| Space-filling fractal surfaces play a fundamental role in how organisms function at various levels and in how structure determines function at different levels. In this thesis, we develop a quantitative theory of oxygen transport to and across the surface of the highly branched, space-filling system of alveoli, the fundamental gas exchange unit (acinar airways), in the human lung. Oxygen transport in the acinar airways is by diffusion, and we treat the two steps---diffusion through the branched airways, and transfer across the alveolar membranes---as a stationary diffusion-reaction problem, taking into account that there may be steep concentration gradients between the entrance and remote alveoli (screening). We develop a renormalization treatment of this screening effect and derive an analytic formula for the oxygen current across the cumulative alveolar membrane surface, modeled as a fractal, space-filling surface. The formula predicts the current from a minimum of morphological data of the acinus and appropriate values of the transport parameters, through a number of power laws (scaling laws). We find that the lung at rest operates near the borderline between partial screening and no screening; that it switches to no screening under exercise; and that the computed currents agree with measured values within experimental uncertainties. From an analysis of the computed current as a function of membrane permeability, we find that the space-filling structure of the gas exchanger is simultaneously optimal with respect to five criteria. The exchanger (i) generates a maximum oxygen current at minimum permeability; (ii) 'wastes' a minimum of surface area; (iii) maintains a minimum residence time of oxygen in the acinar airways; (iv) has a maximum fault tolerance to loss of permeability; and (v) generates a maximum current increase when switching from rest to exercise. |
