Polyimide Film Antennas for a Lunar Radio Observatory K.P. Stewart1, J. Hartman2, E. Polisensky1, B.C. Hicks1, D. Jones3, J. Lazio3, R. MacDowall4

1Naval Research Laboratory 2NASA Postdoctoral Program - JPL

3JPL/California Institute of Technology 4NASA/Goddard Space Flight Center

The LUNAR consortium is funded by the NASA Lunar Science Institute via Cooperative Agreement NNA09DB30A to investigate concepts for astrophysical observatories on the Moon. Part of this research was conducted at Jet Propulsion Laboratory, California Institute of Technology, under contract to NASA

Regolith ε = 3, tan δ = 0.01

8 m

Cu

Kapton

Figure 1. The antenna was tested in a very dry desert location.

The good agreement between measured and calculated impedance demonstrates the accuracy of our simulations. However, our testing with this antenna in close proximity to the lossy ground reduces the antenna sensitivity and the observable amplitude of the diurnal variation in the Galactic background radiation. Even in the 22–25 MHz band (Fig. 6), chosen for lowest RFI levels, the data are noisy and are correlated with the rising and setting of the Sun rather than the Galactic Plane. At higher frequencies, where ionospheric effects decrease, Galactic emission and antenna efficiency are both reduced. Further field tests are planned at other locations and with the antenna raised a few meters above the ground in order to reduce ground losses. Simulations with soil parameters appropriate to the lunar regolith (Fig. 2) give a gain of 0 dBi and indicate that an array of these dipoles would be useful for low-frequency solar radio observations from the surface of the Moon.

Figure 3. Measured and calculated feedpoint impedance.

Figure 2. Antenna model. Regolith parameters are from the Apollo Lunar Sourcebook.

Figure 5. Daytime absorption by the D layer of the ionosphere was observed below ~7 MHz. This plot shows the average 4–5 MHz received power. Times of sunset (S) and sunrise (R) are marked. The three curves are for successive days.

Figure 6. Average received power at 22–25 MHz (blue) compared to a simulation of the Galactic background emission (black) for this antenna at this location. The antenna power shows a diurnal variation, but ionospheric effects mask the expected Galactic background signal.

The Moon is a unique platform for both cosmological and solar observations at low radio frequencies due to the very tenuous lunar ionosphere and diminished RFI, particularly on the far side. Science antennas constructed of metal dipoles deposited on thin dielectric film substrates would have the advantages of convenient packaging and easy deployment. Dielectric sheets, e.g., 25-µm thick polyimide, 1 to 2 m wide and up to several hundred meters long, deposited with an array of dozens of adjacent dipole antennas could be wound on a spool for transport and unrolled robotically on the lunar surface.

The prototype thin-film dipole antenna was constructed of a 5 µm thick Cu layer deposited on a 25 µm thick Kapton film. Each dipole arm was 8 m long and 30.5 cm wide. The inner 1 m of each arm tapered to a point at which a 1:1 wideband balun was attached. Coaxial cable connected the balun to an RFSpace SDR-14 software-defined receiver. The feedpoint impedance was measured with an AIM4170 vector impedance analyzer. Fig. 1 shows the antenna deployed near the LWA1 site in New Mexico.

During this project we tested the antenna on moist grass-covered soil, dry sandy soil, and asphalt. For the results presented here, the antenna was deployed on very dry desert soil in an attempt to more closely approximate lunar conditions. We found good agreement between numerical simulations and measurements of the electromagnetic properties (gain, response pattern, and feedpoint impedance) of a 16 m long thin-film dipole that would be useful for solar radio studies at frequencies around 10 MHz.

CST Microwave Studio 3D electromagnetic simulation software was used to model the antenna on different types of ground, including lunar regolith (Fig. 2). To simulate measured data, the dielectric constant and conductivity of the ground were adjusted to give the best fit to the measured feedpoint impedance. The best-fit values for the desert soil are ε = 8 and σ = 2 mS/m, giving an estimated gain of -18 dBi. The response pattern is nearly isotropic over the sky in this frequency range. The measured and calculated values of the feedpoint impedance are shown in Fig. 3

The receiver recorded a spectrum from 1 – 30 MHz every 5 seconds for a period of 2.5 days. The spectral resolution was 1.017 kHz. Fig. 4 shows the received power as a function of frequency and local time (MDT). The daytime decrease in power below 15 MHz is due to absorption of terrestrial signals by the D layer of the ionosphere (Fig. 5). At the time of these measurements the maximum of the Galactic background emission occurred just before sunrise, making it difficult to separate this signal from the effects of the rapid changes in ionospheric propagation. Variations in ionospheric propagation of MF and HF broadcast stations were clearly observed. These measurements and simulations with appropriate lunar regolith properties predict that antennas similar to this design would work well for radio heliophysics observations and studies of the lunar ionosphere.

Figure 4. Received power vs. frequency (horizontal axis) and local time (vertical axis, showing MDT in 2012). The local elevation of the Galactic center (blue) and the solar flux at 74 MHz (black) are shown on the right, along with times of sunrise (R) and set (S). The solar flux was measured by the Long Wavelength Array. It is not normalized for the decreased sensitivity of this instrument toward the horizon. The solar radio bursts are not seen in these data.