Re: Recycling, Rosennean Style...

[Leo Caves <***>]

Judith wrote:
> What I'm thinking is that there should be innumerable ways to take any 
> dangerous or toxic molecule and figure out a series of steps which 
> reconfigure it, without needing added energy.

You would think so!

(Conventional) Synthetic organic chemistry is a world apart (well, from 
mine).   Known chemical reactions (transformations) are catalogued; 
certain general principles of reactivity in molecules can be inferred; 
certain predictive frameworks exist for rather specialised subclasses 
of reactions. Otherwise, it appears to be as much an art as science - 
with much information encoded in the neural nets of the chemists 
themselves.  Reaction databases do exist, and there are tools to aid in 
"retrosynthetic analysis" - exploiting routes to synthesis via known 
elementary steps.

The key issues in this field area are finding viable routes to 
synthesis (the complement of what you suggest), which give you what you 
want (selectivity, specificity), in useful quantities (yields can be 
low and compound in multistep processes), in reasonable conditions 
(solvents, temperatures, catalysts).

This area is not a closed book, it is characterised by a lot of trial 
and error and is inhabited by the specialist/artisan -it is not 
engineering.

Biomimetic or bio-organic chemistry (towards biochemistry) takes 
increasing cues from biological systems in mimicking or exploiting 
biocatalysis (the agents of which are enzymes - usually, but not always 
proteins).

> Furthermore, because molecular organization is not complex, we can and 
> do model it in enough detail to accurately predict the behaviors and 
> outcomes of various interactions, so there should be "no surprises" of 
> the sort that genetic engineering and use of living systems is liable 
> to generate. 

My research work has dealt mainly with the computer simulation of 
protein dynamics at the single molecule level.  Proteins are many-body, 
dynamical systems, with non-linear interactions.   They have intricate 
hierarchies of structure and dynamics (power-laws abound), exhibiting 
classic traits of complex systems such as self-organisation and 
emergence (protein folding) and criticality (phase transitions).   We 
can't design an enzyme ab initio.  We can take cues from nature (our 
increasing databases of observed protein sequences and structures) and 
make reasonable inferential models in some cases for structure, but 
reactivity is higher order property.

You can get "surprises" even at this level.   The "prion" theory of 
spongiform encephalopathies attributes the initiation of amyloid 
fibrils (aggregates of proteins, forming the "spongey" stuff) to a 
misfolding of the prion proteins:  the same amino-acid sequence, 
responding to changes in environment, resulting in different structure 
and interactions, with unfortunate results.

>  Where would I take this idea from here, in practical terms?

That's a question!  Obviously, we see "responsible" (communities) 
companies look at the a more complete life-cycle of their products 
(processes) - beginning with the end in mind.  However, what I think 
what you are striving it is to apply these principles more widely and 
deeply, and developing a whole system's approach to this issue.
I would imagine in practical terms there is more to be done in raising 
awareness of "total system accounting" than in any underlying 
technology or science.   Does this make me jaded or cynical?  I hope 
not:  its how I respond in "practical" mode!

You can see I am free-wheeling a little on this one...

Leo