Diffusion transport

From apm
Jump to: navigation, search

Myth's

Myth: "Transport at the nanoscale can only be done via diffusion transport"

The lack of fully guided transport in biological system does not imply their impossibility.
The lack of fully guided transport in biological systems instead shows a limitation of biological systems and natural evolution.

Myth: "Diffusion transport has superior efficiency to non-stochastic fully guided transport (e.g. on rails)"

While diffusion transport takes no energy for the motion of transported molecules the necessary energy expenditure for the capture of transported molecules from the solution and subsequent permanent fixture to the target must not be overlooked.

Why diffusion transport is slow

Inherent slowness

Diffusion transport speed is distance dependent. The farther the transport distance the slower it gets. On average the traveled distance grows with the square-root of time.

Slowness from lack of geometric constraints

Since fully unconstrained diffusion may transport molecules in any direction the situation is even worse. The surface area of a sphere with origin at the starting point and surface crossing the target grows with the square of the traveling distance while the target area that must be hit stays constant.

Speeding up diffusion transport

Improving speed by improving geometry

Diffusion on a lower dimensional subspace (in biosystems e.g. on a lipid membrane) can improve the geometric part of the problem. That is diffusion transport can be sped up by reducing the dimensionality of freedom of motion and or restricting the range of motion to regions.

Improving speed by turning away from pure diffusion based transport

In biology motors work often by prevention of back reactions.

If taken to the extreme one ends up with a one dimensional path with unidirectional erratic motion with unpredictable times for forward steps.

By coupling many of these together in the background one ends up with predictable and continuous forward motion. This is exactly deterministic fully guided transport. The system is now all the way in machine phase.

So when successively improving on diffusion transport one naturally ends up with fully guided transport at the very end.

Increasing speed by raising temperature

By raising the temperature one can speed up diffusion transport. It should be fairly obvious though why this can't be stretched very far. Systems relying on solvents usually have rather stringent constraints on temperature. The boiling point of the solvent (which can vary a bit dependent on pressure) must not be exceeded.

In any system the temperature is limited by what the most delicate transported molecules can survive. See: Consistent design for external limiting factors

When temperature rises in solvent based systems the chemical interactions between transported molecules (mutually) and solvent can become problematic. In a fully guided system where molecules are transported in a practically perfect vacuum one only needs to care about the (usually much less critical) internal thermal stability and not about the (usually much more critical) external chemical stability.

A two by two matrix for detangling the prevalent confusion of ideas

diffusion transport as (exclusive) far term goal (bad)

Exclusively sticking with diffusion transport for ones far term goals and complete shunning of system concepts featuring fully guided transport is detrimental. This often happens due to strong belief in the aforementioned myths. The consequence can be misguided development choices and vast misjudgment of the potential capabilities of future technology. This is essentially an exclusive focus on the brownian technology path).

fully guided transport as (main) far term goal

Fully guided transport is not about erroneously fighting thermal motion instead of using it. It's about sufficiently controlling thermal motion such that one can exploit the advantages of fully digital systems (error margins & error corrections).

Diffusion transport has several fundamental limitations compared to non-stochastic/deterministic fully guided transport including the ones listed above. By switching over one can leave all this limitations behind.

This is essentially a focus on the far term goal of gem-gum factories.

fully guided transport as (exclusive) near term R&D focus (bad)

Trying to move to fully guided transport right away may though be too difficult and less productive.
See: Pathway controversy

diffusion transport as (main) near term R&D focus

In the near term exploiting diffusion transport is very beneficial due to lack of options (bootstrapping problem).
But since diffusion transport is clearly inferior to fully guided transport moving over ASAP is desirable.

Still profitable near term applications that create progress merely accidentally (like molecular biology based medicine) should be kept the only forefronts of development. After all their focus can anytime quickly and permanently veer off sensible far term goals. Even in a way that strongly hiders further progress to those goals.

Related

External links