Researchers at Lund University have developed a video camera that can film at a rate equivalent to five trillion images per second, recording events as short as 0.2 trillionths of a second. The ultrafast camera, which enables 2D videography of single, nonrepetitive events at timescales of femtosecond (fs) and shorter, could be used to capture extremely rapid processes in chemistry, physics, biology and biomedicine that so far have not been caught on film. Existing high-speed cameras capture images sequentially, one image at a time. The novel camera technology, called Frequency Recognition Algorithm for Multiple Exposures (FRAME), is based on an algorithm that enables the camera to capture several coded images in one picture, then sort the images into a video sequence. An object of a FRAME video — for example, a chemical reaction in a burning flame — is exposed to light in the form of laser flashes. Each light pulse is given a unique code, as a form of encryption. Every time a coded light flash hits the object, it emits an image signal response with the same coding. The light flashes that follow will all have different codes. These coded image signals are subsequently separated using an encryption key on the computer. The coding strategy enables laser probes with arbitrary wavelengths and bandwidths to collect signals with indiscriminate spectral information, thus allowing for ultrafast videography with full spectroscopic capability. Current imaging methods that extend into the fs regime are capable of only point measurements or single snapshot visualizations, and thus lack the capability to perform ultrafast spectroscopic videography of dynamic single events. To demonstrate the high temporal resolution of the FRAME method, researchers from Lund University filmed light propagation with record-high 200-fs temporal resolution. To illustrate the technology, the researchers filmed how light travels a distance corresponding to the thickness of a paper. In reality, this process only takes a picosecond, but on film the process has been slowed down by a trillion times. Courtesy of Lund University. The researchers believe the FRAME method is applicable to the study of many dynamic processes in physics, chemistry and biology over a wide range of timescales. The camera is initially intended for use by researchers who want to gain better insight into processes that take place on a picosecond (ps) or fs scale. “This does not apply to all processes in nature, but quite a few, for example, explosions, plasma flashes, turbulent combustion, brain activity in animals and chemical reactions. We are now able to film such extremely short processes,” said researcher Elias Kristensson. “In the long term, the technology can also be used by industry and others. “Today, the only way to visualize such rapid events is to photograph still images of the process. You then have to attempt to repeat identical experiments to provide several still images which can later be edited into a movie. The problem with this approach is that it is highly unlikely that a process will be identical if you repeat the experiment,” he said. The team’s primary area of research is combustion, which is controlled by a number of ultrafast processes at the molecular level — and which can now be captured on film using the FRAME camera. For example, the researchers can use FRAME to study the chemistry of plasma discharges, the lifetime of quantum states in combustion environments, and how chemical reactions are initiated. Because the minimum frame separation is dictated by only the laser pulse duration, the researchers believe that attosecond-laser technology could further increase FRAME video rates by several orders of magnitude. A German company has developed a prototype of the technology, which could mean that the technology will be more widely available within an estimated two years. The research has been accepted for publication in Light: Science & Applications (doi: 10.1038/lsa.2017.45). A research group at Lund University in Sweden has developed a camera that can film at a rate equivalent to five trillion images per second, or events as short as 0.2 trillionths of a second. This is faster than has previously been possible. Courtesy of Lund University.