Fabricating micro scalable thrusters for nanosat propulsion

Nanosatellites (nanosats) are flying high as space platforms. With masses between 1 and 10kg and sizes as small as ‘1U’ (10 x 10 x 10cm, see Figure 1), nanosats are increasingly capable of performing many of the functions carried out by conventional satellites, at a fraction of the cost and with shorter development cycles. Maturation of this technology accelerated significantly over the past decade in many applications, including Earth remote sensing, telecommunications, astronomy, and national security.

Nanosats also make it practical to establish distributed satellite systems composed of hundreds of spacecraft flying together as swarms or constellations. 

To realise their full potential, nanosats need propulsion – be it for orbit-keeping or collision avoidance, or to perform precise manoeuvres such as rendezvous, attitude control, and orbit-changing. Propulsion is therefore key to both extending the life of and enabling new types of missions. In addition, propulsion is needed to assist the deorbit of dated nanosats and reduce the hazards associated with space debris. Of the 2,068 nanosats launched as of 1 August 2022, only 6% have propulsion modules [1]. This is mainly due to the very limited budgets of mass, volume, and power available aboard nanosats making it challenging to add propulsion capabilities. 

Figure 1: A PhoneSat 2.5 Cubesat measuring 1U in size. [Image: NASA’s Ames Research Center]

µSTAMPS electrospray thrusters

Micro Scalable Thrusters for Adaptive Mission Profiles in Space, or µSTAMPS, is the University of Tennessee Research Foundation’s patent-pending electric micro-propulsion technology developed at the university’s Space Institute [2]. µSTAMPS thrusters use electrostatic forces to eject and atomise an electrically conductive liquid propellant into an aerosol of electrically charged fine droplets, or sometimes ions. The preferred propellants are ionic liquids, and it is this process of ejecting the propellant, dubbed electrospraying, that generates thrust. A notional description of a single µSTAMPS electrospray emitter is shown in Figure 2. The emitter capillary wicks the propellant and an electric field, generated by applying a voltage difference between the propellant and the extraction grid, forms the Taylor cone that electrosprays. To attain useful levels of thrust, on the order of at least a few micronewtons, arrays of thousands of these emitters are fabricated on a single thruster chip. The chip is made of a dielectric material (borosilicate glass) and the extraction grid is an electrically conductive coating covering the downstream surface of the chip. 

Figure 2: A single µSTAMPS electrospray emitter

To operate µSTAMPS thrusters at voltages below 1kV, we need 1µm-wide emitter capillaries that are 170µm long. In our lab, we form these vias using Bessel-beam ultrafast laser micromachining. A 1,030nm infrared Gaussian laser beam (6ps, 55µJ), produced by a 20W Amplitude Systèmes Tangerine laser, is transformed into a Bessel beam with an axicon and three pulses (at 100kHz repetition rate) per capillary are focused with a combination of spherical and aspherical lenses [3]. The focused Bessel beam has an ~0.8µm diameter central lobe (0th-order maximum) that extends ~500µm axially, long enough to machine through the entire chip without axial movement.

Properties and applications of Bessel beams

A typical ultrafast (femtosecond/picosecond) Gaussian laser pulse, when tightly focused by a lens with a high numerical aperture such as a microscope objective, forms a compact focal volume ~1µm3 (depending on wavelength). If focused inside a transparent material such as glass, such pulses can easily modify matter in, and around, the focal volume, with low aspect ratios of ~2:1. While particularly versatile for precise ultrafast laser micromachining projects, the limited size of the focal volume makes machining larger-scale volumes, such as long, high aspect ratio channels and waveguides, a challenging and time-consuming process.

Another means of modifying a high aspect ratio volume with just a single pulse, or a few pulses, is a Bessel beam. A Bessel beam can be created by transforming a Gaussian field profile into a radial field profile. This transformation effectively produces a non-diffracting beam that, if tightly focused, results in a long and narrow – essentially columnar – focal volume, which can have an aspect ratio exceeding 500:1 (considering only the most intense, central lobe of a zeroth-order Bessel beam).

Bessel beams have found numerous applications, such as machining microns-wide channels up to 4mm long in glasses [4,5], dicing of glass [6], and optical tweezers [7]. Several methods for creating Bessel beams have been developed. Perhaps the most widely known method is to pass a Gaussian pulse through a conic lens, called an axicon. Other methods include passing the beam through an annular slit. More recently, phase holograms programmed on a spatial light modulator offer numerous advantages in tunability of Bessel beam characteristics, such as enabling higher orders or adding optical angular momentum [8].

Our Bessel beam is created by passing a Gaussian beam through a physical axicon. The basic optical layout is designed and optimised using Zemax [9] ray tracing software. A sample layout is depicted in Figure 3, where the Gaussian laser beam travels from left to right, first passing through the axicon, which reshapes the Gaussian intensity distribution into an approximate Bessel function profile.

Figure 3: Basic layout of optics used to form and focus a Bessel beam

This Bessel beam is subsequently collimated with a spherical lens and then focused into a transparent sample by an aspheric lens with high numerical aperture. An asphere was chosen for the focusing lens because it provides a relatively large working distance (~1.8mm) compared to a microscope objective (~0.1mm) with a similar numerical aperture, which is beneficial for reducing contamination of the lens by ablated material. An assembled Bessel optics set is shown in Figure 4, configured to attach to the objective turret of a microscope used for ultrafast laser micromachining.

Figure 4: A Bessel optics configured for ultrafast laser micromachining

µSTAMPS address new FCC requirements 

In September 2022, the Federal Communications Commission of the United States federal government, adopted a new rule for disposal of satellites to help reduce orbital debris. This rule requires the satellite to deorbit in less than five years following the completion of the satellite’s mission. On its own, a 3U, 7.5kg nanosat at an orbit altitude above 450km will not decay naturally in less than five years. Figure 5 shows the deorbit times for a 3U, 7.5kg satellite at varying altitudes using a propulsion system of four µSTAMPS with a dry mass of 0.11kg and a volume of 0.5U. 

Figure 5: De-orbit times for a 3U, 7.5kg satellite at varying altitudes using a µSTAMPS propulsion system, attained through simulations using a.i. Solutions' FreeFlyer software

Figure 5 shows that µSTAMPS are more than capable of deorbiting a nanosat in less than one year, while having minimal impact on the nanosat’s mass, volume, and payload. 

The authors thank the University of Tennessee Center of Excellence in Laser Applications, the Tennessee Higher Education Committee, the University of Tennessee Research Foundation, NearSpace Launch, Inc. (NSL), and the United States Space Force for their support.

Lino Costa is a Research Associate Professor at the University of Tennessee Space Institute

Brian Canfield is a Researcher and Principal Laser Micromachinist at the University of Tennessee Space Institute

Trevor Moeller is an Associate Professor and the Jack D. Whitfield Professor at the University of Tennessee Space Institute

This article was co-authored by Alexander Terekhov, Joshua Howell, Adam Huller, Tyler Sundstrom and Caitlin Bunce, of the University of Tennessee Space Institute


[1] E. Kulu, ‘Nanosatellite Launch Forecasts - Track Record and Latest Prediction’, Proceedings of the 36th Annual Small Satellite Conference, p. SSC22-S1-04, Logan, Utah, USA, 2022.

[2] J.L.V. da Costa, T.M. Moeller, J.M. Jones, B.K. Canfield, A.Y. Terekhov, and J.H. Howell, ‘Micro Scalable Thrusters for Adaptive Mission Profiles in Space – USTAMPS. Patent Application No. US 2022/0194633 A1, 23 June 2022.

[3] L. Costa, B.K. Canfield, A.Y. Terekhov, J.H. Howell, and T.M. Moeller, ‘Ultra-fast-laser fabrication of electric micro-thrusters for nanosatellites’, Proceedings of the 2022 ICALEO conference, p. Micro 203, Orlando, Florida, USA, 2022.

[4]S. Mitra, M. Chanal, R. Clady, A. Mouskeftaras, and D. Grojo, ‘Millijoule femtosecond micro-Bessel beams for ultra-high aspect ratio machining’, Appl. Opt., vol. 54, pp.7358–7365, 2015.

[5] M.K. Bhuyan, P.K. Velpula, J.P. Colombier, T. Olivier, N. Faure, and R. Stoian, ‘Single-shot high aspect ratio bulk nanostructuring of fused silica using chirp-controlled ultrafast laser Bessel beams’, Appl. Phys. Lett., vol. 104, p. 021107, 2014.

[6] R. Meyer, L. Froehly, R. Giust, J. D. Hoyo, L. Furfaro, C. Billet, and F. Courvoisier, ‘Extremely high-aspect-ratio ultrafast Bessel beam generation and stealth dicing of multimillimeter thick glass’, Appl. Phys. Lett., vol. 114, p. 201105, 2019.

[7] T. Čižmár, V. Kollárová, X. Tsampoula, F. Gunn-Moore, W. Sibbett, Z. Bouchal, and K. Dholakia, ‘Generation of multiple Bessel beams for a biophotonics workstation’, Opt. Express, vol. 16, pp. 14024–14035, 2008.

[8] S.N. Khonina, N.L. Kazanskiy, S.V. Karpeev, and M.A. Butt, ‘Bessel Beam: Significance and Applications—A Progressive Review’, Micromachines, vol. 11, p. 997, 2020.

[9] K. Moore, Zemax OpticStudio, Redmond, WA, 2018.