h a l f b a k e r yThe word "How?" springs to mind at this point.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
This idea is more of an experiment that may yield some interesting results.
Current chemical processing is dominated by steady state processes (constant continuous generation of one or two endproducts), or batch processing (Put x units in, get x units out). The paradigm is rigid and optimization
is usually a simple affair.
Now, imagine a symphony. Many insturments all sounding at different times, with different sounds to create a grand unifying sound. Imagine now, that instead of insturments, we use reactants. Instead of different sounds, we have varying concentrations, and varying properties (temp, pH, pressure). Now, devise a machine whereby you insert your reactants, and the machine forces the interaction in a sort of timed symphonic fashion. A is released in milisecond bursts while B slowly increases in concentration and then large burst of C, another burst of C! Stacatto A... fortissimo C! Okay, so you see the idea, but what's the point?
The hypothesis is that such timing may result in different products, perhaps more desired products. the intermediates of certain reactions may be selected based on proper timing. Also, the rate of yield may be increased due to nonintuitive results of such timing. I propose most of the timing would work on milisecond scales, when the chemical intermediates are all in quantum superpositions and such.
This idea may especially work for protein chemistry, enzymatic processes where there are many more intermediate states in the course of a reaction.
Inspiration for such an idea comes from computational genetics, where gene interplay is nothing short of symphonic, nor short of beyond our grasp, and Mozart, whose music makes me happy.
Imagine using the proper reactants and programming their release to some composition, getting a maximum yield of product A. Now, change the composition, and the yeild changes to product B. This paradigm would require much combinatorial chemistry, whereby all musical compositions would be tested for various sets of reactants, creating a database of yields for each combination.
GEB
http://www.forum2.org/tal/books/geb.html No, but very interesting... [daseva, Jun 01 2005]
Combinatorial Chemistry
http://www.netsci.o...chem/feature02.html [daseva, Jun 01 2005]
[link]
|
|
Somebody's been reading GEB again... |
|
|
//Imagine using the proper reactants
and programming their release to some
composition, getting a maximum yield
of protein A. Now, change the
composition, and the yeild changes to
protein B.// You mean regulation of
translation? Not sure how protein
synthesis fits into this... |
|
|
Would the whole rig look like a giant organ, or a cappucino machine? |
|
|
And thanks for the link, I didn't know there was a 20th anniversary edition of GEB. |
|
|
[basepair], I scurry with my words a bit, and did not mean to include proteins as products. Proteins would be used, instead, enzymatically in the machine. Proteins are hard to produce. Denatured protein synthesis is a slow stepwise process that probably wouldn't benefit from the machine I speak of. fermentation is also too slow. Imagine, instead, this machine capable of shooting amino acids out in some array, and then shooting out some ligases or something, and constructing a protein instantaneously, taking advantage of the precise timing and spatial accuracy of the machine. |
|
|
[Sot], giant organ sounds good. With independently moving flexible pipes, some big, some very small, capable of dispensing chemicals in any preprogrammed 3D array, at any millisecond interval, into a reactor. |
|
|
It's like deliberately injecting a bit of chaos into the experiment, I suppose. Forcing an accidental discovery, if you will. I like it. |
|
|
//Imagine, instead, this machine
capable of shooting amino acids out in
some array, and then shooting out
some ligases or something, and
constructing a protein instantaneously,
taking advantage of the precise timing
and spatial accuracy of the machine.//
Sounds like a harey way of making a
synthetic peptide. If you want to
generate a known a.a. sequence, you do
it either chemically by stepwise
synthesis or, for longer peptides/
proteins, you make the relevant DNA
and let bacteria (or an in vitro
transcription/translation system) make
it for you. If you want randomness, you
can again do this in vitro or make
randomized DNA libraries and put them
into bugs.
However, this is
just a side-niggle, and doesn't relate to
your main idea of precise temporal
control of conventional chemical
syntheses. It is interesting to think
about, particularly if you had two
different irreversible reactions which
compete with eachother for reagents.
Perhaps an example would be some
polymerisations which can form
alternative (say, branched or linear)
polymers, with the balance between the
outcomes being determined by local
concentrations of monomers,
catalysts...
So, a bun for an
interesting idea, though I'm not
completely convinced that the current
chemical industry hasn't explored some
aspects of controlling kinetics along
these lines (maybe?). |
|
|
[Basepair], I follow you. If the machine were to be used for protien synthesis it would use entirely new methods aside from stepwise polymerization or GM bug fermentation. These methods are good but have their pitfalls, as annoed above. |
|
|
Indeed, controlling the layout of branched and linear polymers would surely be a profitable feat. |
|
|
The chemical industry is notoriously "mature". The most interesting thing they've done recently is develop combinatorial chemistry (link), and take strides in modeling proteins. Kinetics seems to be on the backburner. |
|
|
I'm still not getting your method of
protein synthesis or why it would differ
from what's done at present. But
maybe I'm missing a point. |
|
|
Mozart makes you happy, eh? You must be a lover of math and logical progressions. |
|
|
I have no idea what you're all babbling about. + |
|
|
More of a "Mach". <Nigel Tufnell> "I call this 'Lick my Love Pump'"</NT> |
|
|
The Iommi21 multiplexer reads subharmonics in a range from latent molecular static states to twin free radical repulsion transition states, but turn up the volume and there is unmistakable power to affect any reaction face wave design! |
|
|
From what I understand so far, this thing is a big metal box that injects various reagents in some sort of sequence, repeating thousands of times per second. |
|
|
What I'm wondering is how your machine would overcome the random effects of diffusion, because the path each molecule takes to smack into a target will be different, right? And that would augment the timing for each individual molecule. |
|
|
Perhaps the placement of the the chemical injection sites will also come into play? I'm imagining a large box, housing a matrix of millimeter-diameter tubes, with reactants ejected at certain nodes, so as to make the injection more uniform throughout the entire solution. |
|
|
Fluidics - a stupid name for the fluid flow analogy to electronics is capable of analogue and digital (logical) operations albeit much slower than electronics. It is being developed for very small scale testing and self-controlling analyses. I also read somewhere someone's idea of advancing on digital fluidic computing by using chemical kinetics as analogue operations and chemical products as memory or actual end products (manufacturing). Probably not worth necro-ing this thread for but I've done it now; best just not to think about it. |
|
|
Thanks [sneaky], I still love this one. Very vague, but I imagined it too often not to post. Where's [Sot]? |
|
| |