Video shooting the split second
(appeared in August 2014)

(link to main website)

A way has been found to take moving clips of super high speed events, says S.Ananthanarayanan.

‘Winning by a nose’ is a phrase that has come from horse racing, where a near dead-heat was decided by a high speed photograph. Science and technology have done much better, and with fast shutter speed and quick reacting film, pictures of milli-second changes are routine. With more complexity, even micro-second events, and now, things that last a femto-second, which is a billionth of a millionth of a second, can be captured.

But the trouble with the fastest methods is that each picture is instantaneous and many such pictures have to be taken before a workable image of a process can be formed. These methods are then not good enough when the event in question cannot be repeated a number of times, for many pictures to be taken. K. Nakagawa, A.Iwasaki, Y. Oishi, R. Horisaki, A. Tsukamoto, A. Nakamura, K. Hirosawa, H. Liao, T. Ushida, K. Goda, F. Kannari and I. Sakuma, of different universities or institutes in Japan, and also from California, report in the journal, Nature Photonics, a method of exposing a single event to a rapid series of flashes and capturing each one separately, so that the images take their time to form, with good quality.

Current art

Events in sports have time-scales of hundredths of a second, and even movements of insects, like the wings of a honeybee, or events like a hammer stroke, or splintering of glass, happen in thousandths of a second. But processes in physics or chemistry are much faster and following them needs methods of finer time resolution. The most popular method, the ‘gold standard’ of high speed, is the pump-probe, to study the spectrum of that emerges from a substance a split second after it has been exposed to a flash of light, and the time gap can be as low as a ten billionth of a millionth of a second, or a tenth of a femtosecond. In a typical case, the substance is exposed to a pulse, called the pump, from a laser and the experiment is to study how the substance affects the following laser pulse, which is the probe, the effect studied being, for instance, the way a particular frequency of light is absorbed.

Refinements of this method are when more than two pulses are used, or when the pump and the probe are different, like the pump in the ultra-violet and the probe in the infra-red. Yet another method is to monitor the fluorescence, which is the emission of light of some lower frequency, after excitation by the higher frequency ‘pump’. This is the method that captures changes down to the picosecond, or a millionth of a millionth of a second.

These methods, though undoubtedly fast, record the state of the system at a given interval after the excitation event. To get a dynamic picture of how the system develops in time, the trial has to be repeated with gradually increasing time-gaps between pump and probe. This calls for the event to be both simple as well as capable of being repeated. Now processes like the movement of molecules of a gas, explosions or destructive testing, interaction with living cells, protein folding, are examples where each instance is unique, and repetition with increasing time gap will not correspond to the earlier trial with an earlier instance of the process. Pump-probe, hence, often cannot be used.

The new way

.txtWhat the authors of the Nature photonics paper report is a way to capture the dynamics of a process with a single flash of light, and during the course of the same instance of the event. The method is called sequentially timed all-optical mapping photography – STAMP – and it consists of capturing the changes in the target in an ultra-fast series of images and then to separate the images, so that they could be viewed as a sequence. Separating the images allows the images to be detected by slow, but sensitive processors. This permits high speed photography, but with low intensity illumination, so that the process that is being studied does not get affected by the series of probes that is used.

The arrangement is as shown in the picture – the illumination is by a single pulse of light from a laser. The light first passes through a device that splits the pulse into a series of time separated pulses. How does this happen? This is by passing the light through a medium like glass. The components of the slight frequency spread of the laser pulse have different speeds of passing through the glass. The pulse is hence stretched out, the component to the violet end is slowest while the component to the red end leads, when the pulse emerges. If it is a case of a glass rod, the separation between components could be in hundreds of femtoseconds. Else the pulse could be passed through pairs of prisms. Here, again, the component to the red end moves fastest and surges into the lead. The separation is greater in a prism and the separation is in picoseconds. If the pulse is passed through a length of optical fibre, then there is even more separation, in the range of nano seconds, which is of thousandths of a millionth of a second.

The light coming from the laser is thus split in time when it passes through this first device, which is called the temporal mapping device (TMD). The pulse then falls on the target, and the components of the pulse srike the target one after the other, or in sequence, separated by intervals as low as in femto-seconds. The light that emerges then has a record of the state of the target at the times that each of the components of the initial pulse reaches the target. Now the second device, the spatial mapping device (SMD), enters the action. The SMD is basically a disperser, like a prism, or a device called a grating, which separates the different frequencies, which in fact, correspond to the different temporally separated components. The components that are separated by the SMD can now be directed to different sensors, as records of the state of the target at intervals, which may be separated in femtoseconds, and the records can be combined, to recreate a dynamic motion picture of the behavior of the target.

The authors of the paper, tried it out to picture how ionized gas behaves split seconds after a laser caused micro-explosion, like happens when the laser is used in surgery. Or when a laser beam strikes a glass surface.

As shown in the pictures STAMP could yield dynamic pictures of what happened. This demonstrates the use of STAMP, which is a versatile and generic imaging procedure which can be used with different imaging methods and at different frequencies, say the authors.