Vintage Micro: The Original Gamma Ray Observatory

Whenever I think “gamma ray observatory”, impressive orbiting platforms come to mind like NASA’s massive 17 metric ton Compton Gamma Ray Observatory launched in 1991 or maybe some of its more modest brethren launched over the decades before or since like NASA’s 4.3 metric ton Fermi Gamma-ray Space Telescope (FGST) currently gathering data in orbit. Surprisingly, the very first gamma ray observatory was actually a tiny satellite called Explorer 11 launched over half a century ago near the dawn of the Space Age. And with a payload mass of only 37 kilograms, it qualifies as a microsatellite as it is generally defined today (i.e. a satellite with a mass between 10 and 100 kilograms). What is all the more impressive was what was gleaned from the analysis of its meager data set which amounted to just 22 gamma rays detected from beyond the Earth during its seven months of operation.

The Satellite & Payload

Gamma rays are the most energetic form of electromagnetic radiation and are generated only by the most powerful processes involving atomic nuclei as well as highly energetic particle interactions. Because of this, gamma ray astronomy deals with the most violent imaginable events in the Universe.  Since our atmosphere shields the Earth’s surface from gamma rays, one must get above Earth’s atmosphere to detect them. Even detectors carried to high altitudes by balloons, as was first done in 1957, would have a difficult time unambiguously detecting celestial gamma rays because of the larger number of gamma rays produced by the interactions of the thin atmosphere that is still above the balloon with the constant rain of cosmic rays. As a result, it was recognized early on that a satellite was the best means of unambiguously detecting celestial gamma rays especially in the lower energy range.

Explorer 11, designated as payload S-15 by NASA before its launch, was one of the “second generation” Explorer satellites. Unlike the “first generation” Explorers 1 through 5 that were orbited by the Juno I launch vehicle (which was limited to a payload of about 11 kilograms), this new series of Explorer satellites employed the larger Juno II originally developed by Wernher von Braun and his team at the Army Ballistic Missile Agency in Huntsville, Alabama (which became NASA’s Marshall Space Flight Center). The cluster of spinning solid rocket motors that formed the upper stages of the Juno I were retained on the Juno II but the modified Redstone first stage was replaced with a stretched version of the much larger Jupiter IRBM increasing the orbital payload to about 45 kilograms in the process. While the Juno II was kludged together from a variety of components and was far from an ideal satellite launch vehicle, it provided NASA with a means of launching science payloads into Earth orbit while more capable, purpose-built launch vehicles like Scout and Delta were being developed (see “Vintage Micro: The Second-Generation Explorer Satellites“).

Diagram of the gamma ray detector carried by the Explorer 11 satellite. (NASA)

Development of the gamma ray detection payload at the heart of Explorer 11 was started in 1958 by a team led by William L. Kraushaar and George W. Clark of the Massachusetts Institute of Technology (MIT) in Cambridge, Massachusetts. The compact instrument they developed was designed to detect gamma rays with an energy in excess of about 50 MeV within a 17° angle of the long axis of the satellite. Because the flux of cosmic rays and other forms of energetic radiation would be over three orders of magnitude higher than the gamma rays they sought, Kraushaar and Clark designed a clever, three-part detection scheme to filter out the unwanted signals.

The first part of the instrument was a scintillation detector consisting of five alternating layers of cesium iodide and sodium iodide crystals. A gamma ray entering this detector has a high probability of interacting with the atoms in these layered crystals producing an electron-positron pair (which would continue to travel forward in the same general direction as the incident gamma ray) as well as a flash of light that would subsequently be detected using photomultiplier tubes. If the incident gamma ray was traveling within ±17° of the central axis of the detector, one or both of the particles in this electron-positron pair would then enter a Cerenkov counter consisting of a block of transparent plastic producing another flash of light that would be detected by another set of photomultiplier tubes. A pair of simultaneous flashes in both parts of this coincidence detector would indicate the possible detection of a gamma ray within the 34° wide field of view of the instrument.

In order to eliminate the possibility that this signal was actually the result of a cosmic ray or other energetic charged particle, the instrument was surrounded on the front and sides by more clear plastic that acted as an anticoincidence detector. An unwanted energetic charged particle would produce a flash of light as it passes through this block of clear plastic (which would be detected by another set of photomultiplier tubes) while an electrically neutral gamma ray photon would not. A flash detected by this anticoincidence filter would veto any signal simultaneously detected by the central part of the detector filtering out the unwanted signal. In the end, only a signal produced simultaneously in the scintillation and Cerenkov detectors but not accompanied by a signal in the anticoincidence detector would be recorded as a gamma ray detection.

Depiction of Explorer 11 in Earth orbit. (NASA)

In addition to the gamma ray detector with its power supply and its state-of-the-art transistor-based detection and logic circuits, Explorer 11 also included a telemetry and command system as well as a tape recorder to save data when the satellite was not within range of a ground tracking station. The exterior housing of the payload was covered with solar cells to charge the payload’s batteries during its operating life in orbit. Data received by the ground stations were recorded on magnetic tape that were subsequently forwarded to NASA’s Goddard Space Flight Center in Greenbelt, Maryland where the data were transferred to photographic film via a multichannel oscilloscope . This film data were then forwarded to MIT where they were reduced and analyzed with the aid of a then state-of-the art, fully transistorized IBM 7090 computer at the MIT Computation Center (for those who want a good chuckle, the IBM 7090 had enough memory for 32k words at 32-bits per word and a computational speed of 100 kiloflops per second – insignificant by today’s standards). The data, when fully analyzed, allowed the time of a gamma ray detection to be ascertained to a typical accuracy of 0.1 seconds.

Unlike the other second generation Explorer satellites, the spent fourth stage of the Juno II was intentionally left attached to the payload once Explorer 11 was in orbit (just as was the case with Explorer 1 and the other first generation Explorer satellites) giving the satellite a total in-orbit mass of 43.2 kilograms. Since the satellite was too small and simple to accommodate any sort of attitude control system to point its gamma ray detector, an ingenious means was found to have the satellite scan the entire sky for gamma ray sources.

During ascent, the cluster of solid rocket motors that served as the upper three stages of the Juno II were spun at 340 RPM to provide stability. Without any perturbing forces, the fourth stage-satellite configuration would keep spinning indefinitely with a fixed orientation with respect to the celestial sphere. But since this long, top heavy configuration is unstable, subtle perturbing forces from magnetic torques would cause the satellite’s axis of rotation to wobble increasingly over the course of a couple weeks. Eventually the satellite would end up spinning in a more stable configuration about its short axis like a propeller at a rate of 5 RPM scanning a 34° wide swath across the sky with each revolution. A precession rate of about 10° per day would allow the entire celestial sphere to be scanned over the course of a couple of weeks. Light sensors, which would detect the position of bright objects like the Earth and Sun, and a careful measurement of how the strength of the satellite’s transmitted signal varied over time allowed the scientists to determine where the gamma ray detector was pointed at any given time to an accuracy of a few degrees.

The Mission & Scientific Results

Explorer 11 was successfully launched into a 497-by-1,793 kilometer orbit with an inclination of 28.8° by Juno II Round AM-19E on April 27, 1961. As it turned out, this was the last successful launch of the Juno II before it was retired after the unsuccessful launch attempt of Explorer S-45a less than a month later.  With the apogee slightly higher than planned, the satellite would briefly enter the lower reaches of the Van Allen radiation belt during each orbit saturating the gamma ray detection electronics during this part of the orbit and degrading the solar cells over time reducing the satellite’s expected lifetime as a result. Aside from this and the failure of the tape recorder (which limited data collections to real-time detections when Explorer 11 was being tracked from a ground station), the world’s first orbiting gamma ray observatory was ready to make observations.

Launch of Explorer 11 on Juno II Round AM-19E on April 27, 1961. (NASA)

The initial spinning mode about the long axis of Explorer 11 continued for the first two weeks in orbit. Starting May 16, the cone of motion opened up quickly and by May 19 the satellite had, for all practical purposes, started spinning end over end.  The initial period rotation was 12 seconds which increased to about 15 seconds over the next seven months. Explorer 11 operated normally for the first two months but a worsening problem with the power supply over the following five months resulted in varying proportions of unusable data. The transmitter was finally turned off by ground command after 224 days in orbit when the power supply voltage had permanently fallen below usable levels.

Despite the loss of the tape recorder and the issues with the power supply, Explorer 11 secured 141 hours of data acquired from 6,000 useful tracking station passes with 250,000 tumbles during its seven months of operation. Out of a the 1,021 “events” that were accepted by the gamma ray detector’s logic circuits, a total of 22 were determined to be gamma rays from celestial sources (as opposed to gamma rays that definitely were or could have been from cosmic ray-induced events in the Earth’s atmosphere or otherwise questionable data). While it was difficult to make any definitive statements from such a statistically small sample, some broad conclusions were able to be drawn from an analysis of just these 22 gamma ray detections.

First, it was obvious that there are no unexpectedly strong gamma ray sources in the sky. As a result, only rough upper limits could be set for the gamma ray flux from various sources (e.g. the Sun, the center of our galaxy, nearby galaxies or various known active galactic nuclei). The average detection rate of 2.4 gamma rays per hour was about an order of magnitude or so higher than had been predicted to result from the interaction of high energy cosmic rays with interstellar hydrogen hinting that other processes (e.g. high energy electrons in intergalactic space) were responsible for the observed gamma ray intensity. With so few recorded detections, there was no clear statistical correlation evident with galactic latitude.  Today, with far better and larger gamma ray detectors, we know that gamma rays preferentially come from the galactic center and from sources clustered along the galactic equator in addition to transient gamma ray bursts from distant sources far beyond our galaxy.

This all-sky image, constructed from two years of observations by NASA’s Fermi Gamma-ray Space Telescope, shows how the sky appears at energies greater than 1 billion electron volts (1 GeV). Brighter colors indicate brighter gamma-ray sources. (NASA/DOE/Fermi LAT Collaboration)

But even more important than what was seen was what Explorer 11 did not see. Half a century ago, the Steady State Theory of the Universe was quite popular. In this cosmological theory, new material in the form of matter and antimatter was constantly being produced to fill the ever expanding Universe. But unless there was some special unknown mechanism at work that somehow kept the matter and antimatter apart indefinitely, eventually the antiprotons produced would encounter normal protons and emit gamma rays that Explorer 11 could detect at an estimated rate of 3,000 per hour. The lack of such a detections was one more piece of evidence against the Steady State Theory bolstering the relatively new alternative model, the Big Bang Theory, for the origin of the Universe. While it would be several years before an improved gamma ray detector would be flown in orbit, the experience of Explorer 11 just goes to prove how a simple experiment flown on a microsatellite can change the scientific view of the Universe around us.

Follow Drew Ex Machina on Facebook.

Related Reading

“Vintage Micro: The Second-Generation Explorer Satellites”, Drew Ex Machina, September 3, 2015 [Post]

“Vintage Micro: The Amateur Space Telescope”, Drew Ex Machina, April 16, 2014 [Post]

General References

Josef Boehm, Hans J. Fichtner, and Otto A. Hoberg, “Explorer Satellites Launched by Juno 1 and Juno 2 Space Carrier Vehicles”, in Aeronautical Engineering and Science, Ernst Stuhlinger, Frederick I. Ordway III, Jerry C. McCall, and George C. Bucher (editors), pp. 218-239, McGraw-Hill, 1963

Ray V. Hembree, Charles A. Lundquist, and Arthur W. Thompson, “Scientific Results from Juno-Launched Spacecraft”, in Aeronautical Engineering and Science, Ernst Stuhlinger, Frederick I. Ordway III, Jerry C. McCall, and George C. Bucher (editors), pp. 281-297, McGraw-Hill, 1963

W.L. Kraushaar and G.W. Clark, “Gamma Ray Astronomy”, Scientific American, Vol. 206, No. 5, pp. 52-61, May 1962

W. Kraushaar et al., “Explorer XI Experiment on Cosmic Gamma Rays”, The Astrophysical Journal, Vol. 141, No. 3, pp. 845-864, April 1, 1965