Pulsed Inductive Thruster

A Pulsed Inductive Thruster (PIT) is an electrodeless electromagnetic spacecraft propulsion system that ionizes and accelerates propellant gas to exhaust velocities of 20 to 80 km s⁻¹ (I_sp ≈ 2,000–8,000 s) using time-varying magnetic induction rather than plasma-wetted electrodes. Pioneered at TRW (now Northrop Grumman) by C. Lee Dailey and Ralph H. Lovberg between the mid-1960s and 1990s, a PIT injects a thin annular puff of neutral propellant gas across the face of a flat spiral induction coil (typically 0.5 to 1.0 m in diameter) and discharges a high-voltage capacitor bank (10–40 kV, 1–5 kJ per pulse) through the coil over a few microseconds. By Faraday's law of electromagnetic induction, the rapidly rising radial magnetic field B_r induces a high-current azimuthal electric breakdown J_θ inside the gas layer, which crosses with B_r to expel the resulting plasma sheet axially via the Lorentz force f_z = J_θB_r. Because plasma never contacts metallic electrodes, the PIT eliminates the cathode and anode erosion that limits the lifetime of magnetoplasmadynamic thrusters and allows direct operation on oxidizing or corrosive in-situ propellants such as carbon dioxide (CO₂), water vapor (H₂O), ammonia (NH₃), and hydrazine.

The operating cycle of a Pulsed Inductive Thruster consists of three synchronized microsecond-scale stages repeated at pulse repetition frequencies f_rep ranging from 10 to 500 Hz (scaling average power P_avg = E_pulse f_rep linearly from kilowatts to megawatts without changing the physics of an individual pulse):

1. Supersonic Annular Gas Injection: A fast-acting piezoelectric or electromagnetic puff valve mounted on a central pylon injects a metered mass increment Δm (typically 0.5 to 10 mg) of neutral propellant downward through a supersonic conical nozzle, spreading a thin, radially uniform gas layer (thickness δ₀ ≈ 1–3 cm) directly against the flat glass or ceramic insulating face of the spiral induction coil.
2. Capacitor Bank Discharge and Inductive Breakdown: Before the gas puff can drift away from the coil face, an array of parallel solid-state thyristors or spark-gap switches closes, discharging a coaxial capacitor bank of capacitance C charged to voltage V₀ (stored energy E₀ = ½*CV*₀²) into the flat spiral coil of parasitic inductance L₀. The high initial current rise rate dI / dt = V₀ / L₀ (typically 10¹⁰ to 10¹¹ A s⁻¹) generates a strong radial magnetic field B_r parallel to the coil face. By Faraday's law (∇ × E = −∂B / ∂t), an azimuthal electric field E_θ ≈ −(r / 2)(∂B_z / ∂t) of several hundred volts per centimeter is induced inside the neutral gas layer, breaking it down into an annular, highly conducting plasma current sheet of azimuthal current density J_θ within ~0.1–0.3 μs.
3. Magnetic Piston Acceleration (Snowplow Sweeping): Once ionized, the plasma sheet acts as a single-turn short-circuited secondary winding of a pulse transformer coupled to the primary spiral coil. Image currents J_θ flowing opposite to the primary coil current exclude the magnetic field from the plasma interior (skin depth δ_skin = √(2η / μ₀ω)), compressing the radial magnetic field B_r between the coil face and the plasma sheet. The resulting axial Lorentz body force f_z = J_θB_r = −∂(B_r² / 2μ₀) / ∂z acts as an impermeable magnetic piston that accelerates the plasma sheet axially away from the coil while sweeping up ("snowplowing") any remaining neutral gas ahead of it.

Equivalent Circuit and Dynamic Inductance Scaling

Electrically, the coupling between the primary induction coil and the accelerating plasma sheet is modeled as a time-varying total circuit inductance L(z(t)), where z(t) is the axial separation of the plasma current sheet from the coil face (Lovberg & Dailey 1993):

L(z) = L₀ + L_c [ 1 − exp(−z / z₀) ]

Here L₀ is the parasitic (stray) inductance of the capacitor bank, transmission feeds, switches, and coil insulation gap (typically 50–80 nH in the PIT MkV/MkVI), L_c is the maximum coupled inductance of the flat spiral coil (~600–750 nH for a 1-meter diameter coil), and z₀ ≈ 0.075 r_coil (~7.5 cm for a 1-m coil) is the characteristic electromagnetic decoupling scale length over which the coil's magnetic field decays axially.

The instantaneous electromagnetic force F_z(t) acting on the plasma sheet is given by the axial gradient of magnetic energy:

F_z(t) = (1 / 2) I²(t) (dL(z) / dz) = [ L_c I²(t) / (2 z₀) ] exp(−z / z₀)

From energy conservation, the maximum ideal electromagnetic conversion efficiency η_EM,max (in the limit of zero plasma resistance) is determined by the ratio of coupled coil inductance L_c to total final inductance L₀ + L_c:

η_EM,max = 1 − L₀ / (L₀ + L_c) = L_c / (L₀ + L_c)

Minimizing the parasitic inductance L₀ relative to L_c is therefore the central electrical engineering requirement of pulsed inductive propulsion.

A breakthrough in PIT performance was achieved by C. Lee Dailey and Ralph Lovberg in 1991–1993 with the PIT MkV, tested at TRW for NASA Glenn Research Center. In earlier PIT designs, connecting all capacitors in parallel across a single full-turn spiral coil required either a 40-kV main switch or suffered high parasitic inductance L₀. In the PIT MkV Marx-loop geometry, the 1-meter coil was wound from multiple interleaved spiral copper straps divided into two half-turn segments in series: nine pairs of 0.5-μF capacitors (each charged to ±15–16 kV) were arranged azimuthally around the perimeter and switched simultaneously so their voltages added in series around the 1-meter circumference to produce a 30–32 kV single-turn azimuthal ring voltage with a parasitic inductance of only L₀ = 65 nH versus L_c = 680 nH.

Constant-Efficiency Variable Specific Impulse on Ammonia

Testing the PIT MkV on ammonia (NH₃) propellant across mass bits Δm from 0.4 mg to 6.5 mg at 4.05 kJ discharge energy revealed a remarkable propulsive characteristic: as the injected gas puff mass Δm was varied at constant discharge energy, the specific impulse varied smoothly from I_sp = 2,000 s to 8,000 s while the measured total thrust efficiency remained nearly flat at η_T = 48% to 54%.

This constant-efficiency variable-I_sp behavior arises from the matching between the dynamic acceleration time t_acc ~ z₀ / u_ex (how long the plasma sheet remains within the electromagnetic coupling zone z < z₀) and the electrical LC ringing half-period T_1/2 = π√(L_totC) ≈ 2.5 μs of the capacitor bank. When the dimensionless dynamic impedance matching parameter (discovered by Polzin and Choueiri in 2006):

α_dyn = (V₀² C² L_c) / (2 L₀ Δm z₀²)

is held near its optimal resonance (α_dyn ≈ 1–3) by adjusting bank charging voltage V₀ alongside pulse mass Δm, the plasma sheet crosses the decoupling distance z₀ just as the primary capacitor ring current reaches its first current zero, capturing the maximum magnetic energy before the oscillating field reverses polarity and attempts to induce a secondary restrike sheet.

Because electrostatic ion thrusters and Hall-effect thrusters rely on delicate thermionic hollow cathodes (such as barium-oxide or lanthanum-hexaboride emitters) that are rapidly poisoned by trace oxygen, water vapor, or carbon compounds, they are largely restricted to expensive, low-density-storage noble gases such as xenon or krypton.

By contrast, the PIT's induction coil is completely isolated behind a fused-quartz or silicon-nitride dielectric plate. Consequently, a PIT can operate indefinitely on:

• Martian Atmospheric CO₂: Enabling round-trip Mars cargo tugs that refuel directly by compressing atmospheric CO₂ at Mars orbit (η_T ≈ 35%–42% measured on the PIT MkV at I_sp = 2,000–4,000 s).
• Lunar, Asteroidal, or Cometary Water (H₂O) and Ammonia (NH₃): Allowing storable, high-density liquid propellants to be shared with chemical attitude-control engines or harvested via in-situ resource utilization from carbonaceous chondrite asteroids and icy moons.

Why does a Pulsed Inductive Thruster require a large 0.5-to-1-meter coil diameter?

The electromagnetic coupling efficiency of a flat spiral coil scales with the ratio of the characteristic axial field-decay length z₀ ≈ 0.075 r_coil to the resistive skin depth δ_skin = √(2η / μ₀ω) of the ionized plasma sheet. If the coil diameter is made too small (e.g., <15 cm), z₀ shrinks to <1 cm—smaller than the plasma skin depth and neutral gas puff thickness—allowing the magnetic field to diffuse straight through the plasma without exerting an efficient magnetic-piston push. To achieve compact inductive propulsion at smaller diameters (15–30 cm), modern designs replace the flat planar coil with a conical theta-pinch or Rotating Magnetic Field Field-Reversed Configuration (RMF-FRC) plasmoid injector.

How can average thrust and power be throttled in a Pulsed Inductive Thruster?

Because each 3-microsecond discharge pulse is completely independent of the next, a PIT throttles its average power (P_avg = E_pulse × f_rep) and average thrust (F_avg = I_bit × f_rep) simply by varying the pulse repetition rate f_rep—for example, firing at 5 Hz (20 kW) near 3 AU where solar power is scarce, and ramping up to 250 Hz (1 MW) at 1 AU or with a nuclear reactor—while maintaining identical per-pulse efficiency η_T ≈ 50%.

What limits the repetition rate and lifetime of a flight Pulsed Inductive Thruster?

Because there is no electrode erosion, thruster lifetime is governed by the high-voltage capacitor bank and solid-state pulsed-power switches. In the Northrop Grumman PIT MkVI program, replacing legacy spark-gap switches with solid-state thyristor/IGBT stacks and circulating liquid coolant through hollow copper induction-coil tubing demonstrated pathways to >10⁹ pulses (thousands of hours of continuous megawatt operation).

1. Dailey, C. L., & Lovberg, R. H. (1993). The PIT MkV Pulsed Inductive Thruster. NASA Contractor Report CR-191155, Lewis Research Center. https://ntrs.nasa.gov/citations/19930023166
2. Lovberg, R. H., & Dailey, C. L. (1982). Large inductive thruster performance measurement. AIAA Journal, 20(7), 971–977. https://doi.org/10.2514/3.51155
3. Polzin, K. A., & Choueiri, E. Y. (2006). Performance optimization criteria for pulsed inductive plasma acceleration. IEEE Transactions on Plasma Science, 34(3), 945–953. https://doi.org/10.1109/TPS.2006.875732
4. Polzin, K. A. (2011). Comprehensive review of planar pulsed inductive plasma thruster research and technology. Journal of Propulsion and Power, 27(3), 513–531. https://doi.org/10.2514/1.B34188
5. Mikellides, I. G., & Villarreal, J. K. (2007). High energy pulsed inductive thruster modeling operating with ammonia propellant. Journal of Applied Physics, 102(10), 103301. https://doi.org/10.1063/1.2809436