A supercontinuum generated by focusing an 800 nm femtosecond pulsed laser into a yttrium aluminium garnet crystal. Photo: Wikimedia

The present age of information technology – the transformation of daily life by laptop computers, smartphones, so-called artificial intelligence, etc – became possible thanks to the exponential increase in the processing power of microcircuits, which began in the 1970s and continues today.

This process is described empirically by the famous Moore’s law: the number of transistor elements that can be packed into an integrated circuit chip doubles about every two years.

Few people are aware, however, that a analogous process has been taking place in laser technology. The intensities of the light pulses which lasers can deliver has been increasing exponentially since the first laser was built in 1960.

It has become possible to concentrate larger and larger amounts of energy into shorter and shorter pulses, down to a million of a billionth of a second or less. We have entered the world of “extreme light” – forms of light which, as far as we know, have never existed in our solar system up to now.  

I would venture to predict that the impact on the future of society may turn out to be as radical as the information revolution, but in a completely different way.

To use a grandiose and politically incorrect expression, we are talking about an increase in Man’s power over the physical universe. Information technologies are fantastic, indispensable tools; but by themselves, they can’t power cities or get you to Mars. Laser fusion can.

The new, ultra-high-power, ultra-short-pulse lasers will be able to do much, much more. And, by the way, these lasers are key to keeping Moore’s law going, through improvements in lithography of microchips and other applications. 

As one can see from the diagram, the maximum intensity that can be attained by focusing a laser pulse grew extremely fast in the first few years. This was followed by over a decade of much slower progress, only to explode upward beginning in 1985.

That is the point marked “CPA,” the discovery of the technique of “chirped pulse amplification,” which I shall describe in a moment. CPA marked a revolutionary breakthrough in the ability to amplify laser pulses. Since then laser intensities have grown by a factor of 10 every five years.

How short is a moment of time?

Laser technology not only permits light energy to be concentrated in space – by focusing a beam onto a tiny area – but also in time. Lasers are now commercially available, which produces pulses of light lasting no more than a few femtoseconds. A femtosecond is a millionth of a billionth of a second.

To get an idea of the incredible shortness of a femtosecond: the ratio of a femtosecond to a second of time corresponds to that between 1 second and 770 million years. 

In ongoing research, pulse lengths are being pushed down below a femtosecond into the attosecond region, a thousand times shorter. This reaches into the range of fundamental physical processes occurring in individual atoms. For the first time, it is becoming possible to make something like a slow-motion video of the motion of individual electrons in an atom.

 Femtosecond laser pulses leave no time for heating. Right: Femtosecond laser for opthamalic operations. Photo: Wikipedia

The standard methods for producing ultrashort pulses down into the femtosecond range – so-called Q-switching and mode-locking – were developed already in the 1960s. The big challenge was: how much energy can one concentrate into such a pulse?

As a rule, an ultrashort-pulse laser system consists of a laser oscillator followed by an amplifier, which amplifies the initial, relative weak pulse into a powerful one. Efforts to increase the pulse energy confronted an obstacle, however.

Beyond a certain power, the amplifier and optical system suffer catastrophic damage. Among other things, sufficiently intense laser light tends to “self-focus” when passing through a medium, reaching energy densities which no material can withstand.

In the mid-1980s, however, the physicist Gérard Mourou and his (then) PhD student Donna Strickland came up with an ingenious solution to this problem, known as “chirped pulse amplification (CPA).

This invention, which opened up a whole new era of laser development, earned them the Nobel Prize in Physics in 2018.


The term and basic concept of “chirp” originated in the area of military radar systems around the end of WW II, and was developed operationally in the 1950s. The concept was first declassified in 1960. I shall briefly sketch the idea, which is quite interesting.

Radar systems working with a fixed frequency, in striving to increase their range and resolution, were constrained by the need to produce very short pulses with very high powers. But with more and more amplification, components reached their damage limit.

 In “chirp radar” the pulse emitted by the radar transmitter does not have a fixed frequency, but is frequency-modulated – “chirped.” In the course of the pulse, the frequency decreases from a chosen initial value to a lower one (or the opposite) before cutting off. The pulse is then relatively long, permitting it to be amplified to high powers without damage to the transmitter electronics.

The signal, going out to the target object and reflected back will arrive with the same frequency-modulation. (The slight frequency shift due to the Doppler effect for a moving object such as an airplane, has only a marginal impact in most contexts.)

Now comes the trick: the radar receiver contains a “delay network” designed in such a way, that reflected higher-frequency signals take longer than lower-frequency signals to get through the network.

As the reflected frequency-modulated signal travels through the network, the initial, higher-frequency parts of the signal are delayed just enough so that the lower frequency parts can catch up with them. This results in a “pile-up” of energy: the pulse coming out of the delay network is greatly compressed in time relative to the original one.

One can appreciate that the resulting, very short pulse contains the same information about the object, as if the original pulse would have been equally short. From the synthesized pulse the radar receiver can measure the precise time delay of the signal in traveling to and from the target, revealing its distance to the target.

The difference is that the much longer frequency-modulated signal emitted by the “chirp radar” can have much, much more energy than could be transmitted in a short pulse. The radar can “see farther” and attain a higher precision.

Schema of a chirped radar system – originally a military secret.

The birth of ‘extreme light’

It was first in the mid-1980s that Gérard Mourou and his then-student Donna Strickland, working at the University of Rochester Laser Laboratory, succeeded in applying the “chirping” idea to the problem of amplifying ultra-short laser pulses (see the diagram below).

In their scheme, the role of the delay network in chirped radar is played by a pair of diffraction gratings. These separate different frequencies of light, contained in the short pulse, into a “rainbow” of frequencies and force the higher (“bluer”) frequencies to go through a longer pathway than the lower (“redder”) frequencies.

This causes the higher frequency components to lag behind as the pulse travels into the amplifier. The original, weak but ultra-short pulse produced by the laser oscillator at the left, is stretched out by the first pair of diffraction gratings into a 1,000-or-more-times longer pulse of rising frequency, e.g. beginning in red and ending in blue.

The resulting “stretched” pulse runs through a laser amplifier, where the total energy can be increased to gigantic powers without damage to the amplifier medium. Finally, a second set of diffraction gratings, operating in reverse, causes the long pulse to “reassemble” at the exit of the system into a single, ultra-short pulse of gigantic power. Amplification factors of trillions or more times can be achieved.  

Thanks to the advent of chirped pulse amplification, femtosecond-range lasers with powers in the range of petawatts – a million billion watts – are in operation today in laboratories around the world.

When a pulse from such a laser is focused onto a target, wild new things happen, phenomena from the domain of nuclear physics and elementary particle physics: transmutation of atoms and other nuclear reactions, generation of ultra-high energy (relativistic) particle beams and the acceleration of macroscopic objects (plasma blocks) to velocities of 1,000 kilometers per second, and more.

Ironically, despite their huge intensity, the pulses – so long as they have sufficiently sharp edges (so-called high contrast ratio) – generate practically no heat when they interact with matter. Instead, their energy is transformed into organized, directed processes. Heat has no time to develop, and appears only later when the processes subside.

As laser intensities continue to grow at an exponential rate, theoretical physicists are dreaming of reaching intensities at which so-called vacuum breakdown, where matter is created “out of nothing” from energy alone, can be directly observed.

Internationally the race is on to reach higher and higher pulse powers, pulse intensities and number of “shots” that can be produced per minute. The European Extreme Light Infrastructure (ELI) project will provide laser powers in the 10 petawatt range. China’s Shanghai Superintense Ultrafast Laser Facility (SULF) plans to reach 100 petawatts by 2023.

Russia’s projected Center for  Extreme Light Studies (XCELS) is supposed to reach 200 petawatts, with the future option of going up to exawatt levels (1,000 petawatts, or a billion, billion watts). The US is scrambling to regain leadership with the Zettawatt-Equivalent Ultrashort Pulse Laser System (ZEUS) to be built at the University of Michigan.

ZEUS will use colliding laser and electron beam pulses to reach total powers of the order of a zettawatt (a billion billions of billions of watts).

What is the ordinary citizen supposed to get out of these dizzying numbers? There is no lack of promises, generally credible, concerning practical applications such as new materials and production methods, applications to medical diagnosis and treatment, elimination of nuclear waste, and not least of all nuclear fusion.   

However, realizing the most advantageous known form of fusion energy – the radioactivity-free hydrogen-boron reaction – does not depend on the above-mentioned megaprojects.

It looks very much like the reactor concept put forward by the Australian plasma physicist Heinrich Hora, which has attracted growing attention in recent years, will need little more than lasers of the sort already operating in laboratories around the world. And probably much more compact. In the following installment, I shall explain how it works.

Jonathan Tennenbaum received his PhD in mathematics from the University of California in 1973 at age 22. Also a physicist, linguist and pianist, he’s a former editor of FUSION magazine. He lives in Berlin and travels frequently to Asia and elsewhere, consulting on economics, science and technology.