-6-thiol on Au or Ag films possess quantum yields larger than unity, and the reason for monochromatic photoelectron emission of diamondoid SAMs is that like diamonds, diamondoids possess a negative electron affinity (NEA) [22].

FIGURE 9.3 Examples of current applications of diamondoids. (a) Amantadine and memantine used for treating Alzheimer’s and Parkinson’s diseases, respectively. (b) Derivatives of adamantane, such as Adamantate™, are used as additives in photoresists to increase the thermal resistance. (c) Inclusion of diamondoids reduces the wettability of metallic coatings (left) when compared to a non-treated metal surface (right). (The photograph (c) is reprinted with permission from H. Schwertfeger and P.R. Schreiner, Chem. Unserer Zeit, 44(4), 248–253, 2010, Copyright 2010, John Wiley & Sons.)

In-depth studies on the mechanisms of monochromatic photoelectron emission revealed that the NEA of diamondoids results from unoccupied states in the lowest unoccupied molecular orbitals (LUMOs), which, when populated by an electron, directly lead to spontaneous electron emission [23].

The detailed mechanism is considered to consist of several steps (see Figure 9.4(b)). Electrons of the metal substrate are excited by photons to unoccupied energy levels above the Fermi level, E_F, which lies in the middle of the energy gap of the diamondoid. The vacuum level (E_vac) is below the conduction band minimum of the diamondoid, which is characteristic of NEA materials.

In the second step, the excited electrons attain thermal equilibrium, producing more electrons with lower energy. Then electrons with energy above the conduction band minimum transfer to the diamondoids. These electrons further lower their energies by exciting phonons in the diamondoid molecules, finally accumulating at the bottom of the conduction band. In the last step, electrons accumulated at the bottom of the conduction band emit into vacuum spontaneously. The main electron emission takes place at the low-kinetic energy threshold. At higher kinetic energy levels, the photoelectron yield is lower.

FIGURE 9.4 Use of diamondoid SAMs as possible sources for efficient photoelectron emitters. (a) Structure of [121]tetramantane-6-thiol and SAM deposited on Au surface. (b) Schematic of the currently understood processes leading to spontaneous photoelectron emission: (1) Photons excite electrons in the metal substrate, (2) the electrons pass to unoccupied states above the Fermi level (E_F), (3) energy transfer of the excited electrons produces additional electrons with lower energies, (4) electrons with energies above the conduction band minimum of the diamondoid transfer to the diamondoid molecule, (5) these electrons further lower their energies by exciting phonons in the diamondoid molecule, accumulating at the bottom of the conduction band, and (6) because of NEA the electrons accumulated at the bottom of the conduction band emit into vacuum spontaneously.

In contrast to Si and Ge nanoparticles, where the LUMO is core confined, the LUMO of diamondoids is a delocalized surface state [24]. The charge transfer mechanism, illustrated in Figure 9.4(b), that allows the electron to be promoted from the metal into the diamondoid LUMO is very efficient, electrons being emitted almost instantaneously with an upper bound of a few femtoseconds. This suggests that if suitable pulsed light sources are available, diamondoid SAMs could be used as ultrashort (~fs), high-brilliance pulsed electron sources for use in electron microscopy, electron beam lithography, and for field-emission flat-panel displays and photocathodes.

Investigations of diamondoids up to [121]tetramantane also showed that the dielectric constant κ is about half that of bulk diamond (κ ~ 5.6), ranging from 2.46 for adamantane to 2.68 for tetramantane [25]. This suggests that diamondoids could be new candidates as low-κ materials for microelectronics applications.

By density functional theory (DFT) simulations, the effect of different doping strategies on the band gap of diamondoids has been studied [26]. The approaches that were investigated were the enlargement and changing of the shapes/morphologies of the particles, the effect of CH bond substitutions with various functional groups (external doping), and finally, the effect of incorporating heteroatom functionalities of one or more CH or CH₂ fragments (interstitial or internal doping). The authors found that increasing the size of diamondoids from C₁₀H₁₆ (adamantane) to about 2 nm (C₂₈₆H₁₄₄) leads to a decrease of the band gap from 9.4 to 6.7 eV, which is still larger than the band gap of bulk diamond (5.5 eV). It was also found that it is the size, and not the morphology of the diamondoids, that influences the band gap most. The authors also discovered that band gap tuning through external (by CH bond substitution) or internal (by replacing CH or CH₂ groups) doping is nonadditive for the same dopant. This means that doubling the number of same functional groups attached to a diamondoid does not necessarily lead to a linear decrease in its band gap. Functionalization by attaching electron-donating and electron-withdrawing groups (push-pull doping) was found to be most effective and permitted us to reduce the band gap of diamondoids to that of bulk diamond.

The effect of clustering on the band gap has also been investigated. DFT calculations showed that in contrast to bulk diamond, crystalline adamantane has a direct band gap. As the transition of electrons across the band gap can accompany the absorption and emission of photons, the efficiency of this process can be expected to be far better in crystalline diamondoid materials than in indirect band gap materials [27].

In summary, the well-defined structure of diamondoids, which distinguishes them from nanodiamonds, makes them predestined as molecular building blocks in a wide range of applications. However, while certain physical properties of diamondoids have already been demonstrated experimentally, the main problem is that they are not readily accessible. Especially higher diamondoids with cage numbers n ≥ 4 are still not available in large enough quantities to be used at an industrial scale.

In the following section, the methods used until now for obtaining higher diamondoids by isolation from natural gas and oil reservoirs are presented.

9.1.3    NATURAL OCCURRENCE OF DIAMONDOIDS AND THEIR ISOLATION

Diamondoids present in natural sources such as gas and oil reservoirs are believed to have formed from kerogen macromolecules in sedimentary rocks. The presence of mineral catalysts such as aluminosilicates present in such sediments seems to promote the formation of diamondoids [28]. The term diamondoid was coined by the German chemist Decker, who tried to synthesize diamonds. The structure of the first diamondoid, adamantane, was predicted in 1933 by Kleinfeller and Frercks. Adamantane was isolated from oil reservoirs near Hodonin, Czechoslovakia, the same year, while the next diamondoid member, diamantane, was isolated in 1966 [29].

FIGURE 9.5 Schematic of the separation process for obtaining single crystals of higher diamondoids [5]. Oil condensates are further concentrated by vacuum distillation at temperatures above 345°C. Volatile components and non-diamondoids are cracked by pyrolysis conducted in the temperature range of 400–450°C. In a next step, polar and aromatic compounds are removed by silica-gel-based column chromatography. Individual types of higher diamondoids are isolated by reverse phase HPLC, enantiomeric diamondoids being separated by shape-selective HPLC. The separated fractions are finally recrystallized.

Diamondoids were often considered a nuisance by oil mining companies, because of the clogging of oil pipes. In the 1990s, Chevron Oil looked for methods to solve these problems and for how to isolate diamondoids from petroleum. Dahl et al. [5] then gave the first proof for the existence of higher diamondoids up to undecamantane (n = 11). They quickly realized that these compounds could be very interesting for other applications and patented methods for the extraction of diamondoids.

The isolation of higher diamondoids and their purification has to be done in several steps, as indicated in Figure 9.5. First, oil condensates are concentrated by vacuum distillation at temperatures above 345°C. Pyrolysis at temperatures between 400 and 450°C permits removal of volatile components and non-diamondoids. Remaining polar and aromatic compounds are then removed by silica-gel-based column chromatography. To separate individual diamondoids and enantiomers, reverse phase high-performance liquid chromatography (HPLC) followed by shape-selective HPLC is used. The separated fractions can then be recrystallized.

In addition to the isolation and purification processes developed [5] and briefly described above, Iwasa et al. [30] presented a vapor transport technique where a two-zone furnace is used. With this approach, it is possible to both obtain diamondoid single crystals with volumes up to ~ 1 cm³ and reduce the number of impurities considerably. In the next section, we briefly give an overview on conventional synthesis approaches of diamondoids and their limitations.

9.1.4    CONVENTIONAL SYNTHESIS OF DIAMONDOIDS AND ITS LIMITATIONS

A brief history on the synthesis of lower diamondoids up to tetramantane is given in the paper by Schwertfeger et al. [6].

Following the fabrication of lower diamondoids, the organic synthesis of higher diamondoids was the topic of intensive efforts up to the early 1980s [29]. However, it could be demonstrated that while lower diamondoids can be synthesized by carbocation equilibration reactions, this approach does not work for higher diamondoids. The main problems encountered were the lack of suitable precursors, the rapidly growing number of possible isomers, intermediates being trapped in local energy minima, and the formation of unwanted side products. As a result, research on carbocation-mediated syntheses of diamondoids beyond tetramantane was stopped at the beginning of the 1980s [31], and up to now, only one of the higher diamondoids, [121]tetramantane, has been synthesized, but the reaction was complex and the reaction yields very low [32].

The interest in diamondoids was only renewed when it was reported that diamondoids with a number of cages (n) up to 11 were isolated from crude oil [5]. However, the quantities in which diamondoids are available decreases inversely to their size.

In order to fabricate diamondoids, especially those with a large number of cages, and to make them available in larger quantities, it is necessary to develop alternative synthesis methods that allow us to synthesize and functionalize diamondoids. Two approaches that have shown promise are electric discharge and pulsed laser plasmas generated in SCFs. In the following section, we discuss the generation of pulsed laser plasmas in supercritical fluids and their application for the fabrication of nanomaterials, especially diamondoids.

9.2    GENERATION OF PULSED LASER PLASMAS IN SUPERCRITICAL FLUIDS

9.2.1    BASIC CHARACTERISTICS AND CURRENT APPLICATIONS OF SUPERCRITICAL FLUIDS

As discussed in the previous section, one of the major obstacles in conventional organic synthesis of diamondoids is that reactions have to be realized close to thermodynamic equilibrium.

In contrast, owing to the presence of reactive species, i.e., radicals, ions, and electrons present in plasmas, reactions far from thermodynamic equilibrium can be realized. Here we will not discuss the properties of plasmas in detail, as there is a vast amount of literature available [3,4,33].

First, we will give a short introduction on supercritical fluids (SCFs), their main physical properties, and some of their current applications.

FIGURE 9.6 Phase diagram and variation of physical properties of supercritical fluids around the critical point. (a) Phase diagram of a pure substance, the domains corresponding to solid, liquid, gaseous, and supercritical phases. Above the critical point, the interface between liquid and gaseous phases disappears. The inset shows the compressible region around the critical point. (b) Density fluctuation of CO₂ near the critical point. (c) Variation of the density ρ, (d) thermal conductivity κ, and (e) thermal capacity at constant pressure C_p at a temperature of 304.432 K (T/T_crit = 1.001) for CO₂. While in the supercritical domain, the density ρ changes continuously with pressure, and the density fluctuation F_D, the thermal conductivity κ, and thermal capacity C_p reach their maximum near the critical point. (Data retrieved from NIST Chemistry Webbook, Thermophysical Properties of Fluid Systems, Technical Report, NIST, 2011.)

SCFs are defined as media in a state of temperature (T) and pressure (p) above their critical values, T ≥ T_crit and p ≥ p_crit, as shown in the phase diagram of Figure 9.6(a). While for fluids below the critical temperature and critical pressure, the liquid and gaseous phases are clearly separated, at temperatures and pressures above the critical point, the two phases cannot be discriminated anymore.

Therefore from a macroscopic point of view, SCFs represent an intermediate state between liquids and gases. As such, SCFs possess high density, high diffusivity, high solubility, and low surface tension. In addition, there is a continuity in their thermophysical properties, and their density, heat capacity, and dielectric constant can be varied continuously by changing the pressure, temperature, or both. For example, while water at ambient pressure and temperature is a polar solvent and consequently does not dissolve apolar substances easily, the dielectric constant of supercritical H₂O (scH₂O) can be varied so that nonpolar organic substances can be dissolved in scH₂O.

TABLE 9.1
Comparison of Average Physical and Transport Properties of Liquid, SCF, and Gaseous States

Property	liquid	sCf	gas
Density (kg m−3)	600–1600	200–900	0.6–2.0
Viscosity (×105 kgm−1 s−1)	20–300	1–9	1–3
Diffusion coefficient (×108 m2 s−1)	0.02–0.2	1–40	1000–4000
Thermal conductivity (×103 Wm−1 K−1)	80	20–250	4–30
Surface tension (dyne cm−1)	20–450	0	0

Note:      Most of the thermophysical properties of SCFs are intermediate between those of gases and liquids, but there are also properties that are higher in SCFs when compared to liquids or gases. For example, the thermal conductivity κ reaches a maximum near the critical point.

However, not all properties of SCFs are intermediate between those of liquids and gases. In the vicinity of the critical point, the thermal conductivity (κ), specific heat (C_p), and compressibility (β) attain a maximum (Figure 9.6(d) and (e)), and they are significantly higher compared to the respective values of the gaseous or liquid states (cf. [34,35,35,36,37] for data on thermal conductivity and specific heat).

Table 9.1 summarizes typical properties of liquid, supercritical, and gaseous phases, while Table 9.2 lists the values of critical temperature T_crit, pressure p_crit, and density ρ_crit for elements that are typically used as SC media.

In Figure 9.6(c–e), the variation of the density ρ, the thermal conductivity κ, and the heat capacity C_p of CO₂ as a function of pressure between 5.90 and 8.55 MPa (p/p_crit = 0.8 – 1.16) at a temperature of 304.432 K (T/T_crit = 1.001) are shown.

These unique physical characteristics derive from the microscopic fluid structure—molecular clustering—of SCFs. While macroscopically, a SCF appears to be homogeneous, on a microscale the structure of the fluid is very heterogeneous. Especially in close vicinity of the critical point, the competition between weak attractive forces and thermal motion leads to rapid and repeated aggregation and redispersion of clusters.

In SCFs, such interactions between atoms or molecules are mainly governed by weak van der Waals forces. Molecular dynamics studies on solvent clustering in SCF solutions have shown that the average lifetime of a cluster, i.e., the exchange of particles in a cluster shell, is on the order of picoseconds (ps; 10⁻¹² s) [40].

One consequence of molecular clustering is an increase in the density fluctuation (F_D). The density fluctuation is a measure of local density enhancements and is defined by Equation (9.1) ([41], p. 96):

TABLE 9.2
Critical Constants (Temperature T_crit, Pressure p_crit, and Density ρ_crit) of Most Common Media Used in Supercritical Applications

Chemical formula	Tcrit (k)	pcrit (mPa)	ρcrit (g cm−3)
He	5.19	0.227	0.070
Ne	44.4	2.76	0.481
Ar	150.87	4.898	0.533
Kr	209.41	5.50	0.921
Xe	289.77	5.841	1.113
H2	32.97	1.293	0.031
H2O	647.14	22.06	0.322
CO2	304.13	7.375	0.468
N2	126.21	3.39	0.311
O2	154.59	5.043	0.438

Source:  Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

Note:      Because the critical point of CO₂ is close to ambient conditions, it is currently the most commonly used SCF. Due to its very high reactivity, scH₂O is increasingly used for the oxidization of toxic waste.

$F_D=\frac{\langle \langle N\rangle \rangle }{\langle N\rangle }=\frac{\langle {\left(N-\langle N\rangle \right)}^2\rangle }{\langle N\rangle }$

(9.1)

where N is the total number of particles in the system and 〈N〉 is the average number of particles. For estimating F_D, one can use its relation to the isothermal compressibility given by

$Z_T=-\frac{1}{V}{\left(\frac{\partial V}{\partial p}\right)}_{T,N}$

(9.2)

and the isothermal compressibility of a perfect gas

$Z_T^0=\frac{1}{n{k}_{B}T}$

(9.3)

where n = 〈N〉/V and one can show that [41]

$F_D=\frac{Z_T}{Z_T^0}$

(9.4)

The plot in Figure 9.6(b) shows the variation of F_D for CO₂ as a function of pressure and temperature near the critical point. Near the critical point, the value of F_D reaches a maximum. Besides this maximum, the region of increased F_D also extends into both liquid and gaseous regions around the critical point. A consequence of the increase of the density fluctuation is that around the critical point, the compressibility of a SCF is greatly enhanced.

The unique properties of SCFs make them attractive for a wide range of applications. For instance, the solvating power of SCFs permits replacement of environmentally much more harmful liquid organic solvants by supercritical CO₂ (scCO₂) or water (scH₂O). In addition, in general SCFs are noncarcinogenic, nontoxic, nonmutagenic, nonflammable, and thermodynamically stable [2].

Owing to their temporal and spatial density fluctuations, high solubility, and density, SCFs are used as solvents in a variety of chemical processes, extraction of components from petroleum using supercritical pentane, or extraction of drugs from plant materials [42].

The use of SCFs may be advantageous for reactions involved in fuels processing, biomass conversion, biocatalysis, homogeneous and heterogeneous catalysis, environmental control, polymerization, and materials and chemical synthesis [43].

In the vicinity of the critical point, where the thermal conductivity [34,35,36] and specific heat [37] attain their maximum values, it is expected that new types of reactions involving clusters can be used, and that by changing the conditions of the SCF medium, the selectivity of reactions can be adjusted.

Furthermore, it has been shown that the use of media in a supercritical state near the critical point can lead to enhanced reaction rates and selectivity of chemical reactions. This has been attributed to the enhanced molecular clustering and density fluctuation near the critical point [44,45,46].

Because of their high dissolving power, SCFs have also shown excellent potential for chemical analysis applications. Supercritical fluid chromatography (SFC), using scCO₂ as a mobile phase, is especially suited for the separation of compounds that are sensitive to organic solvents commonly used in HPLC [47,48]. SFC often shows superior performance compared to conventional HPLC because, besides avoiding the use of organic mobile phases such as acetonitrile, the diffusion coefficients are an order of magnitude higher than in liquids (cf. Table 9.1). Consequently, the transfer of solutes through the separation column encounters less resistance, with the result that separations may be realized more rapidly or with higher resolution in comparison to HPLC.

In addition to chemical synthesis and as solvents in chemical analysis, SCFs have also shown promising potential for use in materials processing.

Among the main commercial applications of SCF media are textile dyeing and food processing, e.g., for the decaffeination of coffee [49,50]. Furthermore, in recent years, the use of SCFs, in particular scCO₂, for reducing water and organic solvents in the microelectronics industry has been investigated [51].

Because of the high dissolving power, scCO₂ can be used for photoresist stripping or for the cleaning of substrates. For instance, Namatsu et al. [52,53] demonstrated the drying of microstructured silicon and photoresist patterns without pattern collapse. The pattern collapse in high aspect ratio Si patterns is avoided owing to the low surface tension of the supercritical medium, which reduces the stresses due to capillary forces on opposite trenches.

Another application of SCFs in materials processing is chemical fluid deposition (CFD). CFD involves the reduction of organometallic compounds that are soluble in SCFs, such as scCO₂, using hydrogen, which is soluble in scCO₂, as a reducing agent. Owing to the low surface tension, CFD allows deposition of conformal thin films of various materials. Compared to thermal CVD where typically temperatures of about 600°C are required, CFD processes can be realized at more modest temperatures. In addition, CFD allows a wider choice of precursors, the reason being that in contrast to CVD, where all species have to be in the gas phase, also precursors with lower vapor pressures can be used. Thus far CFD has been used to deposit Pt films by hydrogenolysis of dimethyl-(cyclooctadiene) platinum(II) (CODPtMe₂) on Si, on polymer substrates, and in nanoporous Al₂O₃ in scCO₂ (p = 15.5 MPa) at modest temperatures of 80°C [54]. Blackburn et al. [55] used CFD for the conformal deposition of Cu and Ni films in high aspect ratio trenches etched in silicon. It has been shown that besides noble and near-noble metals (Cu, Ni, Pt, Pd, Ru), CFD could also be used to deposit alloys and oxides such as RuO_x, TiO₂, Al₂O₃, and ZnO [56].

In recent years, SCFs have been used increasingly for the design of advanced nanostructured materials [57]. In particular, SCFs have been used for the production of a wide range of nanomaterials or the functionalization of nanomaterials: nanofibers, nanorods, nanowires, nanotubes, conformational films, core-shell structures, supported nanoparticles, polymers impregnated with nanoparticles, and organic coatings of particles [58]. Such SCF-based technologies have also been proposed for polymer processing for pharmaceutical and medical applications [59].

All these applications that take advantage of the specific properties of SCFs have turned out to be generally more flexible, simpler, and with a reduced environmental impact because the SCF can be completely eliminated at the end of the process.

However, the main drawback of chemical synthesis in SCFs is that they have to be conducted near thermodynamic equilibrium, which often makes it necessary to work at high temperatures. In the next section, it will be shown that plasmas generated in SCFs offer the possibility of realizing reactions far from thermodynamic equilibrium. This permits the synthesis of nanomaterials that cannot be obtained easily by conventional synthesis techniques in SCFs.

9.2.2    GENERATION OF PULSED LASER PLASMAS IN SUPERCRITICAL FLUIDS

While SCFs show higher reaction rates than synthesis in the liquid or gas phase, it has been suggested that the combination of plasmas with SCFs could lead to even higher efficiency for realizing chemical reactions. The reason is the formation of highly reactive species, ions, electrons, and radicals in the plasma that would allow realization of reactions far from thermodynamic equilibrium.

There are two main approaches for generating plasmas in SCFs: The first is by electric discharges, and the second by pulsed laser ablation.

It has been shown that plasmas generated in SCFs show interesting properties. For instance, it has been discovered that by using very small electrodes with gap distances below 10 μm, the breakdown voltage reaches a minimum near the critical point [60]. This behavior is contrary to what could be expected from classic discharge theory. This decrease of the breakdown voltage has been the same irrespective of whether it was an atomic gas such as Xe, a nonpolar molecule such as CO₂, or a polar molecule such as H₂O [61].

Following these experiments, microplasmas in SCFs, mainly scCO₂ and scXe, have been used for the fabrication of nanomaterials. For example, sp²-hybridized carbon nanomaterials, e.g., CNTs, nanohorns, and nano-onions, were synthesized by generating either short pulse direct-current (pulse duration: 400 μs) plasmas or low-temperature dielectric barrier discharges (DBDs) in neat scCO₂ [62,63,64].

In contrast to SCF processes that were used for the synthesis of CNTs [65,66] and fullerenes [65], conducted at either high temperature or high pressure and using catalysts, plasmas generated in scCO₂ allowed fabrication of CNTs at milder conditions (i.e., temperatures close to room temperature) and without the use of any catalyst.

However, while plasmas in SCFs enable reactions without any catalysts and far from thermodynamic equilibrium, the generation of electric discharges in very high pressure or even SCFs is not very straightforward. The breakdown voltage depends a lot on the electrode geometry (electrode gap, dielectric layer thickness) and the conditions (pressure, temperature, and composition) of the SCFs. In contrast, pulsed lasers allow the formation of dense plasmas under various conditions, from high vacuum, liquid, and SCFs.

By using pulsed lasers, the generation of a plasma in high-pressure or supercritical conditions is easier and more independent of the medium.

Plasmas can be generated either by irradiation of a solid target or by inducing breakdown of the supercritical medium by focusing the laser beam in a small spot.

Pulsed laser ablation (PLA) by plasmas that are generated by irradiating a solid target has been used since the appearance of Ruby lasers in the early 1960s. The PLA process was studied in detail for the first time by Patil and coworkers [67]. They found that the important chemical reactions take place at the interface between the plasma plume and surrounding medium [68].

There are two main mechanisms of plasma formation by PLA: The first is by nanosecond lasers, while the second is for femto- or picosecond lasers. It is generally a highly nonequilibrium process, and the heating, forming of the plasma plume, and material ejection occur after the laser pulse [69,70,71,72].

In Figure 9.7, the mechanisms leading to plasma formation by irradiation of a target by different types of pulsed lasers are shown. While the detailed mechanisms are still not understood yet, depending on the type of laser, continuous wave (CW), nanosecond, or pico- or femtosecond, the mechanisms leading to the removal of material or plasma formation are different. In the case of CW lasers, material is removed primarily by melting, which creates a large heat-affected zone (HAZ), and material ejection is mainly dominated by thermal processes [73]. In nanosecond lasers, there are three main stages that lead to the formation of a plasma. In the first, laser photons couple with both electrons and phonons of the target material. The photon-electron coupling then results in an immediate rise of the electron temperature, leading to vaporization of the target. Compared to CW lasers, the HAZ created by nanosecond pulsed lasers is smaller.

FIGURE 9.7 Mechanisms of plasma formation and material removal by pulsed laser ablation. Depending on the type of laser, the interaction between the laser light and the target material varies. CW lasers remove material primarily by melting, which creates a large heat-affected zone (HAZ). Nanosecond laser pulses create a smaller HAZ, and material is removed by melt expulsion driven by the vapor pressure and recoil pressure. With ultrafast pulses of the order of pico- or femtoseconds, the laser pulse duration is much shorter than the timescale for energy transfer between free electrons and the material lattice. This leads to very high pressures and temperatures limited to a very shallow zone of only a few micrometers, where the material is vaporized immediately.

With ultrafast pico- and femtosecond pulses, the laser pulse duration is much shorter than the timescale for energy transfer between free electrons, and the material lattice and electrons are excited to only a few or few tens of electron volts. Consequently, the lattice temperature of the target remains unchanged, and the main amount of the laser pulse energy is primarily absorbed in a thin layer of only a few microns close to the surface, where extremely high pressures and temperatures can be attained. The absorbed energy heats the material very quickly past the melting point, directly to the vapor phase with its high kinetic energy, and the material is removed by direct vaporization. Consequently, in the case of pico- and femtosecond pulsed lasers, mainly the photon absorption depth governs the heated volume, the influence of thermal diffusion depth being smaller.

With nano-, pico-, and femtosecond pulsed lasers becoming more and more available, PLA has been gaining increasing attention due to its potential for a wide range of materials processing: deposition of thin solid films, nanocrystal growth, surface cleaning, and the fabrication of microelectronic devices.

Besides the generation of pulsed laser plasmas in vacuum, pulsed laser ablation plasmas have recently found increased use in liquids, for both the synthesis and functionalization of nanomaterials [74].

Pulsed laser ablation in water has also been used for the synthesis of nanodiamonds [75]. The nanodiamond particle consists of a diamond core with diameters ranging from 5 to 15 nm, surrounded by a graphitic shell with a thickness ranging from 3 to 4 nm. It was found that the synthesis of nanodiamonds occurs in a very narrow temporal window, at the early stages of plasma formation. The shock wave duration was discovered to be roughly twice the pulse duration (~10 ns), during which the pressure inside the shock wave reached values between 4.5 and 22.5 GPa. After the shock wave subdued, the pressure values were too low for generating diamondoids, and only sp² carbon was formed.

FIGURE 9.8 Experimental setup for realizing diamondoid synthesis by pulsed laser ablation in scCO₂ and scXe. (a) Schematic of the experimental setup consisting of a stainless steel high-pressure cell, cooling/heating circuit, liquefaction loop (typically cooled by liquid nitrogen) for condensing the gas, and pressure and temperature indicators (PI and TI). After the end of the experiment, the chamber is purged and the synthesized materials collected in a trap connected to the outlet. (b) Top view of the high-pressure cell and cooling/heating circuit that allows adjustment of the temperature in the inner cell. (c) Side view of the high-pressure cell. The pulsed laser light is focused on a target through a sapphire window, and the high intensity of the laser leads to the rapid heating of a zone close to the laser spot and formation of the plasma. The schematics of the high-pressure cell were provided by Taiatsu Techno Co.

For generating plasmas in SCFs, the use of high-pressure cells that can withstand the high-pressure of SCF media is necessary. The most common high-pressure cells are fabricated out of stainless steel, but for applications needing optical access, cells consisting of quartz are used. When using highly reactive SCFs such as supercritical water, special alloys, e.g., hastelloy, are employed.

A schematic of a high-pressure cell that was used for realizing pulsed laser plasmas in scCO₂ and scXe for the synthesis of diamondoids is presented in Figure 9.8.

The high-pressure stainless steel cell consists of two parts, an inner cell that contains the pressurized gas and an outer cell that is connected to a water-circulating circuit and that allows adjustment of the temperature in the inner cell. The temperature and pressure in the inner cell are monitored by a thermocouple and pressure sensor, respectively. A window made out of sapphire is placed on top of the cell for optical access and for focusing the laser beam on the surface of the target that has been placed inside the high-pressure cell.

For condensing Xe and to reach SCF conditions, the gas was flown through a liquefaction loop that was cooled by liquid nitrogen. In the case of CO₂, the gas was compressed using a high-pressure pump. After the experiment, the synthesized materials have to be collected in a trap connected to the outlet of the high-pressure cell.

9.2.3    APPLICATION OF PULSED LASER PLASMAS IN SUPERCRITICAL FLUIDS TO NANOMATERIALS SYNTHESIS

In contrast to PLA in liquids, PLA in SCFs has not been used as frequently up to now. Probably the main difficulty lies with the necessity to use chambers that can withhold the high pressures of SCFs.

While xenon and CO₂ are the two gases whose critical temperatures are closest to room temperature, Xe is about 1000 times more expensive than other SCF media, such as Ar and CO₂ [2]. This makes Xe less attractive as a supercritical medium for both science and industry. On the other hand, scCO₂ is one of the most widely used SCFs, having already found applications in various fields. Because of this, most of the PLA in SCFs so far is based on using scCO₂.

As one example of PLA in SCF, the fabrication of gold nanoparticles by pulsed laser ablation in scCO₂ has been investigated [76]. In addition to noble metal nanoparticles, the same authors also investigated the formation of silicon nanocrystals (Si-nc) by pulsed laser ablation of Si targets [77]. The authors show that the size and crystal structure of the Si-nc (core vs. shell) could be changed by adjusting the density of the SCF medium.

Sasaki and Aoki [78] investigated the effect of the pressure of the supercritical medium on the formation of nanoparticles. They found that the quenching rates near the critical point reach a maximum. Therefore the conditions of the SCF can be adjusted in such a way as to optimize the size of the nanomaterials.

In short, while compared to liquids, PLA in SCFs offers finer tuning of the medium, and therefore can be used to adjust the characteristics of the nanomaterials formed, it is still not widely used. In the next section we describe the application of PLA in scXe and scCO₂ for the synthesis of diamondoids.

9.3    SYNTHESIS OF DIAMONDOIDS BY PULSED LASER PLASMAS

Despite the intricacy of synthesizing higher diamondoids consisting of a minimum of four cages, two groups independently reported the synthesis of higher diamondoids using lower diamondoids as precursors using electric discharge plasmas generated in supercritical Xe (scXe) [79,80,81] and by hot-filament CVD [82].

In these previous studies, dielectric barrier discharges (DBDs) [79,80,81] were used to produce a highly dense plasma reaction field in scXe. Higher diamondoids consisting of up to 10 cages were synthesized by the generation of plasmas in supercritical Xe (scXe; T_crit = 289.7 K, p_crit = 5.84 MPa) with dissolved adamantane (n = 1) as a precursor. The key factors for the artificial synthesis of diamondoids were considered to be the high density and highly nonequilibrium reaction field of the plasmas generated in the SCF media, especially in the vicinity of the critical point, and the use of dissolved adamantane as a precursor.

TABLE 9.3
Experimental Conditions for the Pulsed Laser Ablation in scCO₂ with and without Cyclohexane

Parameter	scCo2	scCo2 with Cyclohexane
Adamantane concentration (mol/L)	(1.2–12) × 102	(3.1–6.2) × 102
Cyclohexane concentration (mol/L)	—	7.7 × 101
Temperature T (K)	304.5–305.2	304.5–304.8
Reduced temperature (T/Tcrit)	1.001–1.004	1.001–1.002
Pressure p (MPa)	7.42–7.59	7.54–7.56
Reduced pressure (p/pcrit)	1.005–1.028	1.022–1.024

Source:  Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

Note:      The addition of cyclohexane facilitates the dissolution of the precursor adamantane in the scCO₂.

9.3.1    EXPERIMENTAL PROCEDURE

An HOPG target (mosaic spread: 3.5°–5°) was placed in a batch-type high-pressure cell (inner volume: 3 cm³), as described in Figure 9.8. Adamantane (purity > 99.0%; Tokyo Chemical Industry) was then dissolved in scCO₂ with and without cyclohexane as a cosolvent; see the experimental conditions listed in Table 9.3.

The high-density CO₂ was realized by condensation of the CO₂ gas inside a liquefaction loop cooled by liquid nitrogen before introducing it into the inner cell. A second-harmonic neodymium-doped yttrium aluminum garnet (Nd:YAG) laser (λ = 532 nm) with a pulse width of 7 ns at a repetition rate of 10 Hz was used for the ablation of the HOPG targets in both scXe [84] and scCO₂. The maximum energy was 7 mJ per pulse, giving a maximum fluence of approximately 18 J/cm² on the surface of the target. On average, the ablation experiments were conducted for periods ranging from 60 to 240 min. The synthesized products were retrieved by evacuating the gas through a collection trap containing 10 ml of cyclohexane.

In addition, the cyclohexane solutions used for cleansing the cell were also collected. The HOPG target was dipped in the collection trap and sonicated. The concentration of the products was then increased by partial evaporation of the collected organic solvents.

Following the procedure established for the isolation of higher diamondoids [5], to remove non-diamondoids and to improve the analysis of the diamondoids by gas chromatography–mass spectrometry (GC-MS), volatile compounds, and non-diamondoids, the samples were treated by pyrolysis in Ar [83].

After retrieval and concentration of the synthesized products, micro-Raman spectroscopy and GC-MS measurements were conducted for as-collected samples, and both micro-Raman spectroscopy and GC-MS measurements were also conducted for the sample after pyrolysis.

Samples of the synthesized products for assessment by micro-Raman spectroscopy were prepared by spotting small volumes (~ 0.1 ml) of the suspensions on aluminum substrates. Micro-Raman spectroscopy was performed using a second-harmonic Nd:YAG laser (λ = 532 nm) with an excitation power of 12 mW and a spatial resolution of around 1 μm. Spectra were measured using an acquisition time of 1 min. For the GC-MS measurements, the GC was set up in splitless sample injection mode, which, compared to split injection, allows analysis of very small quantities of sample. The MS measurements were conducted by electron-impact mass spectrometry at 70 eV, in both selected ion monitoring (SIM) mode and scan mode on the products dissolved in 1 L of cyclohexane. SIM ions included combinations of m/z varying from 188 to 590 that could correspond to diamondoids with cage numbers varying between n = 2 and 12. Mass scans were acquired over an m/z range from 40 to 1000 [83].

9.3.2    MICRO-RAMAN SPECTROSCOPY

The characteristics of the Raman spectra of diamondoids can be summarized by dividing them into three regions: (1) strong peaks in the low-energy region below 500 cm⁻¹, due to vibrational modes of the CCC structure, such as CCC bend and deformation modes; (2) well-resolved peaks in the range 1000–1500 cm⁻¹, due to CH twist and wagging modes; and (3) strong and broad peaks in the high-energy region [85]. For the high-energy region, it is reported that the stretching mode of sp^3- bonded CH_x shows peaks in the region of 2800–2950 cm⁻¹, while sp²-bonded CH_x shows peaks in the region of 2980–3060 cm⁻¹ [86]. Comparing the measured spectrum with these characteristics might help to identify the synthesized products, even though these characteristics are common in not only diamondoids but also other hydrocarbons, such as paraffin [87]. In Figure 9.9, the Raman spectra of the products obtained by PLA in scCO₂ with and without cyclohexane as a cosolvent (both before and after pyrolysis) are presented together with the Raman spectrum of the material obtained by PLA in scXe [84].

For illustrating the effect of increasing cage numbers on the Raman spectra, in Figure 9.9(f) and 9(g) the spectra of commercial adamantane and [12312] hexamantane [85] are shown. The Raman spectra of the products obtained by PLA in scCO₂, regardless of the presence of cyclohexane, present almost the same features that are known to be characteristics of Raman spectra of diamondoids, i.e., well-resolved peaks in the range 1000–1500 cm⁻¹ and strong peaks in the high-energy region, 2800–2950 cm⁻¹. On the other hand, the peaks in the measured Raman spectra are supposed to originate from a mixture of different hydrocarbons. Moreover, it is known that the Raman scattering signal varies largely for each material; e.g., sp^2- hybridized carbon shows a much stronger Raman signal than that of sp³-hybridized carbon [86,88]. Therefore it is possible that the peaks from synthesized diamondoids and those from other hydrocarbons, such as paraffin and aromatics, get balanced out, which could explain the lack of strong peaks in the low-energy region below 500 cm⁻¹. The features in the wavenumber range between 100 and 1500 cm⁻¹, and the peaks in the range between 2800 and 2950 cm⁻¹, which are characteristic of sp³ CH_x stretching modes [86], indicate the presence of diamond-structured hydrocarbons in the synthesized products. The existence of diamond-structured hydrocarbons in the synthesized products was also indicated for PLA in scCO₂, both with and without cyclohexane. In all the Raman spectra of the synthesized materials, no indication of the precursor adamantane could be found. This can be explained by the fact that adamantane sublimes even at room temperature, and that for increasing the concentration of the samples, the solutions collected after PLA were evaporated. While the Raman spectra collected from the products obtained by PLA in scCO₂ with and without cyclohexane showed similar characteristics, smaller differences that are found in the Raman spectra for the samples obtained in scCO₂ with cyclohexane as a cosolvent can be attributed to the possible synthesis of species formed by the dissociation of cyclohexane. In contrast, the differences between the products obtained by PLA in scCO₂ and scXe [84] were larger. For example, the strongest intensities in the low-energy region (300–1500 cm⁻¹) appear around 1440 cm⁻¹ for PLA in scCO₂ with and without cyclohexane, while the most intense Raman peaks appear around 1050 cm⁻¹ for PLA in scXe. Moreover, the shapes of the strong peaks in the high-energy region (2800–3000 cm⁻¹) vary considerably. These dissimilarities indicate that the synthesized materials are different for PLA in scCO₂ and scXe. The effects of pyrolysis for the samples obtained by PLA in scCO₂ and in scCO₂/cyclohexane are shown in Figure 9.9(a–d). Compared to the spectra before pyrolysis, the biggest difference can be observed in the high-wavenumber region between 2980 and 3000 cm⁻¹. The intensity of the peaks that are attributed to sp² CH_x stretching modes [86] is lower after pyrolysis. In addition, the peaks in the high-wavenumber region become better resolved. This shows that pyrolysis permits reduction of the number of most non-diamondoid hydrocarbons. However, as-collected products from plasmas generated in SCFs are typically a mixture of hydrocarbons, including many kinds of non-diamondoid products, and consequently, micro-Raman spectra of as-collected samples contain peaks of mixed materials. Accordingly, the analysis of the micro-Raman spectra only allows a confirmation of the presence of diamondoids. Therefore except for pure, recrystallized samples obtained by isolation and purification, it is not possible to identify individual diamondoids.

FIGURE 9.9 Micro-Raman spectra of synthesized products obtained by PLA and compared to spectra obtained on diamondoids isolated from crude oil. (a) Products synthesized in scCO₂ before pyrolysis. (b) Spectrum obtained on products after pyrolysis. (c) In scCO₂ with cyclohexane before pyrolysis. (d) After pyrolysis. (e) In scXe before pyrolysis [83]. For illustrating the effect of increasing cage number, the Raman spectra of (f) commercial adamantane and (g) [12312] hexamantane (taken from [85]) are included as examples. (f) The three strongest Raman active vibration modes for adamantane, corresponding to C-C breathing, CH wag/CH₂ twist and CH_x stretching modes, respectively, are also indicated. (Adapted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.)

9.3.3    GAS CHROMATOGRAPHY–MASS SPECTROMETRY

The technique of choice for the identification of individual synthesized diamondoids is gas chromatography–mass spectrometry (GC-MS). The method allows us to both separate individual diamondoids as a function of the temperature ramp of the GC and identify individual diamondoids or diamondoid derivatives. The main difficulty is to achieve conditions where the diamondoids are put into the gas phase. Consequently, relatively high temperature ramps of up to 360°C are necessary for being able to detect higher diamondoids. Typical features of unsubstituted higher diamondoids found in GC-MS analysis are a strong molecular ion peak M⁺ and a few fragment ions, including one at m/z M⁺/2, which corresponds to the doubly charged molecular ion.

9.3.3.1    Synthesis of Diamantane

In order to prove the synthesis of diamondoids from the precursor adamantane by PLA in scXe and scCO₂, the retention times for m/z 136 and m/z 188, which correspond to the molecular weights (MWs) of adamantane and diamantane (n = 2), respectively, were compared for products obtained by PLA in scCO₂ with and without cyclohexane, and a reference solution containing dissolved commercially available adamantane and diamantane. The comparison revealed the same time lag, 3.5 min, for all experimental conditions. As shown in Figure 9.10, the mass spectra of the GC-MS measurements at the retention times of SIM for m/z 188 for both products from PLA in scCO₂ with and without cyclohexane present almost the same fragments as those of the standard diamantane.

FIGURE 9.10 C-MS mass spectra acquired at a retention time of t = 7.3 min. (a) Synthesized product obtained in neat scCO₂. (b) Product synthesized in scCO₂ and cyclohexane. (c) Mass spectrum of diamantane of the standard sample. The inset shows the molecular structure of diamantane. The main features of the mass spectra are the molecular ion peak at m/z 188 and fragment peaks that are characteristic of diamantine. (Adapted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.)

To estimate the production rate of diamantane, a calibration curve has to be used. For this, cyclohexane solutions with varying diamantane concentrations are prepared and the SIM peak areas corresponding to diamantane are measured for each concentration. By comparing the SIM peak areas of the synthesized products, the production rate of diamantane can then be estimated. In the case of the PLA experiments conducted in scCO₂ both with and without cyclohexane, the estimated production rate was 0.2 μg/h [83] (see Table 9.4).

9.3.3.2    Synthesis of Diamondoids with n ≥ 3

In a first approximation, the larger the diamondoid, the longer the retention time. SIM curves and mass spectra indicated the possible synthesis of diamondoids consisting of three or more cages by PLA in scCO₂. Besides triamantane (n = 3), peaks were observed that could be attributed to diamondoids consisting of up to 12 cages. The cage numbers of eluted diamondoids obtained by PLA in scCO₂ with and without cyclohexane as a function of the experimentally determined GC-MS retention times are listed in Tables 9.5 and 9.6, and Figure 9.11, respectively.

The relative retention times were calculated by dividing the individual retention times by the retention time of diamantane in the range 7.148–7.157 min. The GC-MS signals could be assigned to the following diamondoids: triamantane at m/z 240; tetramantane at m/z 292; hexamantanes at m/z 342, 382, and 396; heptamantane at m/z 394; octamantane at m/z 446; decamantanes at m/z 456 (C₄₅H₄₆) and 590; and dodecamantane at m/z 586. From Figure 9.11 one can see that the GC-MS retention times increase almost linearly with the number of cages, which is in agreement with previous work [89]. Furthermore, for the relative retention times of the diamondoids obtained by PLA in scCO₂ both with and without cyclo-hexane, the authors found good agreement: triamantane (1.188–1.189), tetramantane (1.509–1.510), hexamantanes (1.986–1.987 for a MW of 382 and 2.038–2.039 for a MW of 396), and dodecamantane (2.493–2.495). In addition, some of the observed diamondoids, corresponding to MWs of 382, 586, and 590, were not found in previous works in which the diamondoids were isolated from petroleum [5]. These molecules were detected at high GC temperature, higher than the maximum GC temperature used by another group (593 K). For example, molecular ion peaks corresponding to a MW of 586 appeared when the GC oven temperature was raised to 620 K. In GC-MS measurements, each diamondoid is expected to elute at a distinct retention time. In contrast, for higher diamondoids, there are MW groups that have many structural isomers. For example, MW 292 (tetramantane) has possible isomers and MW 396 (hexamantane) has 28. Therefore in the GC-MS measurements, diamond molecules corresponding to the same MW group but possessing different structures are expected to elute at different retention times.

TABLE 9.4
SIM Peak Area and Estimated Diamantane Concentration of the Product Obtained by PLA in scCO₂

Supercritical Medium for PLA	Solution for GC-MS Analyses	SIM Peak Area (Arb. Unit)	Diamantane Concentration (mg/L)
scCO2	Sample in 1 ml cyclohexane	2100	0.23
scCO2 with cyclohexane	Sample in 1 ml cyclohexane	1900	0.21

Source:  Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

TABLE 9.5
List of Possibly Synthesized Diamondoids by PLA in scCO₂, with Their Molecular Formulae, MWs, GC-MS Relative Retention Times, and Dual Graph Codes

Cage Number (n)	Molecular Formula	M+ (m/z) Base Peak	GC-MS Relative Retention Time	Structure
2	C14H20	188	1.000	Diamantane
3	C18H24	240	1.189	Triamantane
4	C22H28	292	1.509	E.g., [121]
6	C29H34	382	1.986	E.g., [12(1)31]
	C30H36	396	2.038	E.g., [12121]
7	C30H34	394	1.843	E.g., [123121]
10	C35H36	456	2.111	[1231241(2)3]
12	C45H46	586	2.495	E.g., [12131431234]

Source: Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

TABLE 9.6
List of Possibly Synthesized Diamondoids by PLA in scCO₂ with Cyclohexane as a Cosolvent, with Their Molecular Formulae, MWs, GC-MS Relative Retention Times, and Dual Graph Codes

Cage Number (n)	Molecular Formula	M+ (m/z) Base Peak	GC-MS Relative Retention Time	Structure
2	C14H20	188	1.000	Diamantane
3	C18H24	240	1.188	Triamantane
4	C22H28	292	1.510	E.g., [121]
6	C26H30	342	1.627	[12312]
	C29H34	382	1.987	E.g., [12(1)31]
	C30H36	396	2.039	E.g., [12121]
8	C34H38	446	2.156	E.g., [1213(1)21]
10	C45H50	590	2.423	E.g., [12(3)1(2)3(1)23]
12	C45H46	586	2.493	E.g., [12131431234]

Source:  Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

FIGURE 9.11 Variation of the number of diamondoid cages obtained by scCO₂ with and without cyclohexane as a function of the GC-MS relative retention time. The retention time of diamantane (t = 7.148–7.157 min) is taken as a reference retention time, and the grayscale map indicates the molecular weights (MWs) of the detected diamondoids. Diamondoids with higher molecular weights need increasingly longer elution times for being detected by MS, the increase is almost linear.

FIGURE 9.12 C-MS mass spectrum of possible superadamantane (MW 456) obtained by PLA in scCO₂. The inset shows the molecular structure of superadamantane. (Adapted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.)

Depending on the setting of the heating cycle and the resolution of the GC-MS, it is expected that different isomers can appear at the same scan times, even if they elute at different times. Slower heating rates or faster scan rates can help to improve the resolution of GC-MS analyses. Figure 9.12 shows an example of mass spectra corresponding to a MW of 456. This diamondoid is the most compact with n = 10, [1231241(2)3] decamantane, also known as superadamantane.

Strong peaks at low m/z in Figure 9.12 are supposed to be due to either other hydrocarbon molecules such as paraffin and aromatics that co-elute at the same retention time, or other materials that originated from the GC column and the septum that appear in the entire GC-MS analysis.

Moreover, non-diamondoids and other unresolved complex mixtures co-eluting with the diamondoids affect the resolution of GC-MS analyses [90].

Other studies on the isolation of diamondoids from petroleum reported that the extremely high thermal stability of diamondoids can be used to separate them from non-diamondoids [5]. All higher diamondoids are thermally stable up to a temperature of 723 K, which is high enough to decompose most other non-diamondoid hydrocarbons [91].

9.3.4    EFFECTS OF PYROLYSIS ON THE SYNTHESIZED PRODUCTS

The mass spectra acquired after the pyrolysis experiments and corresponding to possible higher diamondoids are listed in Tables 9.7 and 9.8.

The observed peaks were different before and after pyrolysis. However, the molecular ion peak with a MW of 456 was observed both before and after pyrolysis. Moreover, the relative retention times of GC-MS were very close to each other. This result suggests that the synthesized products with a MW of 456 survived after pyrolysis. It is expected that non-diamondoid hydrocarbons of high MW will decompose after thermal treatment at 723 K. Therefore the present results suggest that the peak corresponding to a MW of 456 originated from superadamantane. Many MW groups of diamondoids, such as those with n = 5 and 7 cages, were not found before pyrolysis, although they were found after the pyrolysis experiments (see Figure 9.13).

TABLE 9.7
List of Observed Higher Diamondoids for PLA in scCO₂ after Pyrolysis

Cage Number (n)	Molecular Formula	M+ (m/z) Base Peak	GC-MS Relative Retention Time
5	C25H30	330	1.702.25
5	C26H32	344	2.302.31
7	C30H34	394	2.02
7	C34H40	448	2.45
9	C34H36	444	1.66
9	C37H40	484	2.20
10	C35H36	456	2.122.13
10	C44H48	576	2.57

Source:  Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

TABLE 9.8
List of Observed Higher Diamondoids for PLA in scCO₂ with Cyclohexane after Pyrolysis

Cage Number (n)	Molecular Formula	M+ (m/z) Base Peak	GC-MS Relative Retention Time
4	C22H28	292	1.34
5	C26H32	344	1.83
6	C30H36	396	1.67
7	C32H36	420	2.01
7	C34H40	448	2.45

Source:  Reprinted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.

In scCO₂, species originating from the decomposition of CO₂ could possibly participate in the formation of new materials. Assuming those materials were non-diamondoid hydrocarbons and were decomposed after pyrolysis, this might explain why the observed spectra contained peaks corresponding to diamondoids with five and seven cages only after pyrolysis.

FIGURE 9.13 C-MS mass spectra of possible diamondoids after pyrolysis: MW 448 (heptamantane, C₃₄H₄₀) obtained by PLA in scCO₂ with cyclohexane. The inset shows the molecular structure of [121212] heptamantane, which is one of the possible isomers possessing MW 448. (Adapted with permission from S. Nakahara et al., J. Appl. Phys., 109(12), 123304, Copyright 2011, American Institute of Physics.)

9.3.5    COMPARISON BETWEEN PLA IN scCO₂ AND SCXE

There are MW groups that were only synthesized in scXe and not in scCO₂ [84]. Moreover, the production rate of diamantane of PLA in scCO₂ was 0.2 μg/h, which is lower than that of PLA in scXe. This could be attributed to the lower solubility of adamantane in scCO₂ than that in scXe. The production rate of diamantane is approximately the same in neat scCO₂ and in the scCO₂/cyclohexane mixture. Therefore while the solubility of adamantane in scCO₂ can be enhanced using cyclo-hexane as a cosolvent, this is probably not the only factor for increasing the production rate. In the case of scXe, the molecules of the medium do not participate in the reaction, whereas in the case of scCO₂, CO₂ itself can also dissociate and form reaction products. Consequently, it is suspected that the lower reaction yield of diamantane in scCO₂ is due to both the lower solubility of adamantane and competing reactions resulting in the formation of non-diamondoids. In contrast, the mass spectrum showing a MW of 456 was only obtained by PLA in pure scCO₂. C-C and C-H bonds must be dissociated to allow the formation of successive diamondoids from adamantane. In the previous work where diamondoids were synthesized by DBD in scXe [80], it was suggested that the absence of oxidants resulted in a preferential dissociation of C-C bonds and led to the synthesis of diamondoids with a high H/C ratio. On the contrary, the products obtained from PLA in scCO₂ included relatively low H/C ratio diamondoids compared with those in scXe. It is supposed that oxidant species originating from scCO₂ might lead to selective dissociation of C-H bonds, enabling the synthesis of low H/C ratio diamondoids, such as superadamantane.

9.4    CONCLUSIONS AND PERSPECTIVES

Here we have reviewed diamondoids, their structure, and physical and chemical properties and given a short overview of their current main and possible future applications. While smaller diamondoids up to tetramantane can be synthesized, synthesis beyond tetramantane has turned out to be not possible by conventional organic chemical methods.

Plasmas generated in high-pressure or even supercritical media present a very reactive environment that allows the fabrication of nanomaterials far from thermodynamical equilibrium.

Compared to electric discharges, the generation of pulsed laser plasmas is relatively easy. PLA was performed in both scXe and scCO₂, with and without cyclohexane as a cosolvent, for the synthesis of diamondoids. Raman spectra of the synthesized products point to the presence of sp³-hybridized materials, including diamondoids. The GC-MS measurements indicate the synthesis of diamantane and possibly other, higher-order diamondoids, including those with a number of cages larger than reported so far (n = 12) and superadamantane (n = 10). Because oxidant species originating from scCO₂ might lead to the selective dissociation of C-H bonds, synthesis of low H/C ratio higher-order diamondoids, such as superadamantane, could be realized. Moreover, there were more newly found higher-order diamondoids with n = 5–10 for PLA in scCO₂. It is expected that the fraction of non-diamondoid hydrocarbons might be effectively removed by pyrolysis. Therefore PLA in scCO₂ is considered to be a promising approach for the facile synthesis of higher-order diamondoids.

As a final conclusion, what are the perspectives for the use of PLA in SCFs for nanomaterials processing, in particular, diamondoids?

In comparison to electric discharges, PLA in SCFs allows generation of plasmas more easily, and it is less dependent on the conditions of the supercritical medium. In combination with SCFs, such plasmas permit realization of reactions that might not be possible to achieve with conventional chemical synthesis methods. The main drawbacks of the PLA approach are the difficulties related to scaling up, but this could be alleviated with further advances in laser research and the possible availability of smaller, cheaper pulsed laser systems.

Another difficulty is that this type of approach needs two main steps, synthesis and separation, of the products. However, one could imagine a continuous flow process, where the separation, isolation, and purification of the products is done directly after the synthesis.

In summary, we are convinced that the combination of PLA in SCFs offers many possibilities for the synthesis of not only diamondoids, but also other classes of nanomaterials.

ACKNOWLEDGMENTS

This work was supported financially in part by a Grant-in-Aid for Scientific Research on Innovative Areas (Frontier Science of Interactions between Plasmas and Nano-interfaces, Grant 21110002) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. The authors thank the Materials Design and Characterization Laboratory, Institute for Solid State Physics, University of Tokyo for providing access to the pulsed laser facility, and Prof. M. Suzuki for assistance with the GC-MS measurements and helpful discussions.

REFERENCES

1.  T.L. Chester, J.D. Pinkston, and D.E. Raynie. Supercritical fluid chromatography and extraction. Anal. Chem., 68(12), R487–R514, 1996.

2.  P.G. Jessop and W. Leitner, Eds. Chemical Synthesis Using Supercritical Fluids. Wiley-VCH Verlag GmbH, Berlin, 1999.

3.  M.A. Lieberman and A.J. Lichtenberg. Principles of Plasma Discharges and Materials Processing. 2nd ed. Wiley-Interscience, New York, 2005.

4.  A. Fridman. Plasma Chemistry. 1st ed. Cambridge University Press, Cambridge, 2008.

5.  J.E. Dahl, S.G. Liu, and R.M.K. Carlson. Isolation and structure of higher diamondoids, nanometer-sized diamond molecules. Science, 299(5603), 96–99, 2003.

6.  H. Schwertfeger, A.A. Fokin, and P.R. Schreiner. Diamonds are a chemist’s best friend: Diamondoid chemistry beyond adamantine. Angew. Chem. Int. Ed., 47(6), 1022–1036, 2008.

7.  O.A. Shenderova, V.V. Zhirnov, and D.W. Brenner. Carbon nanostructures. Crit. Rev. Solid State Mater. Sci., 27(3–4), 227–356, 2002.

8.  R.R. Tykwinski, W. Chalifoux, S. Eisler, A. Lucotti, M. Tommasini, D. Fazzi, M. Del Zoppo, and G. Zerbi. Toward carbyne: Synthesis and stability of really long polyynes. Pure Appl. Chem., 82(4), 891–904, 2010.

9.  H.W. Kroto, J.R. Heath, S.C. Obrien, R.F. Curl, and R.E. Smalley. C-60—Buckminsterfullerene. Nature, 318(6042), 162–163, 1985.

10.  S. Iijima. Helical microtubules of graphitic carbon. Nature, 354(6348), 56–58, 1991.

11.  K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, and A.A. Firsov. Electric field effect in atomically thin carbon films. Science, 306(5696), 666–669, 2004.

12.  W. Humphrey, A. Dalke, and K. Schulten. VMD: Visual molecular dynamics. J. Mol. Graph., 14(1), 33, 1996.

13.  A.T. Balaban and P.V. Schleyer. Systematic classification and nomenclature of diamond hydrocarbons.1. Graph-theoretical enumeration of polymantanes. Tetrahedron, 34(24), 3599–3609, 1978.

14.  J.Y. Raty and G. Galli.Ultradispersity of diamond at the nanoscale. Nat. Mater., 2(12), 792–795, 2003.

15.  R.A. Freitas and R.C. Merkle. A minimal toolset for positional diamond mechanosynthesis. J. Comput. Theor. Nanosci., 5(5), 760–861, 2008.

16.  H. Schwertfeger and P.R. Schreiner. Future of diamondoids. Chem. Unserer Zeit, 44(4), 248–253, 2010.

17.  B. Reisberg, R. Doody, A. Stoffler, F. Schmitt, S. Ferris, and H.J. Mobius. Memantine in moderate-to-severe Alzheimer’s disease. N. Engl. J. Med., 348(14), 1333–1341, 2003.

18.  W.J. Geldenhuys, S.F. Malan, J.R. Bloomquist, A.P. Marchand, and C.J. Van der Schyf. Pharmacology and structure-activity relationships of bioactive polycyclic cage compounds: A focus on pentacycloundecane derivatives. Med. Res. Rev., 25(1), 21–48, 2005.

19.  V.G.H. Evidente, C.H. Adler, J.N. Caviness, and K. Gwinn-Hardy. A pilot study on the motor effects of rimantadine in Parkinson’s disease. Clin. Neuropharmacol., 22(1), 30–32, 1999.

20.  W.L. Yang, J.D. Fabbri, T.M. Willey, J.R.I. Lee, J.E. Dahl, R.M.K. Carlson, P.R. Schreiner, A.A. Fokin, B.A. Tkachenko, N.A. Fokina, W. Meevasana, N. Mannella, K. Tanaka, X.J. Zhou, T. van Buuren, M.A. Kelly, Z. Hussain, N.A. Melosh, and Z.X. Shen. Monochromatic electron photoemission from diamondoid monolayers. Science, 316(5830), 1460–1462, 2007.

21.   T.M. Willey, J.D. Fabbri, J.R.I. Lee, P.R. Schreiner, A.A. Fokin, B.A. Tkachenko, N.A. Fokina, J.E.P. Dahl, R.M.K. Carlson, A.L. Vance, W.L. Yang, L.J. Terminello, T. van Buuren, and N.A. Melosh. Near-edge x-ray absorption fine structure spectroscopy of diamondoid thiol monolayers on gold. J. Am. Chem. Soc., 130(32), 10536–10544, 2008.

22.  W.A. Clay, Z. Liu, W.L. Yang, J.D. Fabbri, J.E. Dahl, R.M.K. Carlson, Y. Sun, P.R. Schreiner, A.A. Fokin, B.A. Tkachenko, N.A. Fokina, P.A. Pianetta, N. Melosh, and Z.X. Shen. Origin of the monochromatic photoemission peak in diamondoid monolayers. Nano Lett., 9(1), 57–61, 2009.

23.  S. Roth, D. Leuenberger, J. Osterwalder, J.E. Dahl, R.M.K. Carlson, B.A. Tkachenko, A.A. Fokin, P.R. Schreiner, and M. Hengsberger. Negative-electron-affinity diamondoid monolayers as high-brilliance source for ultrashort electron pulses. Chem. Phys. Lett., 495(1–3), 102–108, 2010.

24.  N.D. Drummond, A.J. Williamson, R.J. Needs, and G. Galli. Electron emission from diamondoids: A diffusion quantum Monte Carlo study. Phys. Rev. Lett., 95(9), 096801, 2005.

25.  W.A. Clay, T. Sasagawa, M. Kelly, J.E. Dahl, R.M.K. Carlson, N. Melosh, and Z.X. Shen. Diamondoids as low-κ dielectric materials. Appl. Phys. Lett., 93(17), 172901-1–172901-3, 2008.

26.  A.A. Fokin and P.R. Schreiner. Band gap tuning in nanodiamonds: First principle computational studies. Mol. Phys., 107(8–12), 823–830, 2009.

27.  T. Sasagawa and Z.X. Shen. A route to tunable direct band-gap diamond devices: Electronic structures of nanodiamond crystals. J. Appl. Phys., 104(7), 073704, 2008.

28.  Z.B. Wei, J.M. Moldowan, J. Dahl, T.P. Goldstein, and D.M. Jarvie. The catalytic effects of minerals on the formation of diamondoids from kerogen macromolecules. Org. Geochem., 37(11), 1421–1436, 2006.

29.  R.I.K. Carlson, J.E.P. Dahl, S.G. Liu, M.M. Olmstead, P.R. Buerki, and R. Gat, Diamond molecules found in petroleum—New members of the H-terminated diamond series. In Synthesis and Properties of Ultrananocrystalline Diamond. Springer, Dordrecht, Berlin, 2005, pp. 63–78.

30.  A. Iwasa, W.A. Clay, J.E. Dahl, R.M.K. Carlson, Z.X. Shen, and T. Sasagawa. Environmentally friendly refining of diamond-molecules via the growth of large single crystals. Cryst. Growth Des., 10(2), 870–873, 2010.

31.  E. Osawa, A. Furusaki, N. Hashiba, T. Matsumoto, V. Singh, Y. Tahara, E. Wiskott, M. Farcasiu, T. Iizuka, N. Tanaka, T. Kan, and P.V. Schleyer. Application of force-field calculations to organic-chemistry. 2. Thermodynamic rearrangements of larger polycyclic-hydrocarbons derived from the 38.5 and 41.5-degrees-c melting dimers of cyclooctatetraene—Crystal and molecular-structures of 5-bromoheptacyclo [8.6.0.02,8.03,13.0 4,11.05,9.012,16] hexadecane (5-bromo-(c2)-bisethanobisnordiamantane), 6,12-dibromoheptacyclo [7.7.0.02,6.03,15.04,12.05,10.011,16] hexadecane, and nonacyclo [11.7.1.12,18.03,16.04,13.05,10.06,14.07,11.015,20] docosane (bastardane). J. Org. Chem., 45(15), 2985–2995, 1980.

32.  W. Burns, M.A. McKervey, T.R.B. Mitchell, and J.J. Rooney. New approach to construction of diamondoid hydrocarbons—Synthesis of anti-tetramantane. J. Am. Chem. Soc., 100(3), 906–911, 1978.

33.  Y.P. Raizer. Gas Discharge Physics. Springer-Verlag, Berlin, 1991.

34.  C.E. Pittman, L.H. Cohen, and H. Meyer. Transport properties of helium near the liquid vapor critical point. 1. Thermal-conductivity of He-3. J. Low Temp. Phys., 46(1–2), 115–135, 1982.

35.  V. Vesovic, W.A. Wakeham, G.A. Olchowy, J.V. Sengers, J.T.R. Watson, and J. Millat. The transport-properties of carbon-dioxide. J. Phys. Chem. Ref. Data, 19(3), 763–808, 1990.

36.  Z.H. Chen, K. Tozaki, and K. Nishikawa. Development of thermal conductivity measurement for fluids which is convenient and effective for samples near the critical point. Jpn. J. Appl. Phys., 38(1AB), L92–L95, 1999.

37.  Stanley M. Walas. Phase Equilibria in Chemical Engineering. Butterworth-Heinemann, London, 1985.

38.  D.R. Lide, Ed. Handbook of Chemistry and Physics. Taylor & Francis, Boca Raton, FL, 2006.

39.  NIST Chemistry Webbook. Thermophysical Properties of Fluid Systems. Technical Report. NIST, 2011.

40.  C.C. Liew, H. Inomata, and S. Saito. Molecular-dynamics study on solvent clustering in supercritical-fluid solutions based on particle radial kinetic-energy. Fluid Phase Equilib., 104, 317–329, 1995.

41.  H.E. Stanley. Introduction to Phase Transitions and Critical Phenomena. Oxford University Press, Oxford, 1971.

42.  C.A. Eckert, B.L. Knutson, and P.G. Debenedetti. Supercritical fluids as solvents for chemical and materials processing. Nature, 383(6598), 313–318, 1996.

43.  P.E. Savage, S. Gopalan, T.I. Mizan, C.J. Martino, and E.E. Brock. Reactions at supercritical conditions—Applications and fundamentals. AICHE J., 41(7), 1723–1778, 1995.

44.  T. Toriumi, T. Kawakami, J. Sakai, D. Ogawa, and N. Higashi. (In Japanese). J. Ind. Chem., 49, 1, 1946.

45.  Y. Ikushima, S. Ito, T. Asano, T. Yokoyama, N. Saito, K. Hatakeda, and T. Goto. A Diels-Alder reaction in supercritical carbon-dioxide medium. J. Chem. Eng. Jpn., 23(1), 96–98, 1990.

46.  R.L. Thompson, R. Glaser, D. Bush, C.L. Liotta, and C.A. Eckert. Rate variations of a hetero-Diels-Alder reaction in supercritical fluid CO₂. Ind. Eng. Chem. Res., 38(11), 4220–4225, 1999.

47.  D.R. Gere. Supercritical fluid chromatography. Science, 222(4621), 253–259, 1983.

48.  M.L. Lee and K.E. Markides. Chromatography with supercritical fluids. Science, 235(4794), 1342–1347, 1987.

49.  K. Zosel. Separation with supercritical gases—Practical applications. Angew. Chem. Int. Ed. Engl., 17(10), 702–709, 1978.

50.  G.A. Montero, C.B. Smith, W.A. Hendrix, and D.L. Butcher. Supercritical fluid technology in textile processing: An overview. Ind. Eng. Chem. Res., 39(12), 4806–4812, 2000.

51.  G.L. Weibel and C.K. Ober. An overview of supercritical CO₂ applications in microelectronics processing. Microelectron. Eng., 65(1–2), 145–152, 2003.

52.  H. Namatsu, K. Yamazaki, and K. Kurihara. Supercritical drying for nanostructure fabrication without pattern collapse. Microelectron. Eng., 46(1–4), 129–132, 1999.

53.  H. Namatsu, K. Yamazaki, and K. Kurihara. Supercritical resist dryer. J. Vac. Sci. Technol. B, 18(2), 780–784, 2000.

54.  J.J. Watkins, J.M. Blackburn, and T.J. McCarthy.Chemical fluid deposition: Reactive deposition of platinum metal from carbon dioxide solution. Chem. Mater., 11(2), 213–215, 1999.

55.  J.M. Blackburn, D.P. Long, A. Cabanas, and J.J. Watkins. Deposition of conformal copper and nickel films from supercritical carbon dioxide. Science, 294(5540), 141–145, 2001.

56.  E. Kondoh, K. Sasaki, and Y. Nabetani. Deposition of zinc oxide thin films in supercritical carbon dioxide solutions. Appl. Phys. Express, 1(6), 061201, 2008.

57.  F. Cansell and C. Aymonier.Design of functional nanostructured materials using supercritical fluids. J. Supercrit. Fluids, 47(3), 508–516, 2009.

58.  E. Reverchon and R. Adami.Nanomaterials and supercritical fluids. J. Supercrit. Fluids, 37(1), 1–22, 2006.

59.  E. Reverchon, R. Adami, S. Cardea, and G. Della Porta. Supercritical fluids processing of polymers for pharmaceutical and medical applications. J. Supercrit. Fluids, 47(3), 484–492, 2009.

60.  T. Ito and K. Terashima.Generation of micrometer-scale discharge in a supercritical fluid environment. Appl. Phys. Lett., 80(16), 2854–2856, 2002.

61.   M. Sawada, T. Tomai, T. Ito, H. Fujiwara, and K. Terashima. Micrometer-scale discharge in high-pressure H₂O and Xe environments including supercritical fluid. J. Appl. Phys., 100(12), 123304, 2006.

62.  T. Ito, K. Katahira, Y. Shimizu, T. Sasaki, N. Koshizaki, and K. Terashima. Carbon and copper nanostructured materials syntheses by plasma discharge in a supercritical fluid environment. J. Mater. Chem., 14(10), 1513–1515, 2004.

63.  T. Tomai, T. Ito, and K. Terashima. Generation of dielectric barrier discharge in high-pressure N₂ and CO₂ environments up to supercritical conditions. Thin Solid Films, 506, 409–413, 2006.

64.  T. Tomai, K. Katahira, H. Kubo, Y. Shimizu, T. Sasaki, N. Koshizaki, and K. Terashima. Carbon materials syntheses using dielectric barrier discharge microplasma in supercritical carbon dioxide environments. J. Supercrit. Fluids, 41(3), 404–411, 2007.

65.  M. Motiei, Y.R. Hacohen, J. Calderon-Moreno, and A. Gedanken. Preparing carbon nanotubes and nested fullerenes from supercritical CO₂ by a chemical reaction. J. Am. Chem. Soc., 123(35), 8624–8625, 2001.

66.  Z.S. Lou, Q.W. Chen, W. Wang, and Y.F. Zhang. Synthesis of carbon nanotubes by reduction of carbon dioxide with metallic lithium. Carbon, 41(15), 3063–3067, 2003.

67.  P.P. Patil, D.M. Phase, S.A. Kulkarni, S.V. Ghaisas, S.K. Kulkarni, S.M. Kanetkar, S.B. Ogale, and V.G. Bhide. Pulsed-laser induced reactive quenching at a liquid-solid interface—Aqueous oxidation of iron. Phys. Rev. Lett., 58(3), 238–241, 1987.

68.  S.B. Ogale, P.P. Patil, D.M. Phase, Y.V. Bhandarkar, S.K. Kulkarni, S. Kulkarni, S.V. Ghaisas, and S.M. Kanetkar. Synthesis of metastable phases via pulsed-laser-induced reactive quenching at liquid-solid interfaces. Phys. Rev. B, 36(16), 8237–8250, 1987.

69.  T. Sakka, K. Takatani, Y.H. Ogata, and M. Mabuchi. Laser ablation at the solid-liquid interface: Transient absorption of continuous spectral emission by ablated aluminium atoms. J. Phys. D Appl. Phys., 35(1), 65–73, 2002.

70.  T. Sakka, K. Saito, and Y.H. Ogata. Emission spectra of the species ablated from a solid target submerged in liquid: Vibrational temperature of c-2 molecules in water-confined geometry. Appl. Surf. Sci., 197, 246–250, 2002.

71.  K. Saito, K. Takatani, T. Sakka, and Y.H. Ogata. Observation of the light emitting region produced by pulsed laser irradiation to a solid-liquid interface. Appl. Surf. Sci., 197, 56–60, 2002.

72.  K. Saito, T. Sakka, and Y.H. Ogata. Rotational spectra and temperature evaluation of c-2 molecules produced by pulsed laser irradiation to a graphite-water interface. J. Appl. Phys., 94(9), 5530–5536, 2003.

73.  M.N.R. Ashfold, F. Claeyssens, G.M. Fuge, and S.J. Henley. Pulsed laser ablation and deposition of thin films. Chem. Soc. Rev., 33(1), 23–31, 2004.

74.  V. Amendola and M. Meneghetti.Laser ablation synthesis in solution and size manipulation of noble metal nanoparticles. Phys. Chem. Chem. Phys., 11(20), 3805–3821, 2009.

75.  D. Amans, A.C. Chenus, G. Ledoux, C. Dujardin, C. Reynaud, O. Sublemontier, K. Masenelli-Varlot, and O. Guillois. Nanodiamond synthesis by pulsed laser ablation in liquids. Diam. Relat. Mater., 18(2–3), 177–180, 2009.

76.  K. Saitow, T. Yamamura, and T. Minami. Gold nanospheres and nanonecklaces generated by laser ablation in supercritical fluid. J. Phys. Chem. C, 112(47), 18340–18349, 2008.

77.  K. Saitow and T. Yamamura.Effective cooling generates efficient emission: Blue, green, and red light-emitting Si nanocrystals. J. Phys. Chem. C, 113(19), 8465–8470, 2009.

78.  K. Sasaki and S. Aoki.Temporal variation of two-dimensional temperature in a laser-ablation plume produced from a graphite target. Appl. Phys. Express, 1(8), 086001, 2008.

79.  H. Kikuchi, S. Stauss, S. Nakahara, K. Matsubara, T. Tomai, T. Sasaki, and K. Terashima. Development of sheet-like dielectric barrier discharge microplasma generated in supercritical fluids and its application to the synthesis of carbon nanomaterials. J. Supercrit. Fluids, 55(1), 325–332, 2010.

80.   S. Stauss, H. Miyazoe, T. Shizuno, K. Saito, T. Sasaki, and K. Terashima. Synthesis of the higher-order diamondoid hexamantane using low-temperature plasmas generated in supercritical xenon. Jpn. J. Appl. Phys., 49(7), 070213, 2010.

81.  T. Shizuno, H. Miyazoe, K. Saito, S. Stauss, M. Suzuki, T. Sasaki, and K. Terashima. Synthesis of diamondoids by supercritical fluid discharge plasma. Appl. Phys. Express, 50(3), 030207, 2011.

82.  J.E.P. Dahl, J.M. Moldowan, Z. Wei, P.A. Lipton, P. Denisevich, R. Gat, S. Liu, P.R. Schreiner, and R.M.K. Carlson. Synthesis of higher diamondoids and implications for their formation in petroleum. Angew. Chem. Int. Ed., 122(51), 10077, 2010.

83.  S. Nakahara, S. Stauss, T. Kato, T. Sasaki, and K. Terashima. Synthesis of higher diamondoids by pulsed laser ablation plasmas in supercritical Co₂. J. Appl. Phys., 109(12), 123304, 2011.

84.  S. Nakahara, S. Stauss, H. Miyazoe, T. Shizuno, M. Suzuki, H. Katakoka, T. Sasaki, and K. Terashima. Laser ablation synthesis of diamond molecules in supercritical fluids. Appl. Phys. Express, 3(9), 096201, 2010.

85.  J. Filik, J.N. Harvey, N.L. Allan, P.W. May, J.E.P. Dahl, S.G. Liu, and R.M.K. Carlson. Raman spectroscopy of diamondoids. Spectrochim. Acta Pt. A Mol. Biol., 64(3), 681–692, 2006.

86.  D. Ballutaud, F. Jomard, T. Kociniewski, E. Rzepka, H. Girard, and S. Saada. Sp(3)/sp(2) character of the carbon and hydrogen configuration in micro- and nanocrystalline diamond. Diam. Relat. Mater., 17(4–5), 451–456, 2008.

87.  N. Zhang, Z.J. Tian, Y.Y. Leng, H.T. Wang, F.Q. Song, and J.H. Meng. Sci. China Ser. D Earth Sci., 50(8), 1171, 2007.

88.  N. Wada, P.J. Gaczi, and S.A. Solin. Diamond-like 3-fold coordinated amorphous-carbon. J. Non-Cryst. Solids, 35(6), 543–548, 1980.

89.  A.T. Balaban, D.J. Klein, J.E. Dahl, and R.M.K. Carlson. Molecular descriptors for natural diamondoid hydrocarbons and quantitative structure-property relationships for their chromatographic data. Open Org. Chem. J., 1, 13–31, 2007.

90.  L. Mansuy, R.P. Philp, and J. Allen. Source identification of oil spills based on the isotopic composition of individual components in weathered oil samples. Environ. Sci. Technol., 31(12), 3417–3425, 1997.

91.  B.S. Greensfelder, H.H. Voge, and G.M. Good. Catalytic and thermal cracking of pure hydrocarbons—Mechanisms of reaction. Ind. Eng. Chem., 41(11), 2573–2584, 1949.