Ultrafast imaging involves a trade-off between temporal resolution and the length of time that can be observed. Shortening the interval between frames makes it possible to capture changes in electrons and plasma that occur on femtosecond ($10^{-15}$ s) to picosecond ($10^{-12}$ s) timescales. But the total time that can be recorded then becomes short.
Conversely, if you want to observe material deformation or the spread of shock waves that unfold over nanoseconds ($10^{-9}$ s) or microseconds ($10^{-6}$ s), you need to widen the frame interval, which makes it harder to capture changes immediately after a phenomenon begins.
In phenomena triggered by laser irradiation of solids or liquids, microscopic changes immediately after irradiation lead to subsequent material destruction and fluid motion. Seeing the initial rapid changes requires fine time intervals, but following what happens afterward requires a long observation window.
A research team at the University of Tokyo's Graduate School of Engineering, whose work was published in the journal Optica, set out to solve this problem. Instead of keeping the imaging interval constant, they widen it step by step as time passes.
Following a Single Event While Widening the Frame Interval
Conventional high-speed imaging records images at fixed time intervals. For example, if images are taken every 1 picosecond, 10 images cover only about 10 picoseconds.
However, the speed of change after intense laser light strikes a material varies enormously over time.
In the femtosecond-to-picosecond regime immediately after irradiation, extremely fast electron-related phenomena occur, such as multiphoton absorption, ionization, and plasma generation. Later, in the range from several hundred picoseconds to nanoseconds, mechanical phenomena become prominent, including the propagation of stress waves and shock waves, the expansion of plasma plumes, and the removal and ejection of material.
If the frame interval is made extremely short to capture the initial changes, it becomes difficult to keep recording all the way into the nanosecond regime.
The standard technique long used to address this problem is the pump-probe method. Laser pulses with the same conditions are repeatedly fired at a sample, and images are acquired while the timing between the pump light that triggers the phenomenon and the probe light used for observation is shifted bit by bit. By stitching together the images obtained at each time point, the time evolution of very fast phenomena can be reconstructed.
However, this approach presupposes that the phenomenon can be reproduced repeatedly under nearly identical conditions.
When there is variation in a material's microstructure or surface condition, or when the phenomenon is irreversible, such as fluctuations in a liquid or fracture, the result may change with each shot. Even if images from multiple experiments are combined, it is not always possible to directly trace how the initial change led to the subsequent destruction or fluid motion within a single event.
The team therefore aimed to extend the time range of "single-shot measurement," which records the time evolution of a phenomenon from one laser shot.
The approach is to image at short intervals during the fast initial stage, then widen the interval as the phenomenon slows. Rather than a constant interval, the time spacing expands like a logarithmic scale.
BRIDGE: Giving Each Wavelength Its Own Delay
To realize this scheme, the team developed an optical technique called BRIDGE (Band-Resolved Individual Delay Generation).
Imaging femtosecond-scale phenomena uses ultrashort-pulse lasers that emit light for an extremely short time. Such laser pulses contain multiple wavelength components spread over a certain bandwidth.
BRIDGE exploits these wavelength differences.
First, an optical filter whose transmitted wavelength changes with angle is used to extract multiple wavelength bands from the continuous spectrum. Each band is then assigned an optical path of a different length.
The longer light travels, the later it arrives. By varying the distance each wavelength travels between mirrors, each band can be made to reach the sample at a different time.
Some conventional wavelength-dispersion-based imaging methods create time differences between wavelengths using the dispersion of diffraction gratings or optical fibers. In this case, the relationship between wavelength and delay depends heavily on the properties of the medium, and the pulse itself stretches in time, degrading temporal resolution.
BRIDGE gives each wavelength band its own independent free-space optical path, making it possible to set delays individually while suppressing the pulse broadening caused by dispersion.
データを表で見る
| 時間 (秒) | |
|---|---|
| 各コマの露光時間 | 0.00000000000026 |
| 最小フレーム間隔 | 0.0000000000006 |
| 設定可能な最大遅延時間 | 0.00000001 |
The system records 11 images, each with a short exposure time of about 260 femtoseconds. The shortest frame interval is 600 femtoseconds, and delays can be set up to a maximum of 10 nanoseconds.
In other words, a single shot can track phenomena from the extremely fast changes that begin in the sub-picosecond regime, through the picosecond range, and into the nanosecond regime.
However, the 11 images obtained with this method are not continuous footage taken at a constant frame rate like ordinary video.
They are 11 still images obtained by illuminating the sample at preset, different times using 11 different wavelength bands. A distinguishing feature of BRIDGE is that the imaging interval itself differs from frame to frame.
Tracking Laser Irradiation of Solids in a Single Shot
Using BRIDGE, the team imaged laser ablation caused by irradiating solids with a femtosecond laser. The samples were glass and hydroxyapatite. Hydroxyapatite is one of the main inorganic components of bones and teeth.
When an intense femtosecond laser is focused on a solid surface, ionization and plasma generation occur immediately after irradiation. As time progresses from picoseconds to nanoseconds, stress waves and shock waves propagate, the plasma plume expands, and eventually material and debris are ejected.
The team tracked these phenomena across different timescales using 11 images obtained from a single laser shot.
The images recorded, in chronological order, filaments, plasma in air, the ablation plume, shock waves, stress waves traveling inside the glass, and the ejection of material and scattering of debris.
In the nanosecond regime, differences were also seen between glass and hydroxyapatite in how material spread and how debris was ejected.
What matters is that this sequence of changes was recorded as images from a single shot, rather than reconstructed by irradiating different samples at each time point.
This allows the change that began in the femtosecond regime and the subsequent mechanical response in the nanosecond regime to be compared within the same event.
However, the images presented do not by themselves reveal how differences in the initial electronic state determined the later form of destruction.
Single-shot imaging is strong at capturing phenomena that differ from one occurrence to the next, but evaluating reproducibility and differences between materials also requires repeated experiments under varied conditions and quantitative analysis.
Shock Waves in Water Also Imaged
Beyond irradiation of solids, the team also applied BRIDGE to optical breakdown and the generation of shock waves when a laser is focused in water.
When intense laser light is focused in a liquid, local ionization produces plasma, and the abrupt pressure change generates shock waves and other effects.
In this imaging, the image sequence obtained in a single shot recorded local breakdown occurring in the femtosecond regime, its expansion in the following picosecond regime, and its development into an asymmetric shock wave in the nanosecond regime.
What this experiment shows is that BRIDGE can observe a local change that begins in a very short time and the larger fluid motion that follows as one continuous event.
At the same time, the result cannot be interpreted more broadly than warranted.
The published study does not quantitatively establish every hydrodynamic quantity, such as the propagation speed, pressure, or released energy of the shock wave. The core achievement is capturing changes unfolding on different timescales as a single-shot image sequence.
Although the use of hydroxyapatite and water may suggest medical applications, this study is a fundamental experiment in optics and laser physics. It is not research that tested medical effects on the human body, laboratory animals, or biological tissue.
Bridging a Wide Range of Timescales in One Shot
The study was published in Optica on September 25, 2026. The paper appears in Volume 13, Issue 10, pages 2028–2037 (DOI: 10.1364/OPTICA.610944).
The authors are Keitaro Shimada, Kohei Azuma, Jun Horiuchi, Dylan Thomas, Masato Ota, and Associate Professor Keiichi Nakagawa. First author Shimada was a project researcher at the University of Tokyo's Graduate School of Engineering at the time of the research and is now a postdoctoral researcher at Lawrence Berkeley National Laboratory in the United States.
The team has long been developing optical techniques for imaging different timescales at once.
In 2023, they reported an imaging method that captures fast phenomena spanning multiple timescales in a single shot using a mechanism called a "spectral circuit." With BRIDGE, assigning an independent optical path to each wavelength band allows the imaging time of each frame to be set more freely.
The team envisions future applications including observing and optimizing laser processing, measuring irreversible structural changes in materials, and observing fast phenomena that are difficult to reproduce repeatedly, such as high-energy-density phenomena.
However, challenges remain on the road to practical use.
BRIDGE splits the probe light into multiple wavelength bands and assigns each to one image. As a result, the amount of light available per frame decreases. Obtaining sufficient image quality and spatial resolution requires highly sensitive detectors and precise optical alignment.
Because each frame uses a different wavelength, care is also needed when the absorptivity or reflectivity of the object varies greatly with wavelength.
It is necessary to distinguish whether changes in image brightness arise from changes over time or from the sample looking different at different wavelengths. Depending on the material, correction or additional calibration may be needed.
How broadly this method can be applied to materials and phenomena beyond the glass, hydroxyapatite, and water used in this study remains a subject for future verification.
The significance of BRIDGE is not simply that it is a "faster camera." It changes the imaging interval itself to match how the phenomenon progresses, connecting the extremely fast initial changes with slower subsequent phenomena in a single shot.
By placing 11 images at different times, from an interval as short as 600 femtoseconds out to 10 nanoseconds, it shows the possibility of following time regimes that were previously easier to observe separately as one event.
If reproducibility and quantitative accuracy are confirmed across a wider variety of materials and phenomena, it could become a new measurement tool for investigating one-off, high-speed phenomena in fields such as laser processing and materials science.