Because I'm irked by people denying basic science, and frankly I'm stressed about something else and looking to vent, I thought an examination of the process of crystallization would be useful. Please understand that there is no way for me to fully address this in a forum post; this is such a complex issue that we still don't understand it fully, and folks are pouring millions into it because whoever works out the details in full will revolutionize human life. Everything I say should be taken as the preamble to an introduction to this topic. But I do want to address some of the basic issues, from a conceptual standpoint. I'm not going to get into the math, in part because this forum doesn't make it easy to do so and in part because I'm not interested in re-learning it at this point in time; got too many other irons in the fire to add this one.
_________________________________________________
How do you make crystals? There are two main ways: precipitating out of a melt, or precipitation out of a solution. However, the thermodynamics and kenetics are identical between these two options; in a very real sense, water at the freezing point can be considered to merely be water saturated with ice. So there's no real point in differentiating between the two, and most discussions of crystalization that I've encountered don't. There are other ways (solid-state reactions, precipitation from a gas, and a few others), but again, thermodynamically and kenetically we can treat them as identical. For the purpose of this discussion, I'll use water crystallizing from a melt (ie, freezing) because it's a system that we're all familiar with.
You can precipitate crystals out of a melt by manipulating the temperature, the pressure, or both--there are diagrams, called phase diagrams, which can show you exactly how to do it. You can see a few here. We call the area shown in this diagram P/T space, for pressure and temperature. To freeze a melt, you have to move the melt in P/T space from liquid to solid. Easiest way? Cool it.
That said, cooling alone is demonstrably insufficient. I have an ex-girlfriend who had a bottle of water in her car once in an Ohio winter. It was liquid when she picked it up, but solid by the time she tiltied it to drink from it. We call such states "supercooled"--the water is supersaturated with ice. From our grade-school introduction to thermodynamics and phase changes, this makes no sense; it's below the freezing point, it should become a solid!
The issue is, ice is (in that part of P/T space) a lower-energy state of matter. Which doesn't seem like a problem, but it is. Nucleation actually happens all the time in supercooled water; it may happen in warmer water too, I don't know. But the process of crystallizatioon is exothermic--a crystal is a lower-energy state, so to go from liquid to solid the system has to give off energy. Where does a three-molecule bit of ice floating in water give off energy? There is only one place: to the water around it. This creates a thin--VERY thin--layer of water that's just warm enough to melt the nucleated ice crystal. (In super-saturated systems, the analogous process is that the crystal forms a layer of unsaturated water around it, re-disolving the crystal).
So why do we see ice at all? Energy--a small amount--is added to the system--at the right time. This is called "activiation energy", and is pretty much universally necessary for any chemical reaction (remember, crystalization--and any phase change--is a chemical reaction). This energy can come from literally anything. If the baby crystals can be physically moved out of that layer of warmer water, they will encounter water that's supercooled and which will add itself to the crystal, to give one example. It's kinetic, not thermal, energy. A seed crystal is also an addition of energy--crystal faces are lattice defects, and in many cases have very small ionic charges (the crystal as a whole is essentially neutral, but a face may have all cations or all anions exposed, giving it a local charge). That electromagnetic energy is sufficient to overcome this energy barrier. Moving that thin film of water away from the crystals--such as stirring the melt gently, which is how they make clear ice--will do it as well; again, this is an addition of kinetic energy.
If there is a large degree of super-cooling, you can sometimes get crystals that, once started, grow so fast that they grow past this layer of warmed liquid. In those cases, you get dendritic crystals--crystals that are perfectly orderly on the molecular level, but completely chaotic (they have made nonlinear equations to reproduce them) at the macroscopic scale. If there's only a little supercooling, you get hopper crystals--because the corners are where crystals grow the most, since that's where there's free space to add new unit cells. You end up with crystals with hollow basket-shaped faces, or which are just long spears fo corners with just enough of each crystal face to hold the corners together (I've seen salt crystals in an X form, for example). To get normal crystals requires relatively slow growth, often with little supercooling--but this is a very complex issue, particularly with complex melts such as magma, which change composition as they cool. You can get huge crystals in cooling magma very fast due to shifts in water content, thanks to fractional distillation.
So does melting require activation energy? Sure. ALL chemical reactions require some amount of activation energy. The thing with melting is, you're already adding energy to the system, in copious amounts, so the activation energy is merely a small portion of the energy of the system. This makes melts tremendously easier to control than freezing--which is why most phase diagrams don't actually show the temperature where something freezes, but rather the temperature at which it melts. In theory, the freezing/melting points are identical; in practice, there are sufficient confounding factors to make freezing an unreliable indicator of the freezing/melting point.
Again, this is a preamble to an introduction to this topic. There is a LOT of math involved--it starts with entropy, entrophy, enthalpy, and Gibbs free energy, and then gets into the complex stuff! No purely conceptual examination of the system will be sufficient. And there's simply no way to adequately address this in a forum post. And finally, we frankly don't understand the process entirely. The mastery of this process by organisms is what gave us the Cambrian Explosion--organisms figured out how to control crystal growth to create skeletons, and suddenly we had a fossil record worth discussing. Humans figuring this process out would be a type of singularity, akin to what happened when we mastered artificial refrigeration or fire; we have no concept of what life would be like with this technology. So understand that literally every phrase in the previous paragraphs is the summary of a tremendous amount of scientific literature.
If you want to delve further into this, I suggest Nesse's "Introduction to Mineralogy"; pretty much every geologist I know has (or had at one time) a copy of it. If you REALLY want to know more about it, get a copy of that book and dig into the literature. It's facinating stuff, if you can wrap your head around it!
_________________________________________________
How do you make crystals? There are two main ways: precipitating out of a melt, or precipitation out of a solution. However, the thermodynamics and kenetics are identical between these two options; in a very real sense, water at the freezing point can be considered to merely be water saturated with ice. So there's no real point in differentiating between the two, and most discussions of crystalization that I've encountered don't. There are other ways (solid-state reactions, precipitation from a gas, and a few others), but again, thermodynamically and kenetically we can treat them as identical. For the purpose of this discussion, I'll use water crystallizing from a melt (ie, freezing) because it's a system that we're all familiar with.
You can precipitate crystals out of a melt by manipulating the temperature, the pressure, or both--there are diagrams, called phase diagrams, which can show you exactly how to do it. You can see a few here. We call the area shown in this diagram P/T space, for pressure and temperature. To freeze a melt, you have to move the melt in P/T space from liquid to solid. Easiest way? Cool it.
That said, cooling alone is demonstrably insufficient. I have an ex-girlfriend who had a bottle of water in her car once in an Ohio winter. It was liquid when she picked it up, but solid by the time she tiltied it to drink from it. We call such states "supercooled"--the water is supersaturated with ice. From our grade-school introduction to thermodynamics and phase changes, this makes no sense; it's below the freezing point, it should become a solid!
The issue is, ice is (in that part of P/T space) a lower-energy state of matter. Which doesn't seem like a problem, but it is. Nucleation actually happens all the time in supercooled water; it may happen in warmer water too, I don't know. But the process of crystallizatioon is exothermic--a crystal is a lower-energy state, so to go from liquid to solid the system has to give off energy. Where does a three-molecule bit of ice floating in water give off energy? There is only one place: to the water around it. This creates a thin--VERY thin--layer of water that's just warm enough to melt the nucleated ice crystal. (In super-saturated systems, the analogous process is that the crystal forms a layer of unsaturated water around it, re-disolving the crystal).
So why do we see ice at all? Energy--a small amount--is added to the system--at the right time. This is called "activiation energy", and is pretty much universally necessary for any chemical reaction (remember, crystalization--and any phase change--is a chemical reaction). This energy can come from literally anything. If the baby crystals can be physically moved out of that layer of warmer water, they will encounter water that's supercooled and which will add itself to the crystal, to give one example. It's kinetic, not thermal, energy. A seed crystal is also an addition of energy--crystal faces are lattice defects, and in many cases have very small ionic charges (the crystal as a whole is essentially neutral, but a face may have all cations or all anions exposed, giving it a local charge). That electromagnetic energy is sufficient to overcome this energy barrier. Moving that thin film of water away from the crystals--such as stirring the melt gently, which is how they make clear ice--will do it as well; again, this is an addition of kinetic energy.
If there is a large degree of super-cooling, you can sometimes get crystals that, once started, grow so fast that they grow past this layer of warmed liquid. In those cases, you get dendritic crystals--crystals that are perfectly orderly on the molecular level, but completely chaotic (they have made nonlinear equations to reproduce them) at the macroscopic scale. If there's only a little supercooling, you get hopper crystals--because the corners are where crystals grow the most, since that's where there's free space to add new unit cells. You end up with crystals with hollow basket-shaped faces, or which are just long spears fo corners with just enough of each crystal face to hold the corners together (I've seen salt crystals in an X form, for example). To get normal crystals requires relatively slow growth, often with little supercooling--but this is a very complex issue, particularly with complex melts such as magma, which change composition as they cool. You can get huge crystals in cooling magma very fast due to shifts in water content, thanks to fractional distillation.
So does melting require activation energy? Sure. ALL chemical reactions require some amount of activation energy. The thing with melting is, you're already adding energy to the system, in copious amounts, so the activation energy is merely a small portion of the energy of the system. This makes melts tremendously easier to control than freezing--which is why most phase diagrams don't actually show the temperature where something freezes, but rather the temperature at which it melts. In theory, the freezing/melting points are identical; in practice, there are sufficient confounding factors to make freezing an unreliable indicator of the freezing/melting point.
Again, this is a preamble to an introduction to this topic. There is a LOT of math involved--it starts with entropy, entrophy, enthalpy, and Gibbs free energy, and then gets into the complex stuff! No purely conceptual examination of the system will be sufficient. And there's simply no way to adequately address this in a forum post. And finally, we frankly don't understand the process entirely. The mastery of this process by organisms is what gave us the Cambrian Explosion--organisms figured out how to control crystal growth to create skeletons, and suddenly we had a fossil record worth discussing. Humans figuring this process out would be a type of singularity, akin to what happened when we mastered artificial refrigeration or fire; we have no concept of what life would be like with this technology. So understand that literally every phrase in the previous paragraphs is the summary of a tremendous amount of scientific literature.
If you want to delve further into this, I suggest Nesse's "Introduction to Mineralogy"; pretty much every geologist I know has (or had at one time) a copy of it. If you REALLY want to know more about it, get a copy of that book and dig into the literature. It's facinating stuff, if you can wrap your head around it!
via International Skeptics Forum http://ift.tt/1C582eR
Aucun commentaire:
Enregistrer un commentaire