NASA’s Viking Mission & The Search for Life on Mars: The Experiments

For young space enthusiasts like myself growing up in the 1970s, NASA’s Viking mission to Mars was one of the more exciting and memorable. The Viking landers, which touched down on the Martian surface on July 20 and September 3, 1976, made the first in situ search for life on another world using a suite of miniaturized laboratories incorporating the cutting edge technology of the time (see “Growing Up in the Space Age: Summer Vacations of the 1970s”). While we all hoped to find life on the Red Planet, in the end the experiments produced disappointing results that can most generously be regarded as ambiguous at best. This failure, in part, was responsible for a 16-year hiatus in new NASA missions to Mars and the reluctance to fly any new experiments designed specifically to detect extraterrestrial life even after almost half a century. What follows below is a close look at the Viking life detection experiments, how they operated, and the assumptions they made.

 

The Viking Program’s Beginning

The Viking program officially began on February 9, 1969, when NASA Administrator Thomas Paine signed the project approval documents. Its predecessor, the Voyager program (not to be confused with the program of the same name that successfully explored the outer planets between 1979 and 1989), proved to be an overly ambitious and expensive project that called for a series of complex landers to be launched to Venus and Mars during the 1970s. When this project was finally cancelled in the summer of 1967, a series of proposals developed by NASA’s Langley Research Center took an early lead as a relatively low-cost replacement for Voyager. One family of proposals, the “Titan Mars 1973” missions, called for a pair of lander-carrying orbiters to be sent to Mars during the 1973 launch opportunity to Mars using a Titan III-class launch vehicle.

Despite its more modest design, Viking was still a very advanced and complex spacecraft in its final incarnation. To minimize risk and development costs, the orbiter simply consisted of an enlarged version of the of the Mariner-class spacecraft bus developed and built by the Jet Propulsion Laboratory (JPL) to explore Venus and Mars. With this cost-saving measure, the bulk of the development effort and budget could be poured into the lander and its instrumentation. While it borrowed heavily from previous studies done for Voyager (see “The Automated Biological Laboratory”), as well as the successful Surveyor program that sent landers to the Moon between 1966 and 1968 (see the Surveyor Program Page), much work remained to be done. Within a year of the Viking project’s official start, the proposed launch date was pushed out to the summer of 1975 to ease budget demands and give more time to design and build the lander and its suite of instruments employing the cutting-edge technology of the time.

 

The Gas Chromatograph – Mass Spectrometer (GCMS)

Almost since its founding in October 1958, NASA and the scientific community had been developing life-detection techniques that could be used on Mars. With a total science payload of 32 kilograms for each Viking lander, only a select few of these proposals could be accommodated. Based on their mass, development costs, and the assessed likelihood of detecting what biologists thought Martian life would be like, five experiments were finally chosen by the end of 1969 to fly to Mars on Viking. The total mass of these instruments was about 15 kilogram or almost half the landers’ total instrument payload.

Probably the most important instrument to investigate the biology of Mars was the Gas Chromatograph – Mass Spectrometer (GCMS) which was developed by JPL and built by Litton Industries. The Molecular Analysis team responsible for this experiment, formed in February 1969 and led by Klaus Biemann (MIT),  consisted of Duwayne M. Anderson (US Army Terrestrial Science Center), Melvin Calvin (University of California – Berkley), Leslie E. Orgel (Salk Institute), John Oro (University of Houston, later NASA Ames Research Center), Tobias Owen (Illinois Institute of Technology Research, later State University of New York – Stony Brook), Garson P. Shulman (JPL), Priestley Toulmin (USGS), and Harold C. Urey (University of California – San Diego).

The dual purpose of the GCMS was to analyze the composition of the Martian atmosphere and soil. The heart of the instrument was the mass spectrometer, which first ionized the gas sample it was to examine, then separated and measured the atomic mass-to-charge ratio (m/e) of the resulting ions by electromagnetic means. Since the ion fragments of every compound have a unique m/e spectrum or fingerprint, the mass and identity of the original compounds contained in the sample can usually be identified. Because it had to analyze both low atomic mass atmospheric gases and heavier organic compounds extracted from the soil, the mass spectrometer’s design was optimized to measure the m/e of ion fragments in the 11.5 to 215 atomic mass range with a mass resolution of about one part in 200. Gases as light as methane (with an atomic mass of 16) to organic compounds containing as many as ten carbon atoms could be detected by this mass spectrometer. Concentrations of atmospheric hydrogen and helium could not be measured by the mass spectrometer but a similar instrument in the lander’s Entry Science package, specifically the Upper Atmospheric Mass Spectrometer (UAMS), housed in the aeroshell could, so there was no significant loss to science by this engineering compromise.

For atmospheric analysis, the gas sample typically was first passed through a chemical filter to remove all of the carbon monoxide and carbon dioxide it contained. These gases constitute about 90% of the Martian atmosphere and their removal allowed for a more sensitive measurement of the interesting minor atmospheric components such as oxygen, nitrogen, argon, and krypton. The presence of certain gases in nonequilibrium concentrations could also serve as an independent biosignature suggesting biological activity on Mars. Nonequilibrium amounts of methane in Earth’s atmosphere are the result of biological activity and the same could be true for other planets with active biospheres exchanging gases with its atmosphere. The finite absorption capabilities of the mass spectrometer’s chemical filter limited the number of filtered atmospheric gas sample to a total of 60. Unfiltered analyses were also possible so that measurements of the concentrations of carbon monoxide, carbon dioxide, and water vapor could be made.

 

Detecting Organic Compounds in the Soil

Since the mass spectrometer could only handle gases, the analysis of solid soil sample presented some difficulties. In a terrestrial laboratory, a soil sample would normally have its organic compounds removed using wet chemical methods followed by solvent extraction and various forms of chemical separation. Miniaturizing such a complicated process with the technology available around 1970 within the budget, mass, and volume constraints imposed on the Viking GCMS was not possible. The simplest and most reliable method available to liberate any organic compounds present in the soil sample was through thermal volatilization. With this technique, the soil sample would first be heated so that its organic compounds would vaporize. Because of the fragility of some organic compounds, the process could break them down into smaller fragments. Still, it is possible to determine the identity of the original parent molecule in cases of thermal degradation.

In the procedure finally adopted, a 100-milligram soil sample secured by the lander’s sampling arm would first be ground and passed through a 0.3-millimeter sieve, then deposited into one of three miniature electric ovens located in a motorized holder. Each ceramic oven had a chamber that was two millimeters in diameter and 19 millimeters long. The sample was then heated to a preselected temperature of 50°, 200°, 350°, or 500° C to drive off its volatile organic compounds. The pyrolysis temperature, which could be reached in one to eight seconds, could be chosen from a preprogrammed sequence or commanded directly from the Earth. Any organic compounds released by the sample during the 30-second heating process were swept away by a two or three cubic centimeter puff of isotopically pure carbon dioxide labeled with carbon-13. This was done to distinguish the purge gas from any carbon dioxide released by the sample (which would be dominated by the carbon-12 isotope). After leaving the test cell, the gas was mixed with a stream of hydrogen carrier gas.

Instead of proceeding directly to the mass spectrometer for analysis, this gas sample first passed through a gas chromatograph (GC), which consisted of a two-meter long column packed with a liquid-modified organic adsorbent consisting of 60 to 80-mesh Tenax-GC coated with polymetaphenoxylene. The GC column allowed the quick passage of water vapor and carbon dioxide through the system while delaying the passage of the more interesting organic compounds. The length of the delay was dependent on the compounds’ adsorptive properties. This lessened the load on the mass spectrometer further downstream and allowed for the differentiation of organic compounds with similar m/e spectra but differing chemical structure and properties.

Beyond the GC was an effluent divider that diverted and vented any excess gas, especially water vapor and carbon dioxide, through a series of restrictor valves and vents. This would prevent components further downstream from becoming overloaded and permanently damaged by the large amounts of gas that a heated soil sample could generate. From here the gas sample continued on to a hydrogen separator which consisted of a 60-centimeter long silver-palladium alloy tube that was porous only to small hydrogen molecules. This device would remove all but 0.5 part per million (ppm) of the hydrogen carrier gas from the sample stream by electrochemical means and would lessen the burden on the mass spectrometer’s ion pump.

The resulting effluent was then monitored by the mass spectrometer which repetitively scanned the entire m/e spectrum every ten seconds. During each scan the mass spectrometer produced a 3,840-sample m/e spectrum that was first converted to a logarithmic scale then digitized to 9 bits. During a typical measurement run, 17 megabits of data would be generated. Originally this raw data was to be analyzed on board the lander and only the most interesting results transmitted back to Earth in order to lessen the data load. But with the availability of a high-capacity magnetic tape recorder on the lander and an increase in the lander’s data transmission rate from 4 to 16 kilobits per second, it became possible for the raw GCMS data to be transmitted directly to Earth. Once in the hand of investigators, they could perform a much more thorough data reduction and analysis than would be possible on the lander.

Ground testing of the GCMS using Antarctic soil samples (the closest analog on Earth of the conditions expected on Mars) indicated that the instrument was capable of detecting and differentiating between a huge variety of organic compounds at levels from part per million to parts per billion or better. Even if the other biological experiments failed to detect any life that might be present due to faulty assumptions in their design, the sensitivity of the GCMS allowed the detection and analysis of its constituent organic compounds. Because the GCMS was optimized for the detection of organic compounds, its ability to analyze volatile-bearing minerals was greatly reduced. The procedure used by the Viking GCMS typically called for rapid heating to 350° to 500° C. Reliable pyrolytic analysis of water or carbon-bearing minerals requires much slower heating to temperature as high as 1,000° C. As a result, the GCMS could offer little data to constrain the proportions of hydrated minerals or carbonates that might exist in the Martian soil.

 

The Biology Instrument Package

The Viking landers’ biology package, built by the Applied Technology Division of TRW Defense and Space Systems group, was the responsibility of the Active Biology team formed in February 1969. Originally, the team consisted of Norman H. Horowitz (California Institute of Technology), Joshua Lederberg (Stanford University), Gilbert V. Levin (Biospherics Research, Inc.), Vance I. Oyama (NASA Ames Research Center), Alexander Rich (MIT), and Wolf Vishniac (University of Rochester). Harold P. Klein (NASA Ames Research Center) would join the team shortly afterwards and become its leader.

Originally the biology instrument package was to consist of four separate miniaturized life-detection experiments packed to fit within a volume of only a single cubic foot (about 0.03 cubic meters). Each biological test sample obtained from the Martian surface using the lander’s remote sampling arm was delivered to the package’s processing and delivery assembly. The sample then passed through a 1.5-millimeter screen and was subsequently split and gravity fed into the test cells of the various experiments. Each experiment analyzes a 0.25 to 1 cubic centimeter sample of loosely packed surface material. Ultimately each experiment was built with the ability to perform tests on four samples. In order to avoid terrestrial contaminants, the instrument was assembled under clean room conditions and was heat sterilized in dry nitrogen for 57 hours at 120°. After installation, the lander and its aeroshell were encased in a protective bioshell and was itself heat sterilized for 40 hours at 112° C. This procedure reduced the risk of biological contamination to less than one chance in 10,000.

 

Pyrolytic Release Experiment

All of the biological experiments made the assumption that Martian organisms used gaseous or liquid organic nutrients supplied by the instrument to yield products that could be detected over the lifetime of the test. The first experiment, originally called the carbon assimilation experiment but later known as the Pyrolytic Release, was performed under the most Mars-like conditions. This experiment assumed that Martian life would make use of common gases in the atmosphere to produce organic compounds via photosynthesis or some other form of autotrophism much as organisms on the Earth do.

At the start of the Pyrolytic Release experiment, a 0.25 cubic centimeter soil sample was sealed in one of four test cells with a volume of four cubic centimeters. Next, 0.02 cubic centimeters of gas consisting of 92% carbon dioxide and 9% carbon monoxide that had been “tagged” with radioactive carbon-14 was added to the test cell. The tracer gas had a total radioactivity of 22 microcuries and its addition resulted in a modest 2.2 millibar increase above the ambient Martian surface pressure of 6 or 7 millibars. If Martian organisms used the carbon-carrying gases in the chamber, the carbon-14 would be incorporated into the organic compounds they produced. The sample could be provided with simulated sunlight using a xenon lamp at an illumination level 20% of Mars’ maximum surface daylight intensity to support photosynthesis. This light source was equipped with a filter to remove ultraviolet light with wavelengths shorter than 320 nanometers. It was felt that the shorter wavelengths could either kill any Martian organisms (giving a false negative experiment result) or unintentionally produce organic compounds through nonbiological photocatalytic reactions (which would produce a false positive result). Water vapor could also be added to the test cell in 80-microgram increments on command.

After an incubation period of 120 hours at a nominal temperature of 10° C, the sample was heated to 120° C to drive off the residual tracer gas. Finally the sample was pyrolyzed at about 625° C to drive off any organic compounds that had formed by the biological assimilation of the tracer gas. The evolved gases were then flushed with helium through a column packed with a hot mixture of 75% Chromosorb-P (a form of diatomaceous earth) and 25% cupric oxide. This mixture allowed the unused tracer gas to pass through while absorbing any organic compounds heavier than methane. After the unused gas obtained from the purge cycle was analyzed, the column was heated to 640° C so that any organic compounds it adsorbed were released and oxidized into carbon dioxide via reactions with the cupric oxide.

The gases from the purge and pyrolysis stages were monitored for the presence of the radioactive carbon-14 tracer using a β particle detector. The first wave of purge gas was referred to as “peak 1” and served as an indicator of unreacted tracer gas. The second wave of gas from the pyrolysis of the sample was known as “peak 2”. The size of peak 2 determined the extent to which any organisms in the sample had assimilated the tracer gas. As a control, the Pyrolytic Release experiment could be repeated with a portion of the original soil sample that was heat sterilized for three hours at 175° C. According to the protocols established by the active biology team before the mission, the presence of peak 2 in the first test followed by none in the sterilized control run indicated the presence of living organisms in the soil.

 

Labeled Release Experiment

The prototype for the second experiment was popularly known by the name “Gulliver”, named after Jonathan Swift’s fictional traveler to strange places. The essential elements of Gulliver were retained and incorporated in Viking’s Labeled Release experiment. Biologist Gilbert Levin started work on this experiment in 1959. Within just seven years its hardware development had advanced to the point where it was the leading contender to fly to Mars.

The Labeled Release experiment was conducted under less Mars-like conditions than the Pyrolytic Release, but used similar principles. In this experiment, a soil sample was placed into one of four test cells and then injected with an aqueous solution containing “labeled” glycine, DL-alanine, sodium formate, DL-sodium lactate, and calcium glyconate. The nutrients were labeled by having some of their normal carbon-12 atoms replaced with radioactive carbon-14 so that they could be tracked. The wetted sample was allowed to incubate in the dark for about eight days at a nominal temperature of 10° C with an atmospheric pressure slightly higher than normal for the Martian surface.

The gas in the chamber was then monitored over time using a β particle detector. The presence of any radioactive carbon dioxide or other volatiles would imply that organisms were metabolizing the labeled nutrients. By measuring the amounts of tagged gases generated by the sample as a function of time, information on the reproduction rate and physiological state of the microorganisms present could be obtained. In case the experiment produced a positive signal, the Labeled Release process could be repeated with a control sample that had been heat sterilized for three hours at 160° C.

 

Gas Exchange Experiment

Next was the Gas Exchange experiment. The conditions under which it operated were significantly less Mars-like than those in either the Pyrolytic Release of Labeled Release experiments. A one cubic centimeter sample was placed into a single, dark 8.7-cubic centimeter test cell with an atmosphere of 91.65% helium, 5.51% krypton, and 2.84% carbon dioxide at a pressure of 200 millibars and a nominal operating temperature of 10° C. The Gas Exchange experiment had three different modes of operation: The Humid Mode introduced 0.5 cubic centimeters of a nutrient solution rich in organic compounds and inorganic salts into the test cell with minimal wetting of the soil sample. In the Wet Mode, two cubic centimeters of nutrients were added to the soil sample to be tested. Finally, there was a Dry Mode where no nutrient solution was added. The relatively high atmospheric pressure under which the Gas Exchange experiment was conducted was needed in part to prevent the nutrient solution from instantly vaporizing as they would under ambient Martian surface conditions. Most of the added gas, helium and krypton, was inert and not be involved in any biological reactions present while carbon dioxide concentration, with a partial pressure of about 6 millibars, was Mars-like. Because of the relatively large amounts of rich nutrients added by the Gas Exchange experiment, it was popularly known as the “chicken soup” experiment.

After a suitable incubation period, a portion of the gas in the test cell was removed for analysis by a gas chromatograph independent of the GCMS. This was done, in part, in keeping with the mission design philosophy that a failure in one of the lander’s instruments should not jeopardize any other experiment addressing a primary mission science objective. Much simpler than the GCMS, the Gas Exchange experiment’s gas chromatograph could measure the abundances of hydrogen, neon, nitrogen, and argon or carbon monoxide as well as nitrous oxide, methane, krypton, carbon dioxide, nitric oxide, and hydrogen sulfide. The krypton added to the test cell’s atmosphere served as a calibration standard for this instrument. The changes with time in the composition of the gases in the test cell could indicate the presence of life and give hints about its metabolism. Methane and carbon dioxide are frequent byproducts of terrestrial organisms living in dark, oxygen-free environment and it was hoped that Martian lifeforms would behave similarly. The single Gas Exchange experiment’s test cell was purged and dried between experiment runs with helium gas and then could be repeated with a sterilized control sample that was heated to 145° C for 3½ hours.

 

The Light Scattering Experiment

The fourth and final experiment in the original suite of biological instruments was the light scattering experiment. It was popularly known as the “Wolf Trap” after its developer, Wolf Vishniac. Vishniac developed the original Wolf Trap between 1958 and 1960 to demonstrate for the first time the feasibility of remote, automatic life detection. Promising early results led to the subsequent development of more complex breadboard models during the 1960s with funding from NASA. Like the Gulliver experiment, this experiment’s advanced state of development made it an early contender to fly to Mars.

In this experiment, which had the least Mars-like conditions, the soil sample was placed directly into a nutrient-rich solution. If microorganisms were present, they would grow and multiply in the solution making it cloudy over time. Any change in the light-scattering properties of the nutrient solution was detected by a simple light sensor. Early versions of the Wolf Trap also measured the pH of the solution over time as an independent check on microorganism growth and metabolism. Early trials with sample from Antarctica and other locales indicated this system was incredibly sensitive and was capable of detecting the presence of as few as a thousand microorganisms after incubation. The high sensitivity was a definite asset if Martian microbes possessed a rather sluggish reproduction rate.

Taken together, these original four experiments were able to test for the presence of life under a wide variety of environmental conditions. The Pyrolytic Release experiment operated under the most Mars-like conditions. Except for the addition of modest amounts of a nutrient solution, the Labeled Release experiment was also performed under conditions that might conceivably be found in favorable oases on the Martian surface. The Gas Exchange and Light Scattering experiments both operated under conditions that have not existed on Mars for millions if not billions of years. While they were both designed under the assumption of more Earth-like conditions on the Martian surface, the designers felt that they were best suited for detection of dormant lifeforms or spores that might be waiting for the return of conditions more amenable for life (at least by terrestrial standards). Given the limits of the then-available technology and development budget, it was the best set of test that biologists could fly on Viking.

 

Preparing for the Viking Mission

The development of the Viking spacecraft and its experiments were wrought with technical problems, schedule delays, and cost overruns. Nobody anticipated the true complexity of developing all the instruments, especially those designed to detect life on Mars. The total cost of the GCMS skyrocketed from the $17.8 million estimate in June 1970 (about $130 million in today’s dollars) to $41.2 million actually spent over the next six years (about $215 million today). Likewise, the costs of the biological package increased from an initial estimate in September 1970 of $13.7 million (about $104 million today) to a whopping $59.5 million ($310 million today) in actual costs by the time of launch.

In the process of reigning in not only the costs of building these instruments, but also their growth in mass and volume, many engineering changes were made during development limiting the flexibility of the experiments as well as the number of analyses that could be performed. The most drastic change was the deletion of the Wolf Trap in March 1972, much to the disappointment of the Active Biology team. Despite the cancellation of the experiment he had developed over 14 years, Wolf Vishniac continued on as the assistant team leader until his untimely death on December 10, 1973 from an accidental fall while he was retrieving equipment during a research expedition to Wright Valley in Antarctica.

Despite the delays, the flight units for the Viking life detection experiments were delivered in early 1975 for integration into the landers. Following assembly and testing, both landers were successfully sterilized by the end of June and integrated with their respective orbiters. By mid-July, the long process of designing, building, assembling, testing, and flight preparations was finally drawing to a close with the pair of Viking spacecraft ready for launch during a 65-day launch window.

 

The details of the results of the Viking life detection experiments will be the subject of a future article.

 

Follow Drew Ex Machina on Facebook.

 

Related Videos

Here is a NASA-produced documentary from 1974, Mars, The Search Begins, about the Viking search for life on Mars:

 

 

Here is another NASA documentary released in 1976, entitled Viking, with an overview of the Viking mission before the spacecraft reached Mars:

 

 

Related Reading

“The Automated Biological Laboratory”, Drew Ex Machina, August 7, 2014 [Post]

“Growing Up in the Space Age: Summer Vacations of the 1970s”, Drew Ex Machina, July 22, 2019 [Post]

“A Cautionary Tale of Extraterrestrial Chlorophyll”, Drew Ex Machina, October 5, 2014 [Post]

 

General References

K. Bieman et al., “The Search for Organic Substances and Inorganic Volatile Compounds on the Surface of Mars”, Journal of Geophysical Research, Vol. 82, No. 28, pp 4641-4658, September 30, 1977

K. Bieman and John M. Lavoie, “Some Final Conclusions and Supporting Experiments Related to the Search for Organic Compounds on the Surface of Mars”, Journal of Geophysical Research, Vol. 84, No. B14, pp 8385-8390, December 30, 1979

Carl W. Bruch, “Instrumentation for the Detection of Extraterrestrial Life”, in Biology and the Exploration of Mars (ed. Colin S. Pittendrigh, Wolf Vishniac, and J.P.T. Pearman), NASA/NRC Publication 1296, pp 487-502, 1966

William Corliss, The Viking Mission to Mars, NASA SP-334, 1975

Edward Clinton Ezell and Linda Neuman Ezell, On Mars: Exploration of the Red Planet 1958-1978, NASA SP-4212, 1984

N. H. Horowitz and G. L. Hobby, “Viking on Mars: The Carbon Assimilation Experiments”, Journal of Geophysical Research, Vol. 82, No. 28, pp 4659-4662, September 30, 1977

Harold P. Klein et al., “The Viking Biological Investigation: Preliminary Results”, Science, Vol 194, No. 4260, pp 99-105, October 1, 1976

R. K. Kotra, E.K. Gibson, and M.A. Urbanic, “Release of Volatiles from Possible Martian Analogs”, Icarus, Vol. 51, pp. 593-605, September 1982

Vance I. Oyama and Bonnie J. Berdahi, “The Viking Gas Exchange Experiment Results, from Chryse and Utopia Surface Samples”, Journal of Geophysical Research, Vol. 82, No. 28, pp 4669-4676, September 30, 1977

Wolf Vishniac, “Extraterrestrial Microbiology”, Aerospace Medicine, pp 678-680, August 1960

R. S. Young, “The Origin and Evolution of the Viking Mission to Mars”, Origins of Life, Vol. 7, pp 271-272, 1976