New Salt-based Propellant Rockets

Researchers from the University of Illinois at Urbana-Champaign used a salt-based propellant that had already been proven successful in combustion engines, and demonstrated its compatibility with electrospray thrusters. This will make dual-mode rocket engines successful. It will function in both combustion and electric propulsion systems.

With electrospray or colloid propulsion, the thrusters electrostatically accelerate ions and droplets from these liquids. It’s a technique that started in the biology/chemistry community, then the propulsion community began looking at it about 20 years ago.

Liquid is fed through a very small diameter needle, or capillary tube. At the tip of the tube, a strong electric field is applied that interacts with the liquid in the tube because the liquid itself is a conductor. The liquid responds to that electric field. Small droplets and ions get pulled out of the liquid—spraying them out of the tube or needle.

In this study, in addition to showing that the propellant could be sprayed, Rovey said they were interested in learning what kinds of chemical species come out in the plume. “Because no one has ever tried this type of propellant before, we expected to see species that no one else has ever seen before and, in fact, we did.”

Propulsion Power Journal – Hydroxylammonium Nitrate Species in a Monopropellant Electrospray Plume

Rovey said they also saw a new swapping of the constituents that make up the two different salts.

“We saw some of the hydroxylammonium nitrate salt bonding with the emim ethyl sulfate salt. The two are mixed together inside the propellant, and are constantly bonding with each other and then detaching.

“There’s a chaotic nature to the system and it was unclear how those interactions within the liquid itself would propagate and show up in the spray. There are no chemical reactions happening. It’s just that we start with A and B separately and when they come out in the spray, A and B are bonded together,” he said.

Rovey said these findings shed a lot of light on what’s happening in these mixtures of salts that are possible propellants for electrosprays. But it also opens doors to a lot of other questions that will lead to fundamental studies that try to understand the interactions within these propellants and how that translates into what comes out in the spray itself.

Additional collaborative researchers were from Missouri University of Science and Technology, Boston College, and the Air Force Research Laboratory in New Mexico. Co-author Mitchell Wainwright used the apparatus at the Kirtland Air Force Base in Albuquerque, New Mexico for the study to take measurements. The team supplied the unique propellant.


A mixture of 1-ethyl-3-methylimidazolium ethyl sulfate ([Emim][EtSO4]) and hydroxylammonium nitrate (HAN) is an energetic monopropellant potentially suitable in a multimode chemical-electric microtube-electrospray micropropulsion system. In this work, electrospray plume mass spectra are compared between the monopropellant mixture and neat [Emim][EtSO4]. This comparison clearly indicates new and additional species present in the plume due to the addition of HAN. Mass spectra from 20 to 600amu/q were obtained over a variety of angles and flow rates from 2pL/s to 3nL/s in both cation and anion extraction modes. Mass spectra dependence on flow rate and angular orientation agree qualitatively well with literature. Results indicate the presence of HAN-based species in anion mode, but no HAN-based species in cation mode. Three of the four monomer species from the monopropellant are apparent in the plume; [Emim]+, [EtSO4]−, and [NO3]− are observed, whereas [HA]+ is noticeably absent. Results also show emission of both proton-transferred covalent forms of HAN: [HA-H] and [HNO3]. Swapping of anion and cation species between the constituents of the monopropellant mixture is also observed.

14 thoughts on “New Salt-based Propellant Rockets”

  1. As someone who is doing research in spacecraft electric propulsion, I can tell you electrospray and colloid thrusters are indeed useful now – of course as you say it’s a special case, but then a lot of space applications are! A specific example is ESA’s LISA gravity wave mission – this requires very tiny thrust levels to maintain the precise positioning needed between the three spacecraft that form the gravity wave detector. Electrospray is perfect for this, exactly because of the tiny flow rates from the thruster.

    On the commercial side, a company called Enpulsion is already flying a slightly different technology, Field Emission Electric Propulsion, that also produces very low thrust and high specific impulse. There is a market for dozens of these per year if cubesat/small sat markets carry on growing.

    On a system level, propellants like HAN, and indium metal used by Enpulsion, have several benefits over xenon which is currently the most common propellant for electric propulsion – they are lower cost, storable as liquids at high density and low pressure, and much more abundant.

  2. Well, yes, it’s inherent that if you’re using limited power to accelerate propellant to speeds like 2 million m/s, you’re going to get really low thrust and really high ISP. But take a very long time to exhaust all that fuel.

    What we *really* need is a thruster that doesn’t require the power to go through power conditioning equipment, and where almost all of the waste heat ends up in the exhaust, (Which consists of the fuel generating the power.) contributing to thrust, so that the efficiency is near 100%.

    I think they call that a “rocket engine”.

    Rocket engines are great, the only issue we’ve got is that they’re not running hot enough. We should work on that, instead.

    Though “colloid thrusters” are great for satellite station keeping.

  3. And after as G-G, doing those calculations, I asked … how long, to what speed?

    It becomes SO depressing, when that question is answered.

    Necessarily — higher charge-per-droplet ionization, higher acceleration field intensity (voltage), … result in some astounding numbers.  

    A ‘mere’ 2,500 volt field, per G-G calc, results a 5,240 YEAR acceleration time; the speed is a whopping 2.4% the speed of light.  Wait…2.4% of c … in over five thousand years of continuous acceleration. Using 100 MW of power each and every second of that trip.  

    Drop to 500 V acceleration, the droplet-rate kicks up a bunch as ISP also drops. 1,100 years, and 1.1% the speed of light.  Well, that’s not terribly encouraging!

    We really need those Mach-Effect reactionless thrusters.  Seriously…

  4. Devil is in the details

    2.5 pL = 2.5×10⁻¹⁵ m³
    … = 2.5×10⁻¹² kg at water density

    (2.5×10⁻¹⁵ m³ × 3 / (4π))(¹/3)×2 = 16.8 µm

    10,000 C/L = 10,000,000 C/m³
    … × 2.5×10⁻¹⁵ = 2.5×10⁻⁸ C 
    … × 6.022×10²³ e/C = 1.51×10¹⁶ electrons

    momentum and energy:
    … × 2,000 V acceleration = 3.01×10¹⁹ eV kinetic
    … × 1.6×10⁻¹⁹ J/eV = 4.82 J/drop
    … √( 4.82 J × 2 ÷ 2.5×10⁻¹² kg ) = 1,960,000 m/s
    … × 2.5×10⁻¹² kg = 0.00000491 N-s/drop

    hypothetical Big Box spacecraft:
    100 MW power, 33% coupling efficiency
    … 33,000,000 W of kinetic energy
    … ÷ 4.82 J/drop = 6,850,000 drop/s
    … × 00000491 N-s/drop = 33.6 N

    So, this shows yet again, that ISP would be phenomenal (200,000!) but thrust would be next-to-nothing for 100 megawatts found in a magic hat that the space aliens bequeathed to Mankind … in a few decades, say. 

    Just saying,
    GoatGuy ✓

  5. Doing a little looking around, we’re talking droplets of about 25nm diameter, (I think that’s a little less than 2.5 pL.) and a charge to mass ratio of about 10,000C/kg.

  6. Might be a bit of a nail-biter … but I don’t think that the droplets get their charge “for free”. TANSTAAFL in physics. The wickedly low vapor pressure is definitely good. Essentially electroconductive motor oil.  

    In any case, the droplets need charging. I would imagine an A/C charging-and-accelerating configuration, as high-Q RF tank circuits are quite effective at increasing ΔVolts with fairly limited up-circuit complications. It also solves the “we don’t want a charged plume…” problem.  

    Kilovolt RF fields. Maybe a billion electrons (or holes) per droplet, 2.5 pL ea.; At 25 kV field, such droplets have an ISP of ‘only’ 100.  But it gets better with higher ionization, or … with perhaps surprising irony, … smaller droplets.  

    Just saying,
    GoatGuy ✓

  7. Few people are comfortable working with algebraic physics equations. Yet, they’re particularly telling.  

    For instance, one can ask the question, “how much energy is needed to get a particular thrust … theoretically (i.e. 100% efficiently) for a given ISP?” 

    On the one hand, you’ve got the kinetic energy equation

      Ek = ½mv²

    And on the other, you’ve got the action-reaction momentum equation:

      u = mv

    So, that becomes

      Ek/u = ½ v

    Tho’ the units … m/s … is less transparent than remembering ‘joules per newton-second’ of the Ek/u equation. 

    But the point you make is well made: the flying alphas from so-called aneutronic p+¹¹B fusion have what, 2.9 MeV kinetic energy. With 4 AMU a copy, 2.9 MeV becomes about 12,000,000 m/s. Great ISP! one megasecond! But… not much thrust at achievable pB fusion rates. So, I see why your group was hoping to catalyze (drag?) indium to whiz along at a lower velocity, carrying far greater mass, and more inertia.  

    Got it.
    GoatGuy ✓

  8. “Raising ISP would be nice, but those lil droplets actually are pretty heavy, compared to ions. billions of atoms. ”

    While that’s true, that doesn’t go to the power to achieve a particular ISP, but only the V vs I tradeoff. Droplet propulsion is much higher voltage, but much lower current. That’s a useful tradeoff, since in practice insulation is lighter than conductors, voltage is easier to handle than current if you’re weight limited. IMHO, the charge to mass ratio for something like Xenon is actually much, much too high!

    The nice thing is that since the ionic salts used here already have charge carriers in them, you get ionization basically for free, whereas a conventional ion engine uses a lot of its power just ionizing the fuel.

    They also remain liquid over a wide temperature range, while having an extremely low vapor pressure, which is a neat combination of properties for something to be used in space.

    All in all it’s a promising technology, I think, which does have the capability to be scaled up to the point where it can be used for something more than station keeping thrusters.

  9. A little less than 10 years ago, I published a couple of papers at UIUC about using variable mass to charge droplets like this (although I used Indium) to mass dope the exhaust plume of a helicon-injected IEC fusion rocket. The bare fusion products were of pB11 – so, alpha particles. The idea was to try to get kN to 10s of kN thrust + low 10^5s ISP instead of 10^6-7 s ISP + 10^-1 low 10^1 newtons of thrust (as you conceivably could with just alphas).

  10. I’m willing to accept the idea of a solar-powered, accelerated ion spacecraft as possibly being manufacturable, sufficiently durable, reasonably cost-effective, endowed with competent science-and-optics payload, and so forth.  

    Yet as the saying goes, “the Devil is in the details”.

    For instance.  With some realistic generating parameters.

    … area = 25,000 m²
    … density = 2,500 kg/m³
    … thickness = 2.5 µm
    … support OH = 1×
    … efficiency = 20%

    … mass = 20 kg

    … perihelion = 0.3 AU
    … apohelion = 10 AU

    RE Mass
    … ISP = 2,500 (25 km/s)
    … mass = 25×
    … thrusters = 1×

    So… that gets to about 22 km/s. Moving up to 100× fuel (over rest-of-rocket mass) to 26 km/s. Not a big help. Raising ISP would be nice, but those lil droplets actually are pretty heavy, compared to ions. billions of atoms.  

    Interesting note: a droplet of ANY SIZE … having ‘one extra electron’ of charge, passing thru a field of 1 volt, attains a ΔV of ‘1 electron-volt’ of energy. So, the acceleration of the drops depends heavily on their charge, and their size, and the electric field gradient.

    Just saying,
    GoatGuy ✓

  11. Yes, actually: It’s been proposed to create thin film devices that integrate solar power, propellant handling, and the electrospray engine, (The design of which is very well suited to integrated circuit manufacturing techniques.) into something called a “brane craft”; In practice you’d use it something like a solar sail, except that the thrust per square meter would be a lot higher as long as the fuel held out.

  12. I’m not quite sure that the consequence of this article’s authors’ research has much real-world applicability, except for a very special case. Namely, that using high tension electric field atomization of a conductive liquid into charged droplets, and they to a more-or-less unidirectional plume of accelerated droplets is good work, but for what end?

    Even if we assume massive scaling … from the experiment’s picoliter to nanoliter per second rates (inverted becomes 11 years per liter (at 3 nL/s) to 15,000 years per liter … at 2 pL/s well, the propulsion DIVIDED BY whatever scaling we wish to entertain … is still only suitable for micro- or so-called nano-satellite station keeping. Because the kilovolt power supply still isn’t something easily compacted.  

    Anyway, it does remain interesting.
    Are there any proposals for actual spacecraft use?

    Just asking,
    GoatGuy ✓

Comments are closed.