High Pressure Molecular Science

The field of laser Raman spectroscopy of liquids and gases is reviewed. After introducing the importance of using pressure as an experimental variable in Raman studies of fluids, a brief overview of the experimental techniques is presented. Illustrative examples of specific high-pressure Raman studies deal with the following topics: reorientational motions in liquids, vibrational dephasing, collision-induced scattering, Fermi resonance and Raman frequency noncoincidence effect.

The effect of pressure has been measured for two non-linear optical phenomena: second harmonic generation (SHG) in organic crystals and one and two photon fluorescence excitations in a molecule dissolved in an organic polymer. Both sets of results are discussed in terms of theory.

There is a large class of “flexible” molecules which, when excited, can isomerize from a planar to a twisted form. In this paper we present both steady state and time resolved data for two such molecules in a solid polymer. We present a model consistent with these data and demonstrate that the steady state results can be predicted from the time resolved data.

The study of high-pressure transformations and the stability of dense materials has been an active area research for many decades. The change of the structure of a solid when pressurized has yielded many interesting new materials and many new phenomena such as insulator → metal transformations, important effects in high T_c superconductors, and pressure-induced amorphization that yields unique disordered materials. Ten years ago, the application of classical molecular dynamics techniques had already yielded results that provided important insight into materials under high pressure. This was made possible by the constant-pressure method developed by Anderson^{1 1} and by Parrinello and Rahman^2,3 which allows the volume as well as the shape of the simulation cell to change. Another important aspect of these simulation methods was that systems could be studied at finite temperatures. These developments complemented the rapidly developing quantum mechanical methods being used in the solid state physics community to study the stability and structures of solids under pressure^4,5. A major advance in the theoretical methods for the study of structural stability and dynamical properties of solids was made with the introduction of the Car-Parrinello method⁶ for ab initio molecular dynamics (AIMD). A primary advantage of this method is the fact that a interatomic potential is not a part of the input. The Car-Parrinello method allows the calculation of all of the properties that can be obtained in the classical MD technique but with the important feature that interatomic forces are obtained from a quantum mechanical calculation. The AIMD method has more recently been extended to include Parrinello-Rahman variable cell dynamics^7,8 so this method has the capability to describe structural phase changes and stabilities from first-principles. Other variations of constant pressure MD have also been proposed and the basic ideas and capabilities are similar⁹.

The operating principles of diamond-anvil high pressure cells are reviewed, with particular attention to the implications for design and construction. The diamond culets and gasket generate the pressure, and their behaviour dictates the requirements for the rest of the cell. The axial alignment mechanism is crucial, while tilt alignment is less important. The implication for piston-cylinder designs is that the clearance of the piston in the bore is critical, while the length of the piston is not. Good practice in the design of drive mechanisms is discussed. Finally, we consider alternatives to the standard piston-cylinder mechanism. Flexure movements, and their basic design rules are presented.

A simple polynomial fit to data which is sublinear with pressure gives coefficients which vary strongly with the pressure range of the fit. Converting pressure to a variable which is linear with that measured enables direct comparison of experiments over different pressure ranges. We illustrate this for high-pressure band-gap data in GaAs. Experimental errors such as non-hydrostaticity can occur and be difficult to detect. Comparison of two spectroscopies provides a simple diagnostic. We find a linear shift of the band-gap with the Raman shift under pressure for ZnTe and propose this as a check of pressure conditions. To measure the Raman shift accurately under pressure, resonant Raman is a useful tool, but requires a tunable source. We have developed a tunable, broadband filter for high-pressure work, which rejects Rayleigh scattered light, allowing measurement of Raman shifts as small as 10cm^-1.

Basic effects of pressure on semiconductors are the increase of direct band gap with the density of the crystal, the consequent band crossovers, and the increase in phonon frequencies. We briefly review experiments on these, with emphasis on those aspects which are not understood. These are the density-dependence of the band gap, the effects of combined pressure and strain, and the equation of state. The theoretical framework is discussed. While thermodynamics, elasticity theory, model solid theory and empirical formulae are all relevant to the experimental work, these different theoretical approaches do not converge to provide a coherent interpretation of the experimental work.

Our understanding of the liquid-vapour equilibrium in metallic systems has increased enormously in the past two decades. Much of this has been stimulated by a series of pioneering papers of Mott [1] on the metal-non-metal transition which shows up when a liquid metal is heated to the region of the liquid-vapour critical point. The existence of this transition implies that the liquid-vapour phase transition of fluid metals is distinct from that of normal insulating fluids such as inert gases. An inert-gas atom retains its identity in the condensed phase and the pair potential which determines the properties of the dilute vapour phase is supplemented to a limited degree by many-body interactions, in the total potential energy of the dense phase [2]. In contrast, the electronic structures of the two coexisting phases, liquid and vapour, of fluid metals may be vastly different. The essntial point is that the metallic state is a collective phenomenon existing only when the density of atoms is sufficiently large. Unlike inert gases the electronic stucture in the high-density liquid is very different from that of an atom in the dilute vapour.

Liquid alloys have been the subject of considerable interest over many years and have attracted the attention of scientists from diverse fields ranging from metallurgy, inorganic chemistry to condensed matter and theoretical physics [1–3]. Much of this work has been motivated by the strong interest in the electronic, thermodynamic and structural properties of systems which are at the borderline between liquid metals, liquid semiconductors, and ionic conductors. In this respect, the chemical bonding and chemical short-range order of liquid alkali-polyvalent alloys are topics of continuing interest. In extreme cases, a metal-insulator or metal-semiconductor transition occurs during alloying, e.g., in Cs-Au [2] and K-Pb [4] at the stoichiometric 1:1 composition. The crystalline binary alloys obtained by combining an alkali metal (e.g. Na, Cs) with polyvalent metals like Sn, Pb or Sb form classical examples of ionic alloy systems, exhibiting anion clustering. In general, if the composition is such that the number of electrons transferred is sufficient to complete the octet shell of the electronegative element, bonding and structure are salt-like, and if the electronic shell remains incomplete, a saturated chemcial bond can be formed only if valence electrons are shared among the electronegative atoms. This results in the formation of polyanionic clusters stabilized by strong covalent bonds and immersed in a matrix of the electropositive species. Zintl and Brauer proposed, that as a consequence of the large difference in electronegativity, an electron is transferred from the alkali atoms to Sn, Pb or Sb [5], and the polyanionic clusters are isostructural to the molecules or crystal structures formed by the isoelectronic neutral species. Examples are the tetrahedral Sn₄ ⁴⁻ or Pb₄ ⁴⁻ polyanions formed in equiatomic alkali-tin or alkali-lead alloys or the chainlike Sb_∞ ⁻ clusters in alkali-antimony alloys, which are isoelectronic and isomorphic to P₄ molecules or to the helical chains formed by the crystalline chalcogen elements, respectively. The role of the alkali ions is merely to separate the anions.

Considerable effort has been put in the development of theories for the description of the behaviour of gas-liquid and liquid-liquid phase equilibria. Literature about the mutual solubility of simple molecular systems in the solid phase is rather limited. More work has been done on metallic systems. It is generally believed that solubility at high density, in particular in solid systems, is mainly governed by geometrical effects. For instance, the well known Hume-Rotary rule, often used by metallurgists, states that a binary mixed solid is only obtained if the diameters of the molecules differ by less than 15 %, otherwise the mixture will separate into the pure solids. We will investigate whether such a geometrical rule is also valuable for molecular systems. Nowadays, the geometrical effects can be accurately calculated by means of computer simulations as well as analytical theories such as density functional theory. We will first consider in section 2 the case of binary hard sphere mixtures, since it is to be expected that the behaviour of those mixtures will give some insight into the effects of the difference in molecular size on the phase behaviour of real binary mixtures of simple molecular systems at high density. Moreover, we will deal with the phase behaviour of colloidal mixtures which is often in close agreement with that of hard sphere mixtures. In section 3 the experimental results on disordered solid solutions are examined. In section 4 we discuss a new class of compounds, the van der Waalscompounds, characterised by only weak intermolecular forces. In section 5 the differences between the experimental behaviour of molecular and hard spheres systems will be analysed in more detail and some conclusions will be drawn. Finally in section 6 a summary will be given.

Vibrational Raman spectroscopy is one of the most useful tools to study the behaviour of systems at high pressure. Phase transitions can be detected and information about the dynamics of the transition may be obtained from the spectra. Information about the microscopic behaviour of the molecules, orientational as well as translational, is revealed by both the peak position and the line width. In mixtures Raman spectroscopy can be used as a tool for determining the concentration or the concentration fluctuations.

For the understanding of basic mechanisms in condensed matter, organic charge-transfer (CT) complexes are of general interest because of their particular electronic tunability strongly coupled with their structural properties. Thus they can refer simultaneously to the chemical physics of organic solid state, by the way of notions like molecular multistability or solid state electron transfer reactions, and to the physics of low-dimensional systems where non-linear structurally-relaxed electronic excitations play a tremendous role, but also to the physics of structural phase transitions for the cooperativity point of view. This is particularly true for the neutral-to-ionic (N-I) transition which takes place in most of mixed-stack CT complexes [1]. These systems are formed of stacks with alternating electron donor (D) and electron acceptor (A) molecules. The transition manifests itself by a change of the degree of CT and a dimerization distortion with the formation of (D⁺A^-) singlet pairs along the stacking axis in the ionic phase [2]. It is known nowadays, that the transformation can be induced by temperature [3], and/or photo-irradiation [4], but it was pressure which allowed the first observation of the N-I transition [1]. After a brief description of the mechanism of the N-I transition, considering pressure as well as temperature effects, this paper will be focussed on the thermodynamics of its recognized ambassador, the tetrathiafulvalene-p-chloranil (TTF-CA) on the basis of on recent experimental results [6] supported by theoretical considerations [7].

High pressure Raman and neutron scattering study of carbon blacks and HOPG graphite is reported. Carbon black particles are composed of graphitic nanocrystallites and amorphous carbon. Relative concentration of amorphous carbon decreases slightly with increased pressure. This non-reversible process differs from temperature induced transformations of amorphous carbon into ordered carbon. Post-production treatment at high temperatures results in the growth of graphitic crystallites whereas no such effect was observed when the pressure was applied to carbon black.

The field of high-resolution nuclear magnetic resonance spectroscopy at high pressure is reviewed. The various applications of the high-pressure NMR spectroscopy to chemical systems are given out in the introduction. After a discussion of high-resolution NMR instrumentation for experiments at high pressures up to 950 MPa, the two main sections cover the recent applications of the high-resolution NMR spectroscopy at high pressure to chemical and biochemical systems. In particular, the use of 2D NMR techniques will be emphasized as the combination of advanced NMR methods with the high-pressure capability represents a new, powerful tool in studies of chemical and biochemical systems.

Raman spectroscopy was used to study intermolecular interactions in liquid dimethyl sulfoxide. The non-coincidence of the vibrational wavenumbers of the in-phase and the out-of-phase SO stretching mode was analyzed in the temperature range of 30 – 70 °C under pressures up to 2200 bar. The observed non-linear increase of this splitting parameter as a function of density can be explained by an increase in local order leading to an enhanced coupling of neighboring oscillators. The isotropic SO stretching vibrational band consists of three components which are attributed to clusters differing in intermolecular association strength. The pressure dependence of the integrated intensities of the three SO band components is interpreted as a pressure-induced formation of clusters consisting of strongly associated molecules.

The elucidation of the mechanisms of thermal reactions in inorganic and bioinorganic chemistry through the application of high pressure kinetic techniques is described. High pressure kinetic and thermodynamic data are used to construct volume profiles for typical reactions that include ligand substitution, activation of small molecules, and electron transfer processes. The mechanistic information obtained from such profiles is discussed in detail.

Volume is a perfectly tangible parameter in current life. On the macroscopic level the volume profile is a pictorial view of a chemical reaction on the basis of volume changes. When reactants are converted into products the volume change is expressed by the reaction volume based on partial molar volumes depending on the medium

Six years ago, during a preceeding NATO ASI devoted to High Pressure Chemistry, Biochemistry and Material Science, we reported on “The Future of High Pressure Organic Chemistry””[1]. The accent was focused on the specific advantage of using high pressure activation vs traditional activation methods apparently competing with pressure in organic synthesis. From 1992 to date, the armory of activation methods has been enriched in such way that the access to sophisticated organic molecules is now possible.

Ethene homo- and copolymerizations are important technical processes. At pressures up to about 3000 bar and temperatures up to 300 °C, approximately 16 million tons LDPE (low density polyethylene) have been produced worldwide in 1997. The continued interest in the LDPE process is primarily due to the enormous flexibility of this reaction which is carried out under supercritical (sc) conditions. A particular advantage of free-radical polymerization in sc fluid phase relates to the potential of widely tuning polymer properties just by continuously varying polymerization conditions. Further advantages consist in the tunability of solvent properties (which allows for choosing p and T conditions such that either homogeneity, e.g., for reaction, or inhomogeneity, e.g., for the subsequent separation step, are achieved) and in heat and mass transfer processes being very efficient under sc conditions. It is occasionally overlooked that the high-pressure ethene polymerization is the archetype of an extremely useful and successful sc fluid phase process. The situation met with the high-pressure ethene polymerization is referred to as reactive sc fluid phase. Ethene acts as both reactant and tunable supercritical fluid medium.

Amphiphilic lipid molecules, which provide valuable model systems for lyotropic mesophases and biomembranes, display a variety of polymorphic phases, depending on their molecular structure and environmental conditions, such as the water content, pH, ionic strength, temperature and pressure [1–5]. The basic structural element of biological membranes consists of a lamellar phospholipid bilayer matrix. Due to the large hydrophobic effect, most phospholipid bilayers associate in water already at very low concentrations (< 10⁻¹² mol·L⁻¹). Saturated phospholipids often exhibit two thermotropic lamellar phase transitions, a gel to gel (L_β′/P_β′) pretransition and a gel to liquid-crystalline (P_β′/L_α) main transition at a higher temperature T _m (see Fig. 1). In the fluid-like L_α-phase, the acyl chains of the lipid bilayers are conformationally disordered, whereas in the gel phases, the chains are more extended and ordered. In addition to these thermotropic phase transitions, also pressure-induced phase transformations have been observed (see, e.g., [6–20]). Upon compression, the lipids adopt to volume restriction by changing their conformation and packing system. When such adjustments are no longer possible, any further increase in pressure results mainly in a reduction of bond lengths, which affects the frequency of the stretching vibrations in the IR spectra. It is now well-known that many biological lipid molecules also form non-lamellar liquid-crystalline phases (see Fig. 1) [1–5,21–23]. Lipids, which can adopt the hexagonal phase are present at substantial levels in biological membranes, usually at least 30 mol% of total lipids. For the double-chain lipids found in membranes, the polar/apolar interface curves towards the water (such phases are called inverted or type II). Some lipid extracts, such as those from archaebacteria (S. solfataricus), exhibit a cubic liquid-crystalline phase [24,25].

High hydrostatic pressure induces changes in protein conformation, solvation and enzyme activities. To have access to these structural modifications, various biophysical techniques have been specially designed for high pressure experiments. Among them, spectroscopic methods are rather simple and easily adaptable both to high pressure and controlled temperatures. In the present paper, different improvements are presented together with some applications.

The recent interest in high pressure biochemistry requires more fundamental studies on the behavior of enzymes under such extreme conditions. In this paper we shortly review some basic knowledge on the high pressure effect on enzyme reactions, together with some new data concerning high pressure micellar enzymology taking as a model for enzyme studies in a membrane environment.

The pressure and temperature behavior of proteins is discussed in the framework of the phase diagram. This gives unique information on the changes in heat capacity, thermal expansion and compressibility of protein unfolding. It also relates the cold, heat and pressure denaturation. The difference in pressure- and temperature-induced aggregation of unfolded proteins shows the unique features of pressure effects. A molecular interpretation of the thermodynamic quantities is not possible on the basis of model systems unless the packing defects are taken into account. High pressure molecular dynamic calculations contribute in a unique way to our understanding of pressure effects.

The study of pressure effects on protein stability has occupied a relatively marginal position in the field of protein folding, with very few thorough thermodynamic, structural and kinetic studies of this phenomenon. Moreover, theoretical treatment of the issue with a few recent exceptions, has been limited to declarations of its complexity and lack of concordance with the results from other approaches. This paucity of data and theory notwithstanding, understanding the fundamental physical basis for pressure effects on proteins is essential to progress in the field of protein folding. Moreover, pressure presents certain advantages as a perturbation methodology that render it an important, useful and complementary approach. In the present review, the issue of the fundamental basis for the effects of pressure is discussed. Reference is made to studies in the literature, but I have concentrated the detailed presentation on the body of work in pressure-induced protein unfolding carried out by my research group and collaborators on staphylococcal nuclease (Snase) over the past 5 years. The origins of the value of the change in volume upon unfolding must be understood prior to any thorough theoretical analysis of pressure effects. The various arguments for the multiple contributing factors are discussed and then recent studies from my research group designed to probe this question are presented, the overall conclusion being that the existence of packing defects in the folded structure represents the most likely candidate for the negative change in volume upon unfolding. Moreover, the results of the temperature dependence of the volume change for unfolding of Snase implicate the difference in thermal expansivity in the temperature dependence of the value of the volume change of unfolding. Next I present results of a characterization of the physical properties of the pressure denatured state of Snase, and compare these to studies on a number of other pressure denatured proteins. Finally, the results of a series of pressure-jump kinetic studies on the folding/unfolding reactions of this protein are discussed. It is too early to conclude whether the results from these pressure studies on Snase stability and their interpretations are general. For this, many more studies on a number of small, reversibly folding proteins will be required.

The contribution of protein folding and protein-nucleic acid interactions to virus assembly has been measured in several bacterial, plant and animal viruses, using hydrostatic pressure as thermodynamic variable. By comparing the pressure stability among native wild-type viruses, single-amino acid mutants or empty particles, we have gained new insights about virus assembly and disassembly. We find that the isolated capsid proteins and the assembly intermediates are not fully folded, and that association of 60 or more subunits into an icosahedral particle is coupled to progressive folding of the coat protein and also to changes in interactions with the nucleic acid. Using pressure, we have detected the presence of a ribonucleoprotein intermediate, where the coat protein is partially unfolded but bound to RNA. These intermediates are potential targets for antiviral compounds. Pressure studies on viruses have direct biotechnological applications. The ability of pressure to inactivate viruses has been evaluated with a view toward the applications of vaccine development and virus sterilization. We demonstrate that pressure causes virus inactivation while preserving the immunogenic properties. There are substantial evidence that a high pressure cycle traps a virus in the “fusion intermediate state”, not infectious but highly immunogenic. Pressure inactivation has been successful with viruses that cause disease in animals, especially foot-and-mouth disease virus (FMDV) and bovine rotavirus and humans, such as rhinoviruses, adenoviruses, alphaviruses, influenza and retroviruses.

The N-domain of troponin C (residues 1–90) regulates muscle contraction through conformational changes induced by Ca²⁺ binding. A mutant form of this domain of avian troponin C (F29W) has been used in previous studies to observe conformational changes that occur upon Ca²⁺ binding, and pressure and temperature changes. In this study we examined the effect of the point mutation on the protein structure and its stability to pressure. We performed 1-D and 2-D ¹H-NMR experiments at 300, 400, and 500 MHz on the wildtype and F29W mutant forms of the N-domain of chicken troponin C in the absence of Ca²⁺. We found that the mutant protein at 5 kbar pressures had a destabilized βl-sheet between the Ca²⁺-binding loops, an altered environment near Phe 26, and reduced local motions of Phe 26 and Phe 75 in the core of the protein, probably due to a higher compressibility of the mutant. Under the same pressure conditions, the wildtype protein experienced little effect. These results suggest that the surface mutation (F29W) significantly destabilizes the N-domain of troponin C by altering the packing and dynamics of the hydrophobic core.

Hydrostatic pressure is a modulator of biochemical processes: it inhibits bacterial growth, activates/inactivates enzymatic reactions. The application of high pressure above 3–4 kbar induces protein denaturation, whereas in the range between 1–2 kbar compression induces dissociation of oligomers[1]. It is commonly accepted that protein conformation is invariant to pressure and that inhibition of biological activity in the pre-denaturant range is due to the dissociation of oligomeric enzymes. But in recent years it has been proposed that inhibition of biological activity at moderate non-denaturating pressures might be due to the reduced flexibility of proteins rather than to the dissociation [2]. Up to date little is known about pressure effects on protein flexibility. In this paper we describe the use of the exquisite sensitivity of the phosphorescence lifetime of tryptophan to the fluidity of the environment to monitor changes in flexibility of several proteins under pressure. The results point out also important pressure-induced predissociational and post dissociational changes of protein structure

The stability of proteins under different conditions of pressure and temperature has already been the subject of many investigations. An elliptical outline of the stability curve can be suggested. In this study, we obtain the stability diagram of lipoxygenase by following the protein conformation with FTIR. Changes in the amide I ((1700-1600 cm^-1) and amide II (1550 + 1450) region were monitored. Plotting midpoints of pressure and temperature induced cooperative changes gave an elliptical outline. In addition to the inactivation measurements of Ludikhuyze, we were interested to see if the typical behaviour could also be found in the hydrogen deuterium (H/D) exchange kinetics. Herefore, we followed the ratio of intensities at 1550 and 1450 cm^-1, which are proportional to the amount of H, resp. D in peptide bonds of the protein. A similar, but not identical, result was obtained. This shows that FTIR is a well suited technique to follow kinetics and thermodynamic equilibria in the protein under different physical conditions. The results of this work are compare with recent work on the inactivation kinetics at different pressures and temperatures.

We studied the denaturation of horse heart metmyoglobin in the diamond anvil cell (DAC) with Fourier transform infrared spectroscopy (FTIR). It is observed that the conformational changes due to the cold and the pressure denaturation are similar, but not identical and that both processes do not lead to a complete loss of the secondary structure. Heat denaturation distinguishes itself from the former two processes by the formation of two new bands at 1615 and 1683 cm^-1 which are characteristic for intermolecular β-sheet aggregation. This also explains why a gel can be observed at the end of the experiment. A gel, however, can also be observed in the case of a pressure denaturation. Some aspects of pressure- and heat-induced aggregation are discussed as well.

The use of thermodynamic quantities deviate occasional from their definitions given in standard textbooks. This could lead to misunderstandings and misinterpretations. This paper reviews some definitions and gives a classification of osmotic pressure and osmotic stress experiments as well as a critical view of the lipid and protein literature.