Despite its long history as possibly the oldest continuous branch of natural philosophy and science, optics remains at the forefront of research and application. It is ubiquitous: as a tool for sensing,imaging, and communications, as well as providing ways to explore, discover, and illustrate new fundamental effects.

Light can generate conditions at the extremes of what is known to be possible according to physics, such as extremes of temperature and extremes of pressure and stress that do not exist naturally,except perhaps in the remotest of stars. And light can be used to observe and even control really fast events, such as the movement of electrons inside atoms.

Further, light can exhibit strange features associated with the quantum world, revealing even in everyday conditions some of the counter-intuitive aspects of the fitful world of randomness that underpins the stable, solid world of our normal experience. In this chapter, I will explore some of the frontiers to which, and across which, light is currently taking us. Exploration of these frontiers is possible because of the great technological strides that have been made in light sources, optical systems, and detectors, which enable exquisite control over the shape and intensity, in both space and time, a light beam can take.

Light mechanics

Light can exert forces on objects. This allows ‘remote control’ of bits of material using shaped light beams. Light can be used to move matter around, and bring it into contact with other objects,or to manipulate the internal configuration of atoms and molecules, forcing them into, for example, simple chemical reactions, in ways that allow both the study and exploitation of unusual material properties. That’s an extraordinarily powerful feature in many areas of research and study.

The concept of mechanical force arising from light has its origins in the momentum carried by each photon. For instance, when a photon is reflected from a mirror, that mirror experiences a force that provides the exertion needed to redirect the photon, just as water from a fire hose hitting a wall exerts a force on the wall by virtue of its bouncing off.

Similarly, when a photon is refracted it changes direction, and this, too, requires a force. Thus the photon exerts a force on the refracting element. If a light beam is incident on a glass bead, rays that are nearly tangential to the bead will experience the greatest change in direction. A photon traversing the ray in the lower half of the bead is directed upward as it propagates through the glass surfaces. The bead therefore experiences a force in the opposite direction. Since the momentum of the photon in the forward direction (the direction it was moving before encountering the bead) is reduced, there is also a net force in the forward direction.The strength of this force depends on how many photons are refracted per second. A light beam that is more intense at a position near the centre of the beam than on its periphery will therefore drag the bead towards the higher-intensity part of the beam.

This effect can be used to make a focused light beam into an‘optical tweezer’, which is able to hold on to a minute object and move it around as the light beam is steered. Optical tweezers find applications, for example, in biology, by enabling control of the position and movement of individual strands of DNA and the characterization of tiny molecular motors. Specifically, DNA,proteins, and other biologically important molecules can be stuck on to these beads and can therefore be handled using optical tweezers. Their position can be controlled with a precision of much less than the wavelength of light. This allows extremely small forces to be measured—such as happens when biological cells adhere to surfaces or other cells—as well as to hold the cells in places as they are operated on using other lasers (so-called cellular surgery). Optical tweezers can be combined with various other test methods, such as light scattering from aerosols, or spectroscopy, to characterize particles that may be pollutants in the atmosphere.

These opto-mechanical forces can also be used to access completely new states of motion of small objects. In particular, it is now possible to build tiny mechanical cantilevers, illustrated in Figure 31, and to both observe and control their motion using light. Light forces can be used to cool or heat the oscillations of the cantilever—like running down or winding up a watch spring—and eventually to bring it into the quietest state possible,where only quantum fluctuations of the motion disturb the complete stasis of the lever. Light forces can also be used to cool atoms—much smaller objects—and this reveals even more strange quantum states of matter.

31. A nano-scale cantilever controlled using light forces. The discs are tiny mirrors about 30 μm in diameter.

Ultra-cold

What’s the coldest thing you’ve experienced? Colder than winter in Oxford (approximately 2°C), or in Ottawa (-20°C), or the South Pole (-50°C)? Or perhaps the effects of liquid nitrogen, at-200°C. These are certainly cold, but by no means the coldest things possible. It turns out that there is a lower limit for temperature: -273°C, or 0 K (Kelvin), below which it is not possible to cool things further. This is the temperature at which things are as still as they’re going to get, with just the effects of quantum mechanics to cause a little jiggling about of atoms and molecules.

It’s not actually possible to build a machine to get to absolute zero but it is possible to get very close using an ‘optical refrigerator’.In fact, you can get cold enough to make the atoms almost stop moving. What this means is that their size gets bigger. (Quantum mechanics tells us that you can’t simultaneously specify the precise location and speed of an object. If the atom is completely stopped,then it must be extended over all space.) Therefore all atoms in the cloud that has been refrigerated occupy the same region of space,and this gives rise to some very strange new phenomena.

An optical refrigerator works by using lasers to ‘cool’ atoms.Imagine a laser beam shining on an atom that is moving from left to right, say. The laser shines from the right to left, so that a stream of photons hits the atom. The laser is tuned in frequency to be absorbed by atoms that are moving at a particular velocity.Now, when the atom absorbs a photon from the laser beam, it gets a kick from the photon, and thus slows down. (More specifically,the momentum of the photon is transferred to the atom. Since it is in the opposite direction to the initial momentum of the atom, it reduces the momentum, and thus the speed of the atom.) The atom must re-emit the photon at some later time, and it will get a kick in the opposite direction to that in which it emits the photon.But the direction in which it re-emits the photon is random—it can go in any direction at all.

If you look at enough of these absorption-scattering events, then you will find that, although the light is always absorbed from one direction (the incoming laser beam) it is emitted uniformly in all directions—no one direction is preferred. The consequence of this is that on average a collection of atoms moving in a direction opposite to the laser beam grinds to a halt, and is left with random motion representing a temperature that is proportional to how long it holds on to the light before re-emitting it.

There are several refinements of this approach, each of which uses light to cool atoms (and molecules) to even lower temperatures,and for which light acts like a ‘viscous fluid’ in which the atoms move slower and slower. It is even possible to use light to trap atoms using optical tweezers once they are slow enough. This allows the application of yet more sophisticated optical cooling techniques, by which it is possible to get to temperatures of a billionth of a degree above absolute zero.

I referred to some residual ‘jiggling’ of the atoms that happens even at zero temperature, arising from quantum mechanics. The range of this jiggling can be thought of as the spatial extent of the atom itself. That is, according to quantum mechanics the atom isn’t just wandering around in a random fashion over a small region of space, but rather it is actually present across all that region at once. For atoms trapped at such low temperatures the size of that region may be several thousandths of a metre. That’s a remarkably large atom, given that the distance of the electron from the atomic nucleus is less than one tenth of a billionth of a metre. What’s even stranger is that several atoms can occupy this region of space at the same time.

That’s conceptually very counter-intuitive. We often think of atoms as being like little billiard balls, that can be packed close together, as in a solid material, but which retain their individual distinction by virtue of their location inside the material. That’s not the case for these very cold atoms. They are each everywhere at once, in a new state of matter identified by Einstein and the Indian scientist Satyendra Nath Bose and called, not surprisingly,a Bose–Einstein condensate.

This very strange state has some remarkable properties. For instance, it is a superfluid, which flows without viscosity. Further,it is possible to split the entire atomic cloud in half and recombine it to show quantum interference between the two separated parts,essentially demonstrating that a big object (containing many atoms and of a palpably visible size) exhibits quantum character,attributable to the uncertainty of whether an atom is in one part of the cloud or the other. One has to think of each atom occupying both separate components at the same time.

Because these cold atoms can be trapped in light beams, it is also possible to create spatial structures out of several light beams that can be used to manipulate the atoms. For instance, when two light beams coincide they form an interference pattern (see Chapter 3)in which there are regions of high and low intensity. Cold atoms like to settle in one or other of these regions (you can adjust which one by choosing a particular wavelength of the light). As the intensity of the light beams is turned up, the atoms fall into the‘egg-crate’-like optical traps that appear in the intensity pattern, as shown in Figure 32a. And they do so in very interesting ways.

32. Cold atoms trapped in an optical lattice: a. a few hundred atoms per cell (several 10s μK), b. individual atoms at each site (a few nK).

When the atoms are cold enough, they don’t like to be located at the same ‘site’ in this egg-crate, so the resulting distribution of atoms is very like a full egg carton—one atom at each site, as shown in Figure 32b. In this case there is no superfluidity, since the atoms like to stay put. It is more like an ‘insulator’, as nothing moves. By turning the light intensity up and down it is possible to explore this interesting transition between completely free flow and no flow at all.

The ability to do this in a system that is fully quantum mechanical allows scientists to explore new properties of matter that are relevant to other types of materials (for example, solid-state metal oxides) over which it is difficult to exert the same degree of precision control and measurement. In cold atomic gases it is now possible to look at atoms in these egg-crates individually and see what they are doing as changes are made to their environment.

It is possible to explore this low-temperature regime with many different kinds of atoms, and to build complicated trapping structures using light. The idea of using cold atoms to ‘simulate’other quantum systems is a current area of research. It allows exploration of complex problems that cannot be solved in other ways, and is expected to lead to a new understanding of materials and structures that will have impact in new ways—perhaps helping to understand and even design new magnets that will be used for applications such as data storage for computers,magnetic resonance imaging machines for healthcare, or even friction-free motors for levitating trains.

Ultrafast

Light pulses can be extremely brief. In Chapter 5, I stated that they can be as short as a single cycle of the optical field. For light in the visible region of the spectrum, that’s about 2 fs. For light in the extreme ultraviolet region, which is of shorter wavelength and higher frequency, the durations can be much shorter. The shortest yet measured is less than 100 as (10-18 s) long. These are currently the shortest pulses that can be controllably generated (although we can observe events that happen on a much shorter timescale by means of particle colliders). And with the advent of bursts of light in the X-ray region of the spectrum, we can expect that even shorter timescales are possible.

Since these numbers are so mind-bogglingly small, it is helpful to put them into context. The age of the universe is approximately5 × 1017 s. Thus the ratio of one second to the age of the universe is approximately the same as the ratio of one attosecond to one second. Or to put it in an economic context, if the national debt of the US is equivalent to a second, then one cent would be equivalent to a femtosecond. On this scale, a single attosecond is virtually worthless.

What sort of things can happen on this timescale? In Chapter 4,I introduced a simple model of an atom—called the Bohr model—in which electrons ‘orbit’ an atomic nucleus, attracted to it by electric forces in much the same way as planets orbit the Sun, attracted by gravitational forces. The time taken to execute these orbits for simple atoms (that is, those with only a few electrons) is about 150as. So if we want to look at this motion, we might need to use pulses shorter than this, so we don’t just see a blur.

The idea of the stroboscope is the one most relevant to our story,since a variant is used by researchers today to look at the really speedy changes that go on at the fundamental level of atoms and molecules. In this application the light pulses from a laser are split into two (or more) parts, with a delay introduced between them.The first pulse in the sequence illuminates the sample, and some of it is absorbed. This ‘triggers’ some changes in the system—electrons move around in the atom, or bonds vibrate in molecules or solids. An instant later the second pulse arrives and some of it is again scattered from the sample and detected.

As the delay between the two pulses increases in repeated runs of this experiment, the detected scattered light maps out the dynamical changes of the sample. In a sense it makes a ‘movie’ of the atom or molecule or solid as it changes. This ‘pump-and-probe’ scheme has been used to get into the guts of what happens, for instance, during a chemical reaction, when two molecules are reconfigured by their interaction. More sophisticated versions of this kind of approach exist using several light pulses. These approaches are being used now to explore many fundamental features of extremely interesting and puzzling materials, from interacting atoms and high-temperature superconductors to biological systems.

I’ve noted that the shortest pulse it is possible to generate is a single cycle of the optical field. It turns out that you can devise an experiment to measure the oscillations of the optical electric field using the extreme ultraviolet (EUV) pulses produced by highharmonic generation. What is needed to measure the pulse field is a very fast process, one that is much faster than the optical cycle itself.That can be provided by a pulse with much shorter wavelengths,about twenty or thirty times shorter than that of the optical wavelength. Pulses of such brevity are generated when an electron is ripped off an atom by means of a strong optical pulse. This requires an optical field whose strength is comparable to the binding force of the electron to the atomic nucleus. Such pulses are readily available by adding optical amplifiers to the output of a mode-locked laser.

When the electron is liberated by such an intense pulse, it finds itself sitting in a rapidly oscillating electric field, and, if its liberation occurs near a time when the field has zero amplitude,the electron can ‘surf ’ along the next cycle of the optical wave,taking an excursion away from the atom, and then back again.When it returns it is moving very fast, and when it recollides with the atom it can be recaptured by emitting all of its extra energy as light. In this case a very short pulse is emitted as the electron recombines, having a very short wavelength of perhaps a few tens of billionths of a metre, in the EUV region of the spectrum, about twenty times shorter than the optical wave that generated it.

Now imagine that this EUV pulse shines on another atom. It has a sufficiently short wavelength such that it is absorbed by the atom,and knocks off an electron, which then sits near the atom. Imagine further that the atom is simultaneously illuminated with the short optical pulse we seek to measure. The field of this pulse accelerates the electron in one direction or another depending on the part of the optical cycle at which the electron is liberated by the EUV pulse. By changing the delay between the EUV pulse and the optical pulse, the acceleration of the electron can be measured since faster electrons, which have been accelerated to a greater degree, have more energy. In this way it is possible to ‘see’ an optical pulse field (Figure 33), despite the extraordinarily short timescale of the oscillations of the field.

An example of an application of the methods of pump-and-probe spectroscopy in biochemistry is the study of the first steps in the process of photosynthesis, by which plants convert carbon dioxide from the air into oxygen using sunlight as an energy source. The processes by which this happens involve transporting energy around a big biological molecule with remarkably high efficiency.The means by which this happens has some very interesting and poorly understood features—it is faster than one might expect and much more efficient. If we could learn from systems that have evolved naturally over eons how to do this, perhaps our understanding would enable us to apply it to things like improving the design of solar cells, which would have an enormous impact on society.

33. Direct image of the electric field of an optical pulse. The time between two adjacent peaks is approximately 2.5 fs.

Ultra-intense

Your electricity bill tells you how much energy you used in the previous month. It is measured in units of kilowatt-hours (kWh),and you are charged for each unit that you consume. Let’s say that you used 220 kWh in some month (this is the average monthly energy consumption in the UK). Now, you could use all this energy at the same rate across all four weeks of the month. Or you could use it in the first week, and use nothing in the subsequent three weeks. But can you imagine using it all in a million billionth of a second? You’d need to have an awful lot of appliances to use that much energy in such a short time, and you’d need to be able to switch them on and off impossibly quickly. But the peak rate or power would be immense.It is possible to produce light pulses that can achieve this. That is,they are incredibly brief and contain this amount of energy. In fact, it is possible to produce a pulse that delivers energy at a rate equal to the entire electricity generating capacity of the planet at a given instant. But the lights in your house won’t go out, because the pulses are so brief that the total energy in them is very small.

Lasers that produce pulses of this kind are massive instruments,occupying large buildings that are a significant fraction of a football field in size. One example is the VULCAN laser at the Rutherford Appleton Laboratory in England. VULCAN produces pulses with 500 Joules of energy (1 J = kWh) in a pulse of 500 fs duration. 500 J is the energy emitted by a 100 W light bulb in five seconds. Yet the brevity of the pulse means that the intensity of the light can be as high as a million suns. The laser at the National Ignition Facility (NIF) in Livermore, California is much, much bigger than this. And the proposed European Light Infrastructure project is set to deliver a system capable of even greater peak power than that at NIF.

Such very brief, very intense light bursts can be used to alter states of matter. The electric field at the most intense moment of the light pulse is larger than the attractive field between the electrons and nuclei that holds atoms together. So it is possible to strip the electrons off the atoms to form a new state of matter—a plasma.And it’s possible to do this in an instant, shorter than the time over which the atomic nuclei can move, so that the plasma is very dense—nearly the same density as in a solid block of material such as a piece of glass, except now at two million degrees Celsius.

These are the conditions inside the cores of the giant planets and even some stars: very high-density plasmas with particles colliding with each other at high speed and at pressures of a million times that of our own atmosphere. It is possible to use this new laboratory-accessible state of matter for several things. For instance, we can begin to understand how stars work, what their life cycle is, and to characterize their stages of evolution, such as supernovae explosions and white dwarfs. Other situations of interest to astrophysicists are also amenable to experimental exploration using lasers. Astrophysicists also use such plasmas to probe the frontiers of planetary science. For instance, it is possible to infer the composition of the gas giants from their mass and size,but only if the degree to which matter can be compressed at such high pressures is known.

Some laser facilities generate pulses using light of very short wavelengths. These pulses are produced by accelerating electrons in magnetic fields, so that they ‘wiggle’ from side to side as they whiz down the accelerator. This produces a kind of synchrotron radiation consisting of short bursts of X-rays. Often these kinds of lasers use techniques, and indeed hardware, from particle accelerators. Examples of these are the Stanford Linear Collider Light Source (LCLS) and the Hamburg X-ray Free Electron Laser (XFEL).

The techniques afforded by the most intense laser pulses, in conjunction with short bursts of X-rays, enable scientists to diagnose plasmas in a wide range of conditions. Further, the immense pressure exerted by lasers between atomic nuclei can cause them, under the right conditions, to fuse together, releasing a large amount of energy in the process. This ‘nuclear fusion’ is a possible route to an almost unlimited source of energy. This application of lasers is extremely technically demanding, and is one of two methods that are being explored to achieve fusion:the other does not involve light except as a way to monitor the process. Both make use of dense plasmas.

As the laser pulses move through these plasmas, they generate a wave, very similar to the wake that trails a boat moving across a water surface. The electric fields in the plasma wave can reach more than 100,000 V over 10-6m (that’s about ten times the voltage in the big power lines on pylons, over a distance of less than one tenth of a human hair’s breadth.) These field strengths are at least 1,000 times bigger than the accelerating fields used in the world’s biggest machines for studying fundamental particles,such as the Large Hadron Collider (LHC) at CERN in Geneva. It may be possible using laser methods eventually to build table-top devices that can accelerate electrons to similar energies as currently possible in the LHC.

It is also possible to accelerate heavier particles, such as protons,by means of the extraordinarily strong electric fields generated by the interaction of laser pulses with matter. Proton beams are currently being explored as a cancer treatment, since the delivery of heavy particles to diseased tissue can be done more precisely and at greater depths than is currently possible using other types of radiation therapy.

The extraordinary properties of light continue to enable new realms of discovery, across a wide range of fields. Light is a ubiquitous tool for science and technology.