Implications of Molecular Nanotechnology Technical Performance Parameters on Previously Defined Space System Architectures

This paper was presented at the
Fourth Foresight Conference on Molecular Nanotechnology


Thomas Lawrence McKendree
Molecular Manufacturing Shortcut Group
8381 Castilian Drive
Huntington Beach, CA 92646

Nanotechnology Implications on Space Systems

PACS Classification Numbers

07.87 85.42

Table of Contents

List of Tables

List of Figures


Molecular nanotechnology offers the prospect for significant increases in various technical performance parameters, such as material strength and density. This performance would enhance the capabilities of many classes of space systems. To begin analyzing this effect, a first step is to take previously defined space system architectures, not originally intended to use molecular nanotechnology, and calculate how well those systems would perform when simply using the MNT technical performance parameters. This paper discusses chemical rockets for putting payloads into Earth orbit, both single stage and two stage architectures, synchronous and rotating skyhooks, solar sails and solar electric ion engines, and large, inhabited space colonies. In all cases, MNT offers the possibility of significant system improvements.

1. Introduction

Molecular nanotechnology (MNT) has been laid out as a vast, long-term project [1], [2], ultimately leading to a general ability to design and build products to atomic precision, thus providing general applications (figure 1). Research along the development pathways are crucial to any realization of this vision. That realization, however, will have dramatic consequences, requiring significant foresight to properly handle [3]. Developing the appropriate foresight rests upon developing a sufficiently accurate and detailed understanding of the actual capabilities MNT will provide.

Figure 1. Simplified roadmap of the effort to develop and apply molecular nanotechnology (MNT). From [2], used with permission.

1.1 Scope and Objectives

This paper develops early explorations of the capabilities MNT will provide in the area of space operations. It is an example of theoretical applied science [1], examining what sorts of applied capabilities are and are not feasible within our current understanding of physical law. The best example of theoretical applied science in the area of molecular nanotechnology is [1], which developed estimated technical performance parameters for molecular nanotechnology (table 1). This paper takes those parameters, and applies them to models of previously defined space system architectures. It also discusses a methodology for extending, deepening and refining out understanding of the actual capabilities molecular nanotechnology will provide, applicable to many fields.

Table 1. Estimates used in this research for the technical performance parameters of MNT are taken from [1].

Parameter Value Unit
Product Tensile Strength 5 x 10^10 Pa
Corresponding Material Density 3,510 kg/m3
Mechanosynthetic Device Operating Rate 10^6 Hz
Marginal Manufacturing Costs 0.1 - 0.5 $/kg

1.2 Main Advances

Most of the space system architectures considered here were not developed to particularly exploit the capabilities of MNT. In particular, one of the major benefits of MNT would be the novel function of a general purpose material processing capability. Since this is a novel function, however, it is not included in previously defined space system architectures.

The capability of MNT which previously defined space system architectures do take particular advantage of is the increased strength-to-density afforded by MNT structural components. The material properties assumed in the prior analyses of space system architectures, and the comparable MNT values, are shown in table 2 below.

Table 2. Values for material properties used in this research.

Material Strength (Pa) Density (kg/m3)
Aluminum (cold formed) 3.52 x 10^8 2,650
Steel (cold drawn) 1.240 x 10^9 7,800
Titanium (cold formed) 9.31 x 10^8 4,540
Graphite crystals 2.1 x 10^10 2,200
MNT Structural Material (Diamondoid) 5 x 10^10 3,510

The main conclusions of this research are that rockets can be made significantly more efficient, and vastly more cost efficient, the feasibility of orbital skyhooks is improved, interplanetary propulsion could have high performance, and very large inhabited space colonies could be structurally sound.

2. Methodology and Larger Goals

This research is meant to embed in a larger methodology, shown in figure 2 below. That methodology has the larger goals of progressively developing a better understanding of the future capabilities for space operations that will be provided by MNT, at the system and mission level, and of progressively developing an understanding of feasible and appropriate strategies for operating with this unfolding capability. The methodology can also help focus ongoing efforts in the theoretical applied science of MNT, by defining which uncertainties about MNT are operationally most significant.

While this larger methodology is expressed in terms of a focus on space operations, it could be useful in all application areas for MNT. The methodology is based on a hierarchy of technical capabilities, system architectures, operational concepts, and grand strategies.

MNT will make dramatically improved systems possible, and it thus is important to develop and analyze system concepts using MNT [4]. Such systems must be evaluated in their operating context. MNT will so change the total environment of capabilities and actions, however, that it will transform the operating environment context for doing the analysis. This "chicken-and-egg" problem requires an iterative analysis methodology to resolve.

Figure 2 describes such an iterative process, built around the hierarchy of technical capabilities, system architectures, operational concepts, and grand strategies. Each level provides building blocks for assembling concepts at the level above. Thus, one can think synthetically, combining building blocks into concepts at the next higher level. Also, one can think holistically, using an intuition or theory about what MNT can achieve at a particular level to develop a concept at that level. In this latter case, one must then do a top down analysis to make sure that the lower levels can actually support this new concept, resulting in either a refutation or added confidence in the concept. Another result of this top-down analysis likely to be additions to the menu of building blocks at the levels below.

Figure 2. An iterative methodology for assessing system architectures using MNT, likely to continue for years or decades by many researchers across many research projects.

A first step in this methodology is to take a broad range of previously defined system architectures, and apply the Technical Performance Parameters of MNT to those architectures. This will provide an initial set of MNT-based system architectures from which to launch further analysis. Such system architectures are likely to be unbalanced, however, and thus will probably perform significantly less well than system architectures designed to exploit the particular capabilities of MNT. The research is this paper is part of that first step.

3. Earth to Earth-Orbit Transportation

The standard concept for launching payload into Earth orbit is rockets, although different staging architectures are often considered. An alternative approach which might better take advantage of the material strength-to-density ratios offered by MNT are very long orbital cables, also called skyhooks and orbital towers. Both concepts are examined here.

3.1. Chemical Rockets

[5] provides a typical top-level comparison of six different architectures for transporting payloads from Earth using chemical rockets. This paper assumes LH/LOx engines with a 462 second specific impulse, and a 200 km circular reference orbit inclined 51deg.. The six architectures examined are:

  1. Single Stage to Orbit (SSTO) with Horizontal Take Off (HTO) using an undercarriage [SSTO(H)];
  2. SSTO with HTO using a sled [SSTO-SL];
  3. SSTO Air Launched from an aircraft at 8 km altitude, traveling at 180 m/s [SSTO-AL];
  4. SSTO with ground-launched Vertical Take Off (VTO) [SSTO(V)];
  5. Two Stage To Orbit (TSTO) using an undercarriage [TSTO(H)]; and,
  6. TSTO with ground-launched VTO [TSTO(V)].
All stages in all architectures use horizontal landing, and thus have wings and undercarriages. The wings and undercarriages are sized for the vehicle's return, unless they are used in take off, as for example by 1) SSTO(H). All architectures are sized to a common take-off mass (GLOW) of 500 tonnes, but their different architectures result in different payloads delivered to the reference orbit (Table 3).

The primary means of taking advantage of MNT without changing these architectures is to use diamondoid materials with much higher strength-to-density ratios. This will reduce the mass of the vehicles' fuselages, wings and tails, undercarriages, and propulsion systems, roughly in proportion to the relative strength-to-density of diamondoid to the materials used. (Many parts of the propulsion system, however, would require appropriately coated surfaces, and external surfaces will require thermal protection for re-entry.) This research assumes that the systems were all be built of titanium in the original paper. The nature of the "on-board equipment" is not clear described in [5]. It probably includes mass for life support for pilots of the vehicles. Since the exact nature of this mass is unclear, the conservative assumption made here is that this mass could not be reduced at all with the use of MNT.

By flying the same trajectories, the mass savings from applying MNT to the vehicle's structure can be applied to the payload. This occurs directly for the SSTO vehicles and the second stages of the TSTO vehicles. Mass savings on the first stage of the TSTO vehicles are added proportionately to all components of those vehicles second stages, including the payload.

The resulting vehicle and payload masses, and mass ratios, are shown in table 3 below. The "Titanium" based numbers (first entry) are from [5]. The "Diamondoid" based numbers are assuming MNT materials. These results are consistent with [6], which describes an MNT-based SSTO with a payload to GLOW ratio of 1/6 and a ratio of payload mass to dry, empty vehicle mass of ~8.

Table 3. Values for material properties used in this research. The difference from old to new is the use of MNT materials in the place of titanium.

Architecture Titanium/ Diamondoid "Dry, Empty Vehicle Mass" Payload Mass Mass Ratio "Payload to Dry, Empty Mass Ratio" Cost per kg to Reference orbit
1) SSTO(H) 92.5 / 16.5 -32 / 44 -6.4% / 8.79% -35% / 267% NA / $5.19 / $375
2) SSTO-SL 57.3 / 10.5 9.8 / 56.6 1.96% / 11.33% 17% / 541% $29k / $3.91 / $185
3) SSTO-AL 54.3 / 9.5 17.0 / 61.8 3.40% / 12.36% 31% / 653% $16k / $3.54 / $153
4) SSTO(V) 57.4 / 9.4 4.8 / 52.5 0.96% / 10.50% 8% / 540% $59k / $4.26 / $185
5) TSTO(H) 107.8 / 20.4 17.0 / 49.5 3.40% / 9.90% 16% / 243% $31k / $4.55 / $412
6) TSTO(V) 70.4 / 14.5 25.0 / 53.5 5.00% / 10.70% 36% / 368% $14k / $4.17 / $272

The costs for the titanium systems were not in [5], but are based on a very simple minded per mission cost estimate of $1000 per kg of dry, empty launch vehicle mass, averaged over the life of the program. This may be extremely optimistic; using traditional aerospace approaches it would require a tremendous number of fleet missions conducted with low overhead. The cost for the diamondoid, MNT-based systems is based on the high end of the MNT manufacturing cost estimate from [1], applied to producing the vehicle and the fuel mass, and assuming the vehicle is used once. The third cost entry applies the per mission cost estimate of $1000 per kg of dry, empty launch vehicle mass to the diamondoid system masses.

3.2. Skyhooks

A totally different architecture for placing mass into orbit is using very long, orbiting cables. The classic example is the synchronous skyhook [7], a cable in geosynchronous orbit reaching down to be anchored at the planet's surface, and extending upward so the entire structure is in tension. It is possible for a mass to climb this skyhook, gaining potential energy directly, and taking momentum from the orbital momentum of the planet. [8] discusses another version, free-orbiting skyhooks counter-rotating to just cancel out the orbit and planet's rotation as the tip makes its closest approach. Such a skyhook can grab a payload, and pull it into orbit.

Skyhooks are generally much too long to support themselves if they are of constant thickness. The answer is to use cables tapered with the load so that they experience constant stress over their length. The ratio of the area of a skyhook at distance r from the planet to the area of the skyhook tip at the surface of the planet is expressed in equation (5) from [8]:


Where A(r) is the area of the skyhook at radius r from the planet's center, A(rp) is the area of the skyhook at the planet's radius, rp is the planet's radius, d is the skyhook material's density, t is the tensile strength, wo is the orbital rate of the satellite, and ws is the rotation rate of the satellite. For a synchronous skyhook, ws = wo = wp (where wp is the planet's rotational rate). One particularly favorable concept is for a rotating skyhook roughly 1/3 the radius of the planet. This can be designed to make its closest approaches near exactly six stable points on the surface. It is also close to the optimum rotating skyhook length.

One measure of skyhook feasibility is the taper ratio, the ratio between the widest point in the skyhook and the tips. The taper ratio is proportionate to the exponent of the density to tensile strength ratio. [8] derived taper ratios, assuming graphite crystals and a substantial safety factor. Properly scaling those taper ratios based on the material properties achievable with MNT, gives improved feasibility, as shown on table 4 below.

Table 4. Improved feasibility of skyhooks, using MNT materials. Taper ratios when using graphite come from [8].
Body Taper Ratio for Synchronous Using Graphite Taper Ratio for Synchronous Using MNT Taper Ratio for 1/3 Planet Radius Using Graphite Taper Ratio for 1/3 Planet Radius Using MNT
Mercury 2.22 1.71 1.42 1.27
Venus 123 25.2 8.32 4.13
Earth 100 21.9 10.1 4.69
Moon 1.3 1.19 1.12 1.08
Mars 2.41 1.8 1.56 1.35
Jupiter 2.8 x 10^26 5.3 x 10^17 7.0 x 10^15 4.1 x 10^10
Saturn 3.3 x 10^6 2.3 x 10^4 17,430 695
Uranus 2,350 182 101 22.1
Neptune 1 x 10^6 1 x 10^4 1,092 168

4. Interplanetary Propulsion

Once a payload is in space, it is then eligible for interplanetary transport. The primary advantage of this is that the vehicle does not have to provide direct lift against the planet's gravity, so propulsion systems with thrusts much less than the vehicle's weights are quite acceptable.

4.1. Performance of Systems in the Literature

In this area, some concepts have already been sketched out which are intended to particularly exploit the capabilities of MNT. [6] examines lightsails, which are propelled by the light pressure from the sun on a large surface, and solar-powered ion engines, which turn solar energy into electrical power and use that power to accelerate ions to high exhaust velocities. (The accelerated ions are neutralized before release so that the vehicle does not build up a net charge.) As discussed, an unloaded sail 20 nm thick of aluminum will develop an acceleration of up to 0.16 m/s2, depending on its solar angle, at one AU from the sun. This is equivalent to a 0.08 m/s2 acceleration with a payload mass equal to the sail mass, or a 0.04 m/s2 acceleration with a payload mass equal to three times the sail mass. A major advantage of solar sails (built with or without MNT) is that they can provide this thrust continuously, without expending on-board reaction mass.

The ion engine discussed in [6] uses solar collectors with specific power on the order of nearly 105 W/kg to drive ion engines with a ~250,000 m/s ideal exhaust velocity, providing ~0.8 m/s2 acceleration. [2], suggests the specific power could be more than an order of magnitude greater, and a 1,000,000 m/s ideal exhaust velocity vehicle could provide ~9.8 m/s2 acceleration.

5. Space Colonies

Gerard K. O'Neill proposed housing citizens of a technological civilization in large space stations he called "colonies," each with up to 20 million people [9], [10], [11]. These O'Neill style colonies were large cylinders with spherical end-caps, rotating to provide a standard 1 g of pseudogravity on the inner surface. O'Neill suggested that if made of steel such a colony might reasonably be 3.2 km in radius, and 32 km long [10].

5.1 Maximum Radius of Space Colonies

The maximum radius of such an O'Neill style colony is limited by the hoop stress of the spinning structure, and the tensile strength to density ratio of the material. The formula is


Where R is the radius, g is the acceleration of pseudo-gravity at the rim, and G is the density. MNT offers a 5 x 1010 Pa tensile strength. Using the design rule of 50% safety factors for O'Neill style colonies [12], a 3.3 x 1010 Pa design tensile strength is reasonable. The associated material density is 3.51 103 kg/m3. One goal of the architecture is for g to equal 9.8 m/s2 [10],[12]. This all gives a possible space station radius of 9.6 x 105 m, or nearly 1000 km. For comparison, the corresponding feasible radius for titanium is 14 km, and even at its ultimate tensile strength with no safety factor, the titanium limit would be 23 km.

At the 9.6 x 105 m radius, the entire available strength (at the safety factor) of the MNT-based material is being used to prevent the rotating structure from bursting, and there is no strength left over to hold the space station's contents, including an atmosphere. To do so, a lower radius must be set.

5.2 Mass of an Example Design

One approach would be to set the design goal of a station radius such that the stressed skin structure would have a surface density of 5,000 kg/m2, which is a mass required in any case to provide shielding against radiation from space[12].

The mass of an O'Neill style colony can be first approximated as the sum of mass of the structure, the mass of the furnishings, and the mass of the atmosphere. For the design sketched here, these sum to 1.6 x 1017 kg., or 1.6 x 106 kg per person.

5.2.1 Mass and Sizing of the Structure

The mass of a stressed skin shell for an O'Neill style colony is derived [12] as:


Where s is the design stress, PA is the internal atmospheric pressure, G is the additional load for other contents, and a is the length to radius ratio of the cylinder. A typical a, used here, is 10. [12] derives a requirement for 50.8 kPa internal atmosphere, and use a G equal to 16% PA. Since the total area of the structure is:


One can directly solve for the structure radius where the shell is 5000 kg/m2. Using MNT materials, the structure will be 461 km in radius. For comparison, the equivalent number for titanium is 6.6 km, or for a titanium shell is at its ultimate tensile strength with no safety factor, 11 km.

A 461 km radius cylinder with endcaps, with a 4610 km long cylinder portion, has a surface area on that cylinder portion of 1.22 x 1012 m2., half of which would be available for habitation (the other half being taken up with windows). The [12] requirement of 87 m2 land area per person (including land for agriculture) yields a possible population for this structure of 76 billion people. The stressed shell masses 8.0 x 1016 kg of carbon.

5.2.2 Mass of the Furnishings

The furnishings are defined as all the mass on the inner surface of the colony, excluding the atmosphere. The G of 16% of PA (or 8,128 Pa) is the load of these furnishings. Since this load is providing a centripetal acceleration of 9.8 m/s2, the furnishings amount to 830 kg/m2, or 1.1 x 1016 kg for the entire colony. (Note, if one assumes that no such load is placed on the window areas, then the total mass for the furnishings is halved. This assumption is reasonable, but less conservative and not used here.)

5.2.3 Mass of the Atmosphere

More difficult to supply could be the mass for the atmosphere. The structure's volume is 3.5 x 1018 m3, which would require 2 x 1018 kg of atmosphere if evenly distributed, but it is not necessary to supply that full mass of atmosphere. The pseudogravity of the station's rotation will collect the bulk of the atmosphere close to the outer perimeter. >From [10] one can derive a simple model of atmospheric density of


where r is any radius of interest within the cylinder, r0 is the atmospheric density at the outer radius, and p0 is the atmospheric pressure at the outer radius. Integrating along r through the cylinder and spherical endcaps gives a total required mass of only 7.2 x 1016 kg, or roughly 3% of what would be required to fill the cylinder uniformly. Near 36 km above the surface, the atmosphere will fall to only 1% of the surface density. Thus, each square meter of surface requires a roughly comparable mass of atmosphere above it, whether that surface is in a very large open space station as described here, or on the surface of Earth with real gravity.

5.3 An Alternate Design

Note, however, that in the previous design the atmospheric density at the zero-g center of the colony is virtually zero. This does not support the secondary goal of allowing shirt-sleeve zero g activities [13]. Both [10] and [12] state lower pressure limits for manned habitation is 1/2 atmosphere. If one sets the design goal of 1 atmospheric pressure at the outer radius (101,350 Pa), and 1/2 atmosphere at the center, this gives a design radius of 11,700 m. This can be achieved with the 50% safety factor by titanium, rotating as an empty shell to simulate one g at the outer radius. An aluminum structure would have a 16 % safety factor, and the safety factor for steel for this structure would be 39 %. While all three materials could survive as empty rotating shells, they could not also hold the requisite atmosphere. The MNT material of diamond satisfies this application, including holding the atmosphere and the internal station furnishings, with a shell thickness of 4 cm. Adding the additional design goal of 5 kg/m2 thick structural material yields a 3,783 % safety factor for diamond.

6. Conclusion

The substantial increases in low level technical capabilities MNT will provide implies significant increases in space system capabilities. Example system architectures for Earth-to-Orbit and interplanetary transport, and large space colonies, have been examined.

The two most obvious extensions of this work would be to analyze the performance of other previously defined space system architectures when using the technical performance parameters of MNT, and to take these system architectures, and refine them to better take advantage of the estimated capabilities MNT will offer.

Further extensions should include designing example missions using estimated MNT space system capabilities and designing novel missions and systems to particularly exploit the technical capabilities of MNT. The ultimate goal should be to develop foresight on future capabilities for space operations adequate to make appropriate plans in the context of a developing molecular nanotechnology.


[1] Drexler K E 1992 Nanosystems: Molecular Machinery, Manufacturing, and Computation (New York: Wiley Interscience)

[2] Drexler K E 1995 Prospects in Nanotechnology: Toward Molecular Manufacturing ed M Krummenacker, J Lewis (New York: John Wiley & Sons) xvii-xviii

[3] Drexler K E 1986 Engines of Creation: The Coming Era of Nanotechnology (New York: Doubleday

[4] McKendree T L 1993 Systems Engineering in the Workplace (Proc. Third Annual Symposium National Council on Systems Engineering) ed J E McAuley, W H McCumber (Seattle: NCOSE) 417-424

[5] Shkadov L M, Denisov V Y, Lazarev V V, Plokhikh V P, Buzuluk V I, Volodin S V, Chervonenko K A, Skipenko V V, 1995 Acta Astronautica 35 47-54

[6] Drexler K E 1992 J. Brit. Interplanet. Soc. 45 401-405

[7] Pearson J, 1975 Acta Astronautica 2 785-799

[8] Moravec H 1977 J. Astronautical Sci. 25 307-322

[9] O'Neill G K 1974a Nature, 250 636

[10] O'Neill G K 1974b Phys. Today, September 32-40

[11] O'Neill G K 1975 Science, 190 943-947

[12] Johnson R D, Holbrow C, 1977 Space Settlements: A Design Study NASA SP-413 (Washington D.C.: NASA)

[13] O'Neill G K 1976 The High Frontier (Princeton: Space Studies Institute Press)