Figure 4-24 — Titanium processing.
Image
³ Information about commercial aluminum ventures is extremely tentative and difficult to obtain because of proprietary interests.
Figure 4-25 — Aluminum processing.
<!-- image -->APPENDIX J: GLASS PROCESSING
The chemical composition of lunar samples returned by the Apollo missions has been determined, and a wide variation of percentages by weight of various constituents is apparent (refs. 38, 39).
Silica ($SiO_2$) composes approximately 40-50 percent of the lunar soil, and in abundance are also found oxides of aluminum ($Al_2O_3$), iron ($FeO$), magnesium ($MgO$), calcium ($CaO$), and titanium ($TiO_2$). Oxides of sodium ($Na_2O$), potassium ($K_2O$), phosphorus ($P_2O_3$), manganese ($MnO$), and chromium ($Cr_2O_3$) are present in less than 1 percent.
In the manufacture of window sheet and plate glass, soda-lime glass is most commonly used. Its composition includes approximately 71-73 percent silica ($SiO_2$), 12-14 percent soda ($Na_2O$) and 10-12 percent lime ($CaO$) (ref. 40). Soda is absent from the lunar soil in percentages needed for producing commercial soda-lime glass and it proves to be a costly item to supply from Earth.⁴ Fortunately, it does not appear to be necessary to supply additional $Na_2O$. Commonly used oxides in commercial glass which, if desired, would by necessity be additives to be mixed with the lunar materials, include lead oxide ($PbO$), used to provide X-ray and gamma-ray protection by absorption (ref. 41) and boric oxide ($B_2O_3$), used when good chemical resistance, high dielectric strength, and low thermal expansion are desirable (ref. 42).
A simplified diagram of industrial glass-making is given in figure 4-26 with lunar soil listed as the raw material to be provided. After additives (if any) are mixed with the lunar soil, the acid leaching stage removes undesirable materials from the mixture, such as iron oxides which degrade the transmissivity of the glass. If it is decided to produce almost pure silica glass (> 95 percent $SiO_2$) almost all of the non-silicate constituents are leached out with acid at this stage. The furnace temperature needed to melt pure silica (~1700° C) is higher than that needed for soda-lime glass (~1550° C) but is well within the limits of the solar furnace to be utilized. Requirements for providing these temperatures are calculated as follows:
Assumptions:
- 40 t/day maximum production schedule
- 24 hr work day
- Mean specific heat (0°-1700° C) 1.13 J/g °C for silica glass (ref. 43).
- Mean specific heat (0°-1550°C) 1.21 J/g °C for soda-lime glass (ref. 43).
- Insolation, 1.39 $kW/m^2$.
For silica glass, 890 kW (640 $m^2$ solar collector) is needed. For soda-lime glass, 870 kW (626 $m^2$ solar collector) is needed.
Volume requirements for the processing plant are governed basically by the volume needed for the melting tank and the annealing lehr. Although most industrial processes for plate glass involve capacities substantially greater than 40 t/day, the following estimates based upon a scaling down from larger systems are given as guidelines: approximately 23 m in length for the tank, 91-122 m in length for the annealing lehr, 1-2 m in depth, and, to allow for trimming, a width of 0.3-0.4 m in excess of the desired width of the glass panels. Note that some processes have annealing lehrs of only 10 m in length (ref. 44). For panels 0.5 m in width, the volume of the tank and lehr assembly using the upper values of the above dimensions is about 280 $m^3$. Additional space must be provided for the preliminary processing and cutting phases.
A rough estimate for the weight of a plant processing 40 t/day is 400 t.
⁴ Blumer, J., Libbey-Owens-Ford Company, Toledo, Ohio, personal communication, Aug. 1975.
Figure 4-26 — Glass processing.
Image
APPENDIX K: THE LUNAR GAS GUN MASS DRIVER
The importance of obtaining the mass of material from the Moon has been emphasized earlier. Overall probability of success of the entire system is substantially improved by having an alternative to the transport linear accelerator (TLA) and the mass catcher that has been selected as the primary or baseline system for transporting 10 million tonnes of lunar material to L5 over a period of 10 yr.
The lunar gas gun is a fundamentally different concept to that of the TLA. Where the TLA launches small payloads with very high repetition rate onto a precisely determined trajectory, and has the small individual loads caught in a localized active net or passive catcher, the gas gun launches large payloads with a much lower repetition rate onto a less precisely determined trajectory and has these large payloads collected by remote controlled interceptor rocket engines.
The gas gun has four primary elements in its system: a) the launching barrel on the lunar surface, b) the energy storage system which uses compressed hydrogen to propel the payload, where the propulsion energy is obtained from the hydrogen gas maintained at a pressure of 200 MPa in a blast hole deep beneath the lunar surface, c) the compressor which maintains the gas at high pressure and is driven by a nuclear turbine, and d) the interceptor rockets.
Launching Barrel
The equations that govern the mass of the launching barrel are:
$\sigma_w = PR/t$ $E = (1/2)mv^2$ $M = 2\pi RLt\rho$
where
- $\sigma_w$ is the working stress of the barrel material, Pa
- R radius of the barrel, m
- t thickness of the barrel, m
- P internal pressure (presumed constant), Pa
- L length of the accelerating region, m
- m mass of the projectile, kg
- v muzzle velocity, m/s
- M mass of the barrel, kg
- $\rho$ density of the barrel material, $kg/m^3$
These equations together yield the following approximate expression for the barrel mass which depends only on the properties of the barrel material, the muzzle velocity, and the mass of the projectile:
$M = \frac{\rho}{\sigma_w} \frac{mv^2}{2}$
For hydrogen at 200° C, when the thermodynamic properties of motion near the speed of sound are taken into account, this expression becomes:
$M = 1.2 \frac{\rho}{\sigma_w} \frac{mv^2}{2}$
For lunar escape velocity of 2370 m/s, and for a boron or graphite filament epoxy such as PRD-49 of density $1.38 \times 10^3$ $kg/m^3$ and an allowable working stress of 1650 MPa, the mass ratio of barrel to projectile is 24.5.
Projectile Mass
Given a desired mass flow rate of $10^6$ t/yr the projectile mass varies inversely as the launch repetition rate for a single barrel. Since large payloads are desired at low repetition rates, the largest projectile that can be handled on the lunar surface represents the best solution to this element of the system taken alone. Since a 10-t shell is a reasonable size for a shell on the Earth's surface, a 57-t projectile is assumed to be manageable on the lunar surface. With this value the desired mass flow rate can be achieved with a repetition rate of 2 launches per hour. A cylindrical projectile of this mass sintered from lunar material and having a density presumed to be 2.5 can be about 2 m in diameter and 4 m long. The mass of the barrel for a projectile of this size is about 1400 t.
Gas Storage
The compressed gas can be stored in a deep sublunar hole which can have a diameter of approximately 30 m and can, if necessary, be lined with a heavyweight plastic film. The mass of the film is not expected to exceed 5 t.
Nuclear Compressor
The average power required to launch 1 million tonnes at lunar escape velocity over the course of a year is 89 MW. The energy for this comes from a nuclear turbine and gas compressor. For a nuclear electrical power generator the study assumes a mass-to-power ratio of 45 t/MW (exclusive of shielding). Although an electrical generator is inherently more efficient than a gas compressor, perhaps by a factor of 2, the compressor has a much lower ratio of mass to power throughput, a factor of from 5 to 10. The combined effect on the overall energy source is assumed to reduce the mass to power ratio to 27 t/MW. Thus the mass of the nuclear compressor may be taken as approximately 2400 t.
Remote Controlled Interceptor Rockets
The large blocks of lunar material arrive in the vicinity of L5 at an average rate of one every 1/2 hr. With an anticipated launch velocity error of ±0.5 m/s, the radius of the scatter circle is approximately 1000 km. If 50 interceptor rockets are used to collect the lunar material the rotation time to intercept and dock is 24 hr. If this maneuver is carried out with as small velocity change as possible, the average power required to collect the lunar material is 1.2 kW, or 37.8 J/kg of material.
A more accurate study of this part of the problem is desirable. Perhaps the interceptors could rendezvous and correct the trajectory at a point closer to the Moon. It might also be possible to schedule the timing of launches and the direction of trajectories to take advantage of the relative velocities of the Moon and the L5 processing location which is in a large orbit around the L5 point.
Lunar Gas Gun Summary
Many aspects of this system need to be studied further. For example, the thermodynamic efficiency should be determined with greater accuracy. It is now merely assumed to be nearly 1 percent because the temperature of the gas never differs from the ambient temperature by more than about 20° C and then only for a short time. The process is, therefore, considered to be a quasi-isentropic, adiabatic, expansion/compression. No losses at the valves are taken into account. The valves can be seen in the schematic of the launcher, figure 4-27, which indicates that the launching gas is not allowed to escape. Some indication is also given in the figure that fine velocity control might be developed to reduce the scatter circle at L5.
For comparative purposes, the component masses of the lunar gas gun are compared in table 4-16 with the corresponding masses of the transport linear accelerator.
TABLE 4-16 — COMPARISON OF THE MASSES OF THOSE PARTS OF THE TLA AND THE LGG MASS DRIVING SYSTEMS THAT MUST BE LOCATED ON THE LUNAR SURFACE (IN TONNES)
| Component | TLA | LGG | | :--- | :--- | :--- | | Launcher | 1000 | 1400 | | Power Source | 4000 | 2400 | | Gas Storage | — | 5 | | Total | 5000 | 3805 |
Figure 4-27 — Lunar gas gun.
Image
APPENDIX L: PASSIVE CATCHERS
Two different designs for passive catchers, although rejected for the baseline, were considered in some detail. Figure 4-28 illustrates one version of a passive catcher. It consists of two major parts: a slowly spinning bag of Kevlar fabric, and a non-spinning rim which contains power, propulsion and other necessary systems. (Such "dual-spin" designs are commonly used in satellites.) To keep the mass of the catcher within reasonable limits, it is limited to 100 m in radius. Consequently the dispersion of the incoming payloads must be less than this. After the incoming masses arrive within the 100 m radius of the target area, they strike a grid of cables across the mouth of the catcher and break up, thereby releasing a spray of fine particles and small gravel-size rocks which fly inward. These particles strike the bag at up to 200 m/s (23 ply Kevlar stops 44 magnum bullets fired point blank) and come to rest against the surface of the bag where they are held by centrifugal force thereby preventing them from drifting free and escaping from the mouth. Uncertainty about achieving the accuracy of launching required for this catcher, and also some doubt about whether the material would break up and be contained as planned, led to the rejection of this alternative.