A high-speed anamorphic pupil-conjugate slit spectrograph for rapid, self-luminous sources

When scientists study things that burn, explode, or flash into existence for a split second, they face a frustrating paradox. To understand the chemistry of a rapidly changing event, like a flame front racing across a surface or a wire exploding in a vacuum, they need to break the light it emits into a rainbow of colors. This process, called spectroscopy, reveals exactly which chemical elements are present. However, traditional tools for this job are built on a simple assumption: that the light source stays still. If the source moves even slightly while the instrument is looking at it, the resulting rainbow smears and shifts, making the data unreadable or misleading. For events that happen too fast to stop and recalibrate, or for explosions that can only happen once, this movement creates a blur of confusion where real chemical signals get lost in the noise.

A team of researchers at Vision Research in New Jersey has built a new kind of camera system that solves this problem by changing where the light enters the machine. Instead of locking the instrument's focus to the position of the object, they designed the device so that the position of the light source does not matter at all. By placing a specific internal stop and a special cylindrical lens in a precise arrangement, they ensured that the angle of the incoming light, rather than its location, determines where the colors land on the sensor. This means that even if a burning piece of material jumps, spins, or expands wildly, the chemical fingerprint it produces remains perfectly steady on the camera's screen. The result is a tool that can capture the true chemistry of chaotic, one-time events without the data drifting away.

The core of this innovation is a shift in how the instrument handles the relationship between space and color. In a standard spectrograph, the slit—the narrow opening that lets light in—is an image of the source. If the source moves, the image of the slit moves, and the colors shift across the detector. The researchers realized that if they could instead make the slit an image of the "pupil," or the aperture that controls the light's angle, the position of the source would become irrelevant. They constructed a system using standard, off-the-shelf camera lenses and a high-speed camera to create a path where the light passes through a 4f relay, a term describing a specific arrangement of lenses that keeps the light's angle constant. Inside this relay, they placed an iris that acts as the system's main stop. Then, they added a cylindrical lens, which focuses light in only one direction, to place the slit exactly where the angle of the light is encoded, not its position.

To prove this concept works, the team first tested the instrument with a simple light source, a flashlight, moving back and forth across their field of view. When they used a traditional setup where the front lens acted as the stop, the spectral image wandered significantly as the light moved, shifting by more than a millimeter on the sensor. This would have made it impossible to track fast-moving chemical changes. However, when they switched to their new design, with the internal iris acting as the stop, the spectral image stayed locked in place. Even as the light source swept across the entire usable view, the colors shifted by less than a thousandth of a millimeter. This stability held true regardless of which camera lens they used to look at the source, proving the instrument is truly agnostic to the choice of fore-optic.

With the stability confirmed, the researchers turned to characterizing the precision of their new tool. They used a helium discharge lamp to map out exactly which column on the camera sensor corresponded to which wavelength of light. By fitting a smooth curve to the known lines of helium, they achieved a calibration accuracy with a residual error of just 0.12 nanometers. They verified this by testing a neon lamp without changing the calibration, finding the neon lines matched their predictions with an error of 0.13 nanometers. They also measured the sharpness of the instrument's vision, known as spectral resolution. By carefully adjusting the slit width, they found the optimal setting where the instrument could distinguish two colors that were 0.92 nanometers apart, a level of detail sufficient to separate many important chemical signatures.

The true test came when they applied the spectrograph to a real, chaotic event: the combustion of nitrocellulose. They placed a strip of the material three meters away and ignited it. As the flame front raced across the strip, expanding and moving unpredictably, the camera captured the light at a rate of 10,000 frames per second, with each frame exposing the sensor for just 99 microseconds. Because the instrument was immune to the motion of the flame, the spectral lines remained steady. The team was able to identify five distinct chemical species in the fire: sodium, potassium, calcium, rubidium, and a molecule called calcium hydroxide. They tracked the sodium line throughout the entire event, and despite the violent motion of the flame, the position of the line on the sensor shifted by only 0.06 nanometers on average. The tiny variations that did occur were not due to the instrument failing to track the motion, but rather to the physics of the flame itself, where cooler outer layers of gas absorbed and re-emitted light in a way that slightly shifted the apparent center of the line.

This work demonstrates that it is possible to build a high-speed spectrograph that does not require the source to be stationary. By decoupling the spectral registration from the pointing of the camera, the researchers have created a tool that can observe fast, self-luminous sources without the need for on-the-fly recalibration or complex fiber-optic scrambling. The device, built from standard catalog optics and a high-speed camera, maintains a fixed spectral position even as the source moves, eliminating the non-physical shifts that usually plague such measurements. While the design sacrifices some ability to see fine spatial details in the direction of the color spread, it gains the ability to see the true chemistry of moving events. The researchers suggest this approach is ideally suited for future studies of exploding wires and pulsed plasmas, where the source expands rapidly and the need for frame rates in the hundreds of thousands or millions of frames per second is critical. For now, the instrument stands as a proof that with the right optical arrangement, the chaos of a moving flame can be tamed into a clear, steady record of its chemical composition.