By Kenneth Freeman
‘Extreme’ is a relative word. At 250m below the surface of the ocean the pressure is 26 times higher than at the surface. This is about as deep as a human has ever free-dived, but is far too shallow and at too low a pressure for the blobfish that lives comfortably around 1km deep.
Pressure isn’t as intuitive to us as, say, temperature. The few times we notice it could be feeling your ears pop on takeoff as the cabin pressure quickly (and only slightly) drops, or feeling the building pressure as you dive towards the bottom of a pool.
The standard scientific unit for pressure is the Pascal (written Pa); typical atmospheric pressure (1 atmosphere) is around 1×105 Pa (100 kPa), the pressure at the top of Everest is about a third of that and the pressure at the bottom of a 6m deep pool is around 160 kPa.
Low Pressure – environments under vacuum
Let’s start at the lower end of the scale; how do we reduce the pressure of something?
First of all, let’s think about exactly what low pressure means. Consider an object, sitting in a sealed box, surrounded by gas. The pressure exerted by the gas on the object can be thought of as a result of the gas particles hitting the sides of the object and exerting a force on it. The more gas particles there are in the box, the more collisions with the sides of the object and the higher the pressure on the object. If we start to remove the gas particles from the box then collisions occur less frequently and the pressure on the object is less.
We typically use low pressures to create clean environments by removing most of the particles from a volume i.e. putting it under vacuum. Removing the particles reduces friction, reduces chemical degradation and allows for the free passage of particle beams, like in cathode ray tubes (remember them?). The ability to achieve high vacuum is vital to the function of electron microscopes and in the processes of coating and etching materials with nano-scale precision.
To describe the quality of vacuum, we talk about it either in terms of pressure or the mean free path of the remaining particles (which is the average distance a particle travels before it encounters another particle). ‘High vacuum’, for example, is defined as where the mean free path of particles in the sample chamber is longer than the size of the chamber. Ultra-high vacuum (UHV), which is what we most often use in the sciences, describes vacuum pressures below 10-6 Pa.
The first challenge in producing ultra-high vacuums is an engineering challenge – building the vacuum chamber. The presence of any contaminants inside the chamber will ruin the vacuum so close attention has to be paid to the chamber design. The surfaces in the chamber must be non-reactive and not allow for any trapped pockets of gas, requiring the use of clean welding techniques and specially designed bolts (as the space between the threads can trap tiny amounts of contaminants). Even the chamber materials themselves must be carefully selected to reduce the outgassing of any adsorbed or trapped gas in the materials.
With a suitable, clean chamber the next step is to remove as many of the particles from inside the chamber as possible. This is done using a series of pumps. First, a positive displacement pump removes most of the atmosphere from the chamber to give a rough vacuum. This pump works by expanding a cavity (increasing the volume) into which the atmosphere from the chamber flows due to the resulting difference in pressure. It then seals off and evacuates this cavity and repeats, leaving less atmosphere in the sample chamber. The same principle as a bicycle pump.
Once we start to reach lower pressures this type of pump becomes less efficient. To continue reducing the pressure we need a molecular pump; for example, a turbomolecular pump. This looks like a little a jet engine, and works by rotating a set of fans to accelerate particles from one side of the pump to the other. As there is no hard seal between the sample chamber and the outside, the outside end of the pump must be held at low pressure in order to avoid atmosphere rushing back in.
Finally, we get to seriously low pressures using even more specialised pumps: usually either ion pumps or titanium sublimation pumps (TSPs). Ion pumps use high electrical potentials between electrodes in the sample chamber to create a Penning trap – a cloud of electrons which ionises any gas particles in the chamber. These ionised particles are then accelerated by the potential and either embed themselves in the solid electrodes, or they sputter some of the cathode material on to the walls of the chamber. The sputtered cathode material acts as a getter – a chemically active material which removes gas particles by chemical reaction or adsorption, thereby reducing the pressure in the chamber. TSPs work in a similar way, with a high current passed through a titanium filament to sublimate a thin film of titanium onto the inner surfaces of the chamber; this film then acts as a getter.
High Pressure – static & dynamic
Compressing a material under high pressure forces its atoms closer together which changes the way they interact and bond with one another. This means that high pressure alters the atomic and electronic structure of a material which can result in interesting, surprising and (hopefully) useful changes in its physical and electronic properties.
Moderately high pressures are of interest to engineers who want to know how materials will behave under their working conditions, such as in the design of engines or civil engineering projects. Extremely high pressures are of more interest to fundamental science and the study of planetary interiors.
There are plenty of well-established methods to produce static pressure up to around 100 GPa (for reference, the centre of the Earth is at around 360 GPa) with more recent methods able to produce static pressures up to 1 TPa (1000 GPa), and dynamic pressures up to hundreds of TPa. But how do we get there?
There are two routes available: static compression and dynamic compression.
Static compression relies on high-school physics:
If we can only produce a certain amount of force, then to increase the pressure we have to reduce the sample volume (reducing the area over which the force acts). Various anvil set-ups squeeze a sample between opposing hard anvils, achieving higher pressures by using smaller and smaller sample volumes; eventually the limiting factor becomes the hardness of the anvil materials.
The highest static pressures are generated using a diamond anvil cell (DAC). The small, flat culets (tips) of two opposing diamonds are used to compress a sample by pushing the diamonds together using mechanical force (typically manual Allen screws or an expanding gas membrane). We take a thin metal foil and drill a tiny hole in it using a laser or spark-eroder. This hole is smaller than the size of the flat diamond culets and is placed between them to form a sample chamber. In the chamber goes a small sample, along with a liquid pressure transmitting medium which ensures the sample experiences hydrostatic pressure i.e. even pressure from all sides.
We can get up to 100 GPa or so using a typical DAC before the gasket or the diamonds fail (either with the sample chamber bursting open, or with a sickening crack as you break several hundred Euros worth of diamonds).
Now that we’re at the limits of our DAC materials, what else can we try to reach even higher pressures? We can produce ultra-hard diamond, but it’s difficult and would be far too expensive to make a full-size DAC using them.
Well, what happens if we put a smaller DAC inside another DAC?
That’s essentially the method behind two-stage diamond anvil cells. Between the culets of a standard DAC, two semispheres of ultra-hard diamond are placed, opposing one another as in the normal DAC. It is between these tiny diamond anvils that a sample is held directly. Work has been published showing the use of nanocrystalline second-stage anvils that have produced pressures up to 1 TPa (ten million times atmospheric pressure) in tiny samples 3 microns across and 1 micron thick.
Diamond anvil cells are very important in the field of high-pressure science. They allow the relatively easy generation of large pressures but at the expense of sample size. The diamonds are not only hard enough to create such pressures, but they are transparent to most electromagnetic radiation. This allows the samples to be easily viewed and studied using x-ray crystallography and Raman spectroscopy, and provides routes for the laser heating of samples under pressure.
To achieve the very highest pressures we need to use dynamic compression methods.
As we mentioned in the Extreme Temperatures article, the shock waves produced in inertially confined fusion produce extremely high pressures in the fuel material. This principle has been used directly to study matter under extremes only seen in the interiors of planets. A sample is prepared with an ablator backing (e.g. plastic) then intense laser light is then fired at the ablator causing it to explode outwards, sending a shock wave through the sample in the opposite direction. If an x-ray beam is directed through this `shocked’ sample at the right moment, the high-pressure portion of the sample can be studied.
Such techniques have successfully produced pressures of up to 5 TPa, with methods to shock pre-compressed samples inside of a diamond anvil cell being developed to regularly reach the 10-100 TPa pressure range. This will allow scientists to probe matter under pressures found in planetary interiors and view exotic states of matter under extreme conditions.
The engineering challenges posed by studying materials at high and low pressures have been overcome in many ways. This has allowed experiment to largely keep up with theory in the field of high-pressure science and for new technologies to be developed in clean high-vacuum environments. This trend of gradual progress is likely to continue, with major facilities’ ongoing upgrade programs and the development of better technologies and methods to get the most out of experiments performed under extremes of pressure. While the limits to the pressures we can produce will no doubt be pushed when required, the field of high-pressure research will probably focus on improving the quality of experiments and the data obtained from them.
DAC picture courtesy of Dr. Ingo Loa.