© 1999 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999

6.3.4.1 Human Chemical Energy Resources

The first step in evaluating chemical power transduction alternatives for medical nanorobots is to assess the chemical energy resources readily available within the human body. A complete inventory is beyond the scope of this book, but a few broad conclusions may be drawn.

Table 6.4 summarizes major representative chemical energy resources available within typical individual human tissue cells, within the human blood volume, and within an entire 70 kg adult human body. Burning a protein or carbohydrate in oxygen releases ~4.1 Kcal/gm (17 x 10⁶ joules/kg); burning lipids in oxygen releases ~9.3 Kcal/gm (39 x 10⁶ joules/kg). In theory, proteins represent the most plentiful energy resource in blood and cells, but extensively accessing this source may require disassembly of essential cellular structures, considerable preprocessing (e.g., denaturation and lysis), and disposal of nitrogen-containing waste products.* Free amino acids and short peptides are too dilute to be of much value. ATP, usually available only inside cells with a high turnover rate, is also of very limited utility (though it is fairly stable -- one can buy bottles of powdered ATP and store it for years in the freezer).

* The human body secretes urea because it lacks the necessary enzymes to further process the nitrogen. Nanorobots could make use of bacterial enzymatic or other chemical processes to obtain energy from nitrogenous wastes -- for example, by exploiting the ammonia oxidation reaction: 4NH₃ (g) + 3O₂ (g) ----> 2N₂ (g) + 6H₂O (g) + 2105 zJ.

Fats potentially offer the most plentiful natural energy source in the human body, providing over 80% of body heat in the absence of carbohydrates. Unfortunately, high energy density serum lipids such as cholesterol are not freely available but instead are present only in three bound forms that ensure solubility:

1. chylomicrons and other plasma lipoprotein carriers (proteincoated lipid droplets <0.5 microns in diameter), to transport lipids throughout the body from the intestine after absorption of dietary fat, or from the liver after lipid synthesis to storage in adipose tissue or for utilization elsewhere in the body;³⁵²⁵

2. fatty acids bound to serum albumin, to transport fat from adipose tissue to elsewhere in the body; and

3. ketone bodies (acetoacetate and beta-hydroxybutyrate) to transport lipids processed by or synthesized in the liver.

Anhydrous lipid droplets 0.25 microns in diameter are often found in the cytosol. Excepting operations inside adipocytes (wherein triglyceride droplets occupy nearly the entire cytosol), fat utilization by medical nanomachines would probably require some physical nanopipetting or other preprocessing to dislodge the lipid from its carrier, and different energy extraction pathways may be required depending on whether the fatty molecules thus liberated are saturated or unsaturated, etc.

Carbohydrates are soluble in water and thus exist in free form throughout the body. Large individual molecules of glycogen (10-40 nm in diameter) float freely as granules in the cytosol, usually coated with a ~5 nm thick monolayer of digestive enzymes (Section 8.5.3.7). Glucose is the preferred human body fuel -- a nearly constant bloodstream inventory of ~5 grams is maintained homeostatically from a buffer supply of ~350 grams of glycogen stored mainly in the liver and muscles. Glucose transporters maintain^3649-3654 cytosolic concentrations at comparable levels to the blood. Nerve cells can only metabolize glucose and ketones; sperm cells are surrounded by a fructose bath, from which they absorb chemical energy to power flagellar engines during their short-lived locomotion (Section 6.3.4.2).

Because of its abundance, high energy density, and relative ease of use (e.g., high solubility), glucose is probably the ideal fuel for most in vivo nanomedical applications. A 1 micron³ onboard store of glucose fuel powers a 10 pW nanorobot for 2400 sec of operation without further refueling. The equilibrium free glucose mass in a typical human tissue cell provides ~10⁴ sec of power for a 10 pW nanodevice; during strenuous exercise, a typical tissue cell can normally replace its entire glucose store in <10³ sec via membrane glucose transporters, providing a continuous fuel supply for at least ~100 pW of nanorobot demand.* If this is still insufficient for the intended application, artificial lipophilic oxyglucose transporter structures may be inserted through the cellular membrane to provide supplementary power levels up to the diffusion limit (Section 6.5.3). (An additional 100-8000 nanojoules of polymerized glucose, present as glycogen granules, may be available in some cells.) For a (20 micron)³ tissue cell, from Eqn. 3.4 the maximum diffusion-limited glucose current is J_glu = 6-30 x 10¹⁰ glucose molecules/sec. Assuming ~50% absorption efficiency at the cell surface (Section 4.2.5), ~50% energy conversion efficiency (see below), and that the local extracellular space has sufficient fluid volume to supply adjacent cells at peak rates, then the maximum theoretical total continuous power draw in cyto is 70,000-300,000 pW for foraging glucose-consuming nanodevices with unlimited access to oxygen (e.g., onboard storage). If oxygen supplies were restricted to that which could passively diffuse through cell walls, the comparable in cyto diffusion-limited total oxygen current would be J_O2 = 2 x 10¹⁰ O₂ molecules/sec, a ~4000 pW power limit. (For comparison, the basal power consumption of a typical 20-micron human cell is ~30 pW; Table 6.8.)

* If 624 micromoles/min of glucose are transported per milligram of purified glucose transporter protein molecules of molecular weight ~100,000 daltons,³⁶⁵⁶ then the mean glucose transport rate is ~1000 glucose molecules/sec-transporter which is equivalent to ~0.0025 pW/transporter assuming 50% metabolic utilization efficiency, or ~250 pW/cell for a 20-micron tissue cell embedded with ~10⁵ transmembrane glucose transporters (membrane surface density ~40 transporters/micron²).

The following Sections include brief technical sketches of several possible approaches to chemoergic power transduction. The methods outlined are rather crude. An ideal design might involve complex nanomachinery that has yet to be designed, which may perhaps more closely resemble in efficiency and function the enzymatic and "proton pump" biological nanomachines that are found in nature, or the natural chemophotonic transducers employed in bioluminescence.³⁵⁶⁴

Last updated on 18 February 2003