Can Software Engineering Contribute to Solving Environmental Chemistry Problems?

[MikeFromOhio]

I am a software engineer of 20 years and would love to use my skills to somehow help work on the Carbon Dioxide problem (or other problems that could help the environment).

I was wondering if simulation software might be written that would allow one to model various processes that attempt to transform CO2 into something not so harmful for the environment.

I was thinking of writing an extendable simulator package where one could add new models (of atoms and/or elements and/or compounds). The software would also have RULES about how such elements would combine or react. Based on such rules, the software could then try out various combinations of LEFT-SIDEs involving CO2 and other element-models. The rules would then be evaluated to provide the RIGHT-SIDE of the equation for each left-side combination.

Maybe such modeling software already exists?
Is it possible to capture the rules mentioned above?
Are there other ways software is being used to help solve such problems?

 

[alxm]

Well, it's nice you'd like to help the environment. But actively removing CO2 from the atmosphere is not the way to solve global warming. Basic thermodynamics dictates that this would require more energy than we got from producing the CO2 in the first place, rendering it pointless as long as we're using fossil carbon as fuel.

That said, yes, software is used in chemistry. Computational chemistry is an entire field of its own and has been around since before digital computers, even. Douglas Hartree, a significant name in the field, once did his calculations on a mechanical integrator he'd built out of Meccano(!). And at practically every point in time since the computer was invented, a substantial portion of supercomputing has been devoted to chemistry or chemical physics calculations.

I think the adage "It is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail." applies here. You assume that the properties of chemical reactions can be formulated in terms of discrete logical 'rules'. That's an assumption based on what would be convenient for a programmer, not on physical theory. The correct approach, is to start with the physical theory and (hopefully) work the problem into a mathematical form that a computer can solve, such as an eigenvalue problem or the minimization of a function.

If you look at some computational chemistry textbooks, you'll also find that reflected; they don't typically include much code, if any, and little on algorithms. The main problem is to derive a soluble problem. That's what the field is about, that's what the limiting factors are.

You don't need to be a master programmer to implement an algorithm for calculating matrix eigenvalues. But it takes grad-level quantum mechanics to understand the math and approximations made in getting from the molecular Schrödinger equation to a matrix eigenvalue problem that can be solved to get the energy of a molecule.

 

[MikeFromOhio]

First off, thanks for the reply.

>>But actively removing CO2 from the atmosphere is not the way to solve global warming.
>>Basic thermodynamics dictates that this would require more energy than we got from >>producing the CO2 in the first place, rendering it pointless

So how do plants/photosynthesis tie into what your saying?
And the various upstart companies (like Joule) using Photosynthesis-like processes.
So there's at least one way to split CO2 that is practical, no?


>>You assume that the properties of chemical reactions can be formulated in terms of
>>discrete logical 'rules'. That's an assumption based on what would be convenient for
>>a programmer, not on physical theory.

Actually, it was an assumption based on my chemistry textbook, but I am not disagreeing. It appears to a non-expert like myself that valence electrons, or lack of them in the outer shell tells us much about what wants to combine with what in both ionic and covalent bonding. So that might be a good starting point for formulating "combining rules".

However, what about "splitting rules"? What can split a stable entity like CO2? Nature uses sunlight to split water, and then uses those high energy hydrogen atoms to split CO2. What other atoms/energy could also be used to split CO2 and are there general principles/rules involved?

Again, thanks for the response.

 

[Borek]

So how do plants/photosynthesis tie into what your saying?

They use a lot of energy in the process, using light. And somehow I doubt we could be able to design much more efficient process doing the same thing, after all, plants have been optimizing their approach for billions of years.

Actually, it was an assumption based on my chemistry textbook, but I am not disagreeing. It appears to a non-expert like myself that valence electrons, or lack of them in the outer shell tells us much about what wants to combine with what in both ionic and covalent bonding. So that might be a good starting point for formulating "combining rules".

At the very basic level it works, but reality is much more complicated. Please read page here and note that text there refers only to relatively simple inorganic systems of already well known chemistry. Even then predictions are quite difficult.

Best we can do is to approach the system using Schroedinger equation and quantum chemistry methods, but even then we are limited in accuracy of our predictions - and not because of the lack of effort! But alxm knows much more than me on this particular subject, so I will stop here.

--

 

[Acut]

@MikeFromOhio
Contrary to popular belief, plants do not use CO2 to produce oxygen (O2). The oxygen comes from water molecules.

Splitting CO2 is not very practical. As someone has already pointed out, it would spend a lot of energy. Plants use sunlight to break water molecules, and CO2 is absorbed to form glucose.

 

[MikeFromOhio]

Guys -- thanks for the link and feedback.

 

[alxm]

So how do plants/photosynthesis tie into what your saying?
And the various upstart companies (like Joule) using Photosynthesis-like processes.
So there's at least one way to split CO2 that is practical, no?

Sure, but it doesn't use any more or less energy. You're basically talking about a different form of solar energy, and producing fuel directly instead of electricity. Photosynthesis isn't actually very efficient. The benefit is that it's cheap to do in bulk - solar panels don't build themselves, but living things do.

Actually, it was an assumption based on my chemistry textbook, but I am not disagreeing. It appears to a non-expert like myself that valence electrons, or lack of them in the outer shell tells us much about what wants to combine with what in both ionic and covalent bonding. So that might be a good starting point for formulating "combining rules".

Well, those are the 'rules' you learn in introductory chem. Lewis structures, 'filling shells' etc. They have a limited range of validity, and also a limited range of what they predict. Basically they don't tell you much more than a chemistry ball-and-stick model set does. Which is actually quite a bit, but the thing is, every chemist knows all this very well. Coming up with possible reactant/product combinations isn't a chemical problem of much dignity. The problem isn't in determining which compounds can be formed, but which ones are formed, and how fast they are formed and such. In other words, quantitative properties such as heats of formation and reaction rates.

The theory covered in basic textbook chemistry doesn't tell us anything about how to arrive at those numbers. It just focuses on teaching the concepts behind them. And these are idealized concepts. There's no binary distinction between covalent bonds and ionic bonds, for instance. It's a continuum. Metal atoms don't actually have integer redox states.

However, what about "splitting rules"? What can split a stable entity like CO2? Nature uses sunlight to split water, and then uses those high energy hydrogen atoms to split CO2. What other atoms/energy could also be used to split CO2 and are there general principles/rules involved?

Well, just look at Photosystem II (a member of my old research group got his thesis on theoretical studies of it; I can send you a copy if you want). It's an incredibly complicated mechanism that can't be described by anything less than explicit quantum mechanics.

But basically what you're asking is like asking "what rules are there for constructing a working clock?". There aren't rules, there's just the fundamental physics and whatever way you can figure out to exploit it to make it do what you want. But with chemistry, the fundamental physics is a lot more difficult. At least clocks are classical.