Proceeding along the path to colonizing Mars. Part 1 described two possible super-heavy lift rockets constructed on paper with mostly reasonable assumptions. Part 2 will cover the transit habitat.
I've already discussed the hazards involved in a manned trip to Mars in previous posts, but the two most important factors are radiation and microgravity. Less critical but still important are life support, food supplies, medical service, psychological health, maintenance and cost.
Read on after the break to see how these challenges can be addressed.
There is only one known way to avoid the sweeping decay of microgravity on the human body: simulate gravity by spinning. We don't yet know how much pseudogravity is required. (From this point forward I will simply say gravity when I mean artificial gravity.) Since this project focuses on colonizing Mars we have to assume that Martian gravity is sufficient for long-term human health. That's important because there are limits to how fast a structure can spin before the passengers can't handle the motion sickness. A bigger structure can spin slower for the same gravity or can spin at the limit of tolerance for higher gravity than a smaller structure. There are some tricks for getting a useful level of gravity in a small module but I'm looking at moving large numbers of people, so a large module is preferred. To be specific, a module that spins at 4 RPM to produce Mars gravity on the outer level must be about 43 meters in diameter. Faster spin rates are likely to cause motion sickness; 2 RPM and Earth gravity would be even better but that would require a huge module.
Such a large module must either be constructed in orbit or be built as an expandable unit and launched as a much smaller package. Expandable is preferred as the work can be done on Earth, inspected and pressure tested before launch. Building a habitat out of small components would take a lot of time and effort. Using larger 'tin can' modules with automated docking would be much easier and is feasible but less efficient. This is a possible fallback option if expandables of this size prove impossible.
The spin gravity of the habitat will be produced by using two sections that spin in opposite directions. Electric motors can spin up the pair without using propellant and without spinning the rest of the craft. A rotary valve will connect the two sections and allow pumping water back and forth to fine-tune the balance.
Likewise with radiation, there is only one known way to prevent damage: mass. Place material between you and the radiation source to reduce your exposure. (On Earth you can reduce exposure by adding distance, but that's not feasible in space where the radiation comes from all directions.) The exact amounts needed vary by the type of radiation, type of shielding material, method of exposure and allowable limits. For reference, the background dose on Earth is around 3.6 milliSieverts (mSv) per year. People working in radiation industries (nuclear power, medical imaging, pilots / air staff) are limited to 50 mSv per year. Astronauts face two limits: 500 mSv in a year and a lifetime limit that depends on age and gender but is typically 2,000 to 4,000 mSv. These values are expected to cause no more than a 3% increased risk of developing cancer, which NASA and the astronaut corps considers acceptable.
Colonists are not likely to make more than one round trip (if they even return), but dedicated crew might need to make several. The actual transit takes around six months (varies, could be 5, could be 7 depending on where we are in the cycle), so a single round trip should take on average one year. The deep space radiation environment averages about 740 mSv per year, way over the limit; this needs to be reduced to a manageable level using the least mass possible.
Radiation shielding can be specified in a couple of ways; most spacecraft design studies state it in terms of grams per square meter and assume that all the mass is aluminum. That's useless for an expandable module since there is little to no aluminum in the outer hull. Another way to define it is by attenuation depth; old fallout shelter manuals from the 60's will list halving thicknesses or tenthing depths of soil or concrete, meaning the thickness of a material necessary to cut radiation by half (or by 90%). Modern sources use the number e as the base, so an attenuation of 1 means reducing radiation by 1/e, or about 36.8%. The benefit of using e is that you can easily calculate how much shielding you need by taking the natural log of expected divided by allowable radiation. ln(740/500) is 0.392 units, so we need at least this much shielding for colonists. If we wanted a permanent workforce to operate habitats then the allowable radiation limit would be 50 mSv and the shielding required would be ln(740/50) = 2.69 units.
Different materials provide different levels of shielding. 1 unit of shielding requires 8cm of titanium, 17 cm of aluminum, 60 cm of water or 52 m of dry air at 1 atmosphere. The units are additive, so once you know the value of each layer of hull you can simply add them up and see if it is enough. The minimum value is 0.392 units and more would be better. The expandable hull provides about 0.02 by itself, but if we build in a 24 cm thick layer of water then it will provide about 0.4 units, leaving a tiny bit of margin. (My previous analysis suggested 20cm of water and assumed the hull, hardware and other mass would make up the deficit; this turned out to be wrong.)
These unitless numbers can be difficult to visualize, so let's look at what that means as a percentage. The fraction of radiation that penetrates a shield is 1 / e^(shield value). For our water shield of 0.4 units that's 0.67, or 67%. Another way to think of it is the percentage of radiation blocked, which is 1 - ( 1 / e^(shield value)). For our water shield of 0.4 units that's 0.33 or 33% as one might expect. A 2.7 unit shield suitable for career crew would allow 1 / e^2.7 or 6.7% of radiation through, blocking 93.3%; this would require a 1.6 meter layer of water (as deep as a residential swimming pool), a 46 cm aluminum shield or a 10.5 cm nickel-iron shield. I also have a design for a permanent colony that uses a meter of packed regolith and half a meter of water plus a thin metal shell to provide adequate protection (4 units).
Lastly, the radiation levels in space are not constant. Sometimes there are solar storms that push the levels much higher than normal for a short time. Surviving these storms requires a storm shelter, a secure area on the craft with much higher shielding than the rest of the habitat. For my modules this will be inside the rigid core section, taking advantage of the mass of all the levels and their equipment as well as a second layer of water shielding. This doubles as the water processing storage, so it is mass that was already needed. The ship points the engine at the sun and points all solar panels and radiators parallel, operating on battery power. If high radiation conditions go on for longer than the batteries can sustain then some of the solar panels will be put back into service; this will reduce their lifespan.
Feeding a person requires about 40m³ of hydroponics volume (including row spacing, aisles, nutrient storage, equipment, seeding areas, etc., etc.). The exact amount varies and can be pushed lower (below 20m³ is my guess) but this is a reasonable number to start with. Plants need light (from LEDs), and hydroponic grow systems need pumps, fans and sensors. This gear collectively eats about 5 kW of electricity per person, nearly all of which ends up as low-grade heat.
That same volume will convert two people's CO2 back into oxygen. This is because most of a plant's dry mass is carbon compounds but only about half of that is food; the rest is waste. That carbon comes from CO2, so crops to feed one person trap a second person's worth of carbon as waste. Fortunately we can simply burn the waste to recover that carbon on demand. This would be done in a pyrolysis unit to form neat little carbon blocks for storage and controlled release; these may even be used as filters before ultimately getting burned to keep the CO2 levels high enough for growth. Since plants and humans have very different 'ideal' atmospheres, CO2 scrubbers (zeolite sorbent beds) will be used to actively move CO2 out of human spaces and into plant spaces.
Nitrogen is a buffer gas; it is inert and allows the oxygen percentage in the atmosphere to be low enough to avoid fires. It's also an important part of amino acids (and thus proteins), and is actively cycled in humans and in plants. Nitrogen gas is very stable, meaning it is hard to convert into active forms like urea or nitrate. Nitrogen fixers like legumes and many bacteria can do the job, plus there are methods of making ammonia and other nitrogen-containing chemicals directly (given enough energy). The active nitrogen compounds in the system are assembled in plants to form proteins, eaten by humans, excreted as waste, separated in a waste processor and converted into hydroponic nutrients to be fed to the plants. Some of it is bound up in plant wastes and is either recovered from there or burned back to nitrogen gas depending on how fancy the recovery system is. Quantities of nitrogen are stored in liquid form to replenish atmosphere (everything leaks). That handles the nitrogen cycle, but note that some of it is unavoidably lost over time and resupply is eventually necessary even with perfect recycling. The same applies to all other volatiles (liquids and gases).
Argon will probably be used as a buffer gas and ion engine propellant since it is readily available on Mars. It doesn't get used in biological processes but otherwise would be used in place of nitrogen for maintaining atmosphere.
The waste processing systems will recover other important nutrients like phosphorus, potassium, calcium, iron, etc., usually as salts that can be used directly as fertilizer after purification. Inputs are wastes, water and energy. Outputs are these salts, clean water, heat and CO2 (for the most part). Water circulates through most of these systems; it is taken up by plant roots and evaporated from leaves, bound up in sugars, eaten, drank and excreted, condensed on cold surfaces, etc. Power consumption is minimal for the baseline (relying heavily on bioremediation and SCWO) but could be higher for more advanced systems.
Humidity and temperature is maintained by the CO2 system, since the air must be dried before it can be filtered properly. Incoming warm, wet 'used' air is cooled and dried, passed through the filter, then heated and moistened to the desired level. The filter regenerates by pumping warm, very dry and very high CO2 air through to the plant sections. Any excess water is sent to waste processing for filtration. This means that the atmosphere system needs active refrigeration and is the primary load on the craft's radiators; it needs to be able to handle the entire heat load of the habitat since only a small amount is lost through the solid metal connections to the rest of the craft. Power consumption is roughly 1 kW per person.
A better use of plant waste would be to make animal feed with it and raise fish and chickens. Food fish have a short enough lifespan that they can be raised inside the radiation shielding water, though breeding pairs would be kept inside the storm shelter. Chicken eggs greatly expand the kinds of foods you can create and would be a big benefit to morale. This kind of microfarming can be done even for a very small population; certainly a ship carrying more than a hundred people would be able to sustain populations of both.
If you've been keeping track, we are up to 6 kW per person for life support. More is needed for lighting, cooking, entertainment, as margin for other equipment I haven't considered and for solar panel degradation over time. Let's assign a nice, round 10 kW per person.
There are a variety of techniques for capturing solar energy. I strongly recommend visiting the National Renewable Energy Laboratory website for comprehensive information on the subject (NREL).
The usual method is to put together a bunch of solar cells, forming panels that are pointed at the sun. Simple silicon photovoltaic panels are cheap, durable, a little bit heavy and efficient enough for most purposes at around 24%. Thin-film panels are much lighter and only a little less efficient (22%). Multijunction cells reach around 38% efficiency but are very complex to manufacture, expensive and usually heavier than single junction cells.
Some improvement can be had using concentrators. This is basically using a mirror to focus extra sunlight onto solar cells. Single-junction and thin-film cells gain a little bit of efficiency (24-28%), but multijunction cells gain quite a bit (44-46%). The benefits are that if your mirror is lighter than the solar cells then you can save a lot of mass, and if you arrange things properly you can shield the cells from radiation and debris. The drawback is that you have to actively cool the cells or they will melt; also, damage to a single cell knocks out a bigger chunk of your total capacity.
If your power needs are truly enormous then concentrating thermal power is an option. Solar heat is used to drive an engine (Sterling usually, though other options are possible) that turns a generator. This is very similar to existing steam-driven generating stations on Earth. In space, however, dissipating waste heat is difficult and expensive. Peak efficiency for a two-stage system can reach 60%.
We need to design for Mars orbit. Solar energy in Earth orbit averages 1,366 W/m², while in Mars orbit it averages 590 W/m². The low and high values are 493 and 717 W/m² respectively. Let's plan for the worst case and revisit if necessary. The baseline area for 10 kW is then 20.3 m² per person, to be modified based on the selected system's efficiency. The next post will include total system mass for power, but we will need to find the per-person values here.
I will select a low-concentrating multijunction system for the initial design. Here is the current state of the art, SOLAROSA (NASA, PDF), with 39% efficiency and 0.404 kg/m² areal density. Advantages of this system are low mass, high efficiency, passive heat management, simple operation and relatively low cost. 52 m² and 21 kg are required per person, though some allowance should be made for structural support and steering for the array.
An alternate design would use high-concentrating multijunction cells. Here is an example (Spectrolab, PDF) from 2011, commercially available, with 40% efficiency. Mass is not specified but with a power output of 200 kW/m² under 500 suns it hardly matters. The mass of the cell would be a rounding error compared to the mass of the radiator equipment. Speaking of, the habitat will already have active radiating equipment, so extending this to the solar panels only increases the necessary capacity. In order to produce this very high concentration we need lightweight mirrors; this can be done with aluminized Mylar or Kapton film as with a solar sail, and with at least 90% efficiency. These materials mass 5 to 12 *grams* per m², so again most of the mass is in the support structure. Mirrors require precise control over shape; an example system to achieve this might be SpiderFab (Tethers Unlimited, PDF), massing about 90 grams per m² with good stiffness. 56.4 m² of reflector per person is required, or 5.1 kg (including solar cell) plus an unknown mass for heat transfer. (That's why it is an alternate; I don't have enough information to estimate the mass of the cooling system for the cells.)
Two main options are available for propulsion: chemical and solar-electric. Chemical systems are fast, which is their primary advantage. Electric systems are much more efficient but can take longer to perform a mission unless they are sized appropriately. I don't see a deal-breaker with either technology, so I will instead design a chemical version and an electric version in the next post.
A ship carrying passengers between Earth and Mars has occasional opportunities to pick up supplies. Anything from fertilizer to 3d printer plastic can be replenished as long as the ship can store a bit under 5 years worth. The ship needs to carry enough spare materials and equipment to complete a round trip even if many components fail.
Colonization will be by intrepid explorer types in the early days, but at some point you need to bring ordinary people who are accustomed to their daily creature comforts. As passengers they will need personal space, privacy, variety in the menu, recreational activities, opportunities to socialize and a way to contribute meaningfully to operations. Even for the best of the best, having some extra space, good food and privacy makes for happier, more productive crew on these very long journeys.