Repurposing an electric toothbrush to measure spectra of hypervelocity impact flashes.

1. Introduction

Impacts are ubiquitous in the Solar System and recently interest has grown into studying the spectrum of the light flashes from these impacts, and the information that the spectra might contain (for example, the composition of the impactor) (see [1] – [6] for examples).

However, such impact events are very short lived (~micro-seconds for mm-sized impactors) and thus very fast spectroscopy is needed to analyse the data from these impacts. High speed, spectrometers are commercially available (i.e. from Teledyne Princeton Instruments) but are expensive and require very accurate triggering. Therefore we set about building our own spectrometer from parts in the lab to see if we could capture impact flash spectra from mm-sized projectiles fired in the University of Kent’s LGG.

2. Experimental setup

The idea was to have a mirror that would vibrate in one plane at high speed. This would reflect light in an arc onto a diffraction grating. A digital SLR would then be focussed on the other side of the diffraction grating and would take an image of the resulting spectrum as the arc was swept from side-to-side. This would give time versus spectral information in a single image.

An old electric toothbrush (a ‘Phillips Sonicare HX6511/43’) was discovered to have a vibrating brush-end that vibrated at ~1000 Hz and through an angle of approximately 30° in a single plane. The brush mount was removed and a polished aluminium mirror with a diameter of 25 mm was epoxied onto the end. This was then inverted and fixed through the top of a black plastic enclosure. A fibre optic feed was mounted on the side of the enclosure to shine light from the target in the impact chamber onto the vibrating mirror. Another hole was drilled in the front of the enclosure to accommodate a diffraction grating and the front of a digital SLR (a Nikon 3200) which was remotely commanded from a Linux laptop (Figure 1).

At the target chamber end a high-speed Nikon f/1.2 lens was mounted on the external surface of the target chamber and focussed onto the impact area of the target in the target chamber. The light from this lens was then focussed onto the top of a microscope objective which then launched the light into an optical fibre. All experiments were carried out in a darkened lab and the optics were made light-tight.

Just prior to a shot the toothbrush was switched on, and the camera commanded to take four x 3-second exposures. When an impact occurred the light from the impact would be seen by the camera during its exposure and recorded as a spectrum.

4. Example results

 4.1 Shot G040321#2.

In this shot the projectile was a 4 mm long by approximately 4 mm diameter cylinder of nylon fired into a mix of glycine and water ice. The measured impact velocity was 6.272 km/sec.

During this shot the projectile broke up into three fragments (a common occurrence) and these fragments were recorded by the time-of-flight oscilloscope. The impact flashes generated by these fragments were also recorded by the toothbrush spectrometer and are shown below. The oscilloscope trace gives temporal information so that the X-axis of the toothbrush spectrometer’s image can be calibrated.

As can be seen the individual flash spectra of the impacts can be clearly seen. ‘ImageJ’ was used to give an intensity scan across the centre of the spectra (right-hand image of Figure 2) and shows that spectral information is evident. Note no attempt has been made to correct the intensity scan from the RGB information given in the Nikon NEF image.

The time difference between the first (left-hand) and second (middle) impacts was approximately 130 microseconds, and between the second (middle) and third (right-hand) was approximately 90 microseconds, thus we have a smear spectrometer with a temporal resolution of a few tens of microseconds

4.2 Shots G200521#1 and G200521#3

 Two shots were performed to attempt to do a spectral calibration of the toothbrush spectrometer. In the first shot (G200521#3) a 1.5 mm diameter copper projectile was fired at a 3 mm thick copper plate at 4.827 km/sec. An Ocean Optics Redtide USB650 spectrometer was used to acquire a single spectrum of the impact flash.

In the second shot (G200521#3) the shot was repeated (v = 6.859 km/sec) but the Ocean Optics spectrometer was replaced with the toothbrush spectrometer. Data from the two experiments were combined to give a spectral calibration as seen in Figure 3.

5. Improvements

 Sadly work on the toothbrush spectrometer ceased shortly after these experiments, but various improvements were planned. These included:

1) Replacing the colour DSLR camera with a Peltier cooled astronomical camera. This should improve sensitivity and remove the need to colour correct the spectrum.

2) Or, alternatively, have the Bayes filter removed from the DSLR turning it into a monochrome camera.

3) Installing a fast flashing LED to act as a temporal and spectral calibration source.

4) Increasing the resolution of the diffraction grating.

6. Conclusions

Using some inventiveness and spare parts a high-speed smear spectrometer was constructed which gave spectra which contained useful scientific data.

The advantage of such a spectrometer was that it would be able to give temporal information at the few microsecond level and thus show how the spectrum of an impact flash evolves which informs us of the evolution of the ions in the plasma and, potentially, how they interact.

References

[1] Avdellidou C., MNRAS, 484, 4, 2019. [2] Yanagisawa M., Icarus, 434, 2025. [3] Tandy J., Meteoritics and Planetary Science, 55, 10, 2020. [4] Simpson G., PNAS Nexus, 2, 7, 2023. [5] Heunoske D., Procedia Engineering 58:624, 2013. [6] Spathis V., Proceedings of the IAC, 64949, 2021.