User blog:Cerne/Geology recap: core composition

I will start off with a short geology recap. Basically I have decided to stick with a nickel-cobalt alloy for my planet's core, which - on its own - will bring the core's density up to 8.9 g/cc (cc = cubic centimetre). I may or may not decide to add something else to the alloy but those two metallic elements will serve as the main component. I had been thinking of creating a portmanteau of the two metals for the new alloy's name...maybe something like "NiCo" after both of their atomic symbols/abbreviations.

I realize I have typed this numerous times before, but the reason why I want a higher density for my planet's core is because the mean density - which I already have - is dependent on what you have for crust density and core density. It should be somewhere in-between those two. Ideally, I should have figured out the density of the planet's core before figuring out the planet's mean density and then somehow this should have affected how big and how small the size of the planet could be before you start running into problems like premature geologic burnout or dissipation of the magnetic field. But I neglected to do this, so now I need to tweak the planet's core density. This is considerably more difficult than tweaking the planet's radius to accommodate the mean density once you've decided what your planet will be made of. So, yeah, important tip for hard sci-fi worldbuilders: start with what you want your planet to be made of first, then establish its mean density, then decide how big your planet will be. Most likely you will choose iron or an iron-based alloy because iron is the most common metallic element a star will produce when it dies and goes supernova, and will consequently be the most common metallic element found in the proto-planetary nebula from which your solar system will form. You may decide to use a different metal, but beyond the 26th element in the periodic table (iron), elements become a lot more rare and thus a lot more difficult to come by in a protoplanetary nebula. From what I have learned, the Period 4 Transitional Metals are your best bet.

Anyway, the basic principle of determining core size in relation to the rest of the planet is as follows: the closer the mean is to the density of the core, the bigger the core will be and the more room it will take up inside the planet. Conversely, the closer the mean is to the density of the planet's crust (which I will get to later), the smaller the core will be and the less room it will take up inside the planet. This is assuming the crust's density and the core's density are on opposite ends of a spectrum and the mean density somewhere between them. Approximate closeness of the mean to either ends should not be understood in an absolute context; you could theoretically have a very thick crust and henceforth justify moving the mean up much closer to the core's density - and I think many of the rocky planets outside of our solar system may have something like this - but for the sake of reality we will assume that "rocky" material has a mean of 3 g/cc. I am sure you will find an exception after looking at the periodic table, though.

Iron has a density of 7.874 g/cc. Take 7.452 away from that and you get 0.422 g/cc. Take 7.452 away from 8.3 g/cc, which is the Earth's current estimated core density, and you get 0.848 g/cc. No wonder I was told my planet would die out within a billion years...that is not even a third of the density of its crust! Take 5.5 g/cc - the Earth's mean density - away from 8.3 g/cc and you get 2.8 exactly. The Earth's continental crust has a density of 2.7 g/cc so I am thinking the difference between the mean density and the core density should be no less than the density of the planet's crust. Adding 2.7 (I am being minimalistic here) to 7.452 g/cc, I get 10.152 g/cc. that is a far cry from the 12.204 g/cc that I got from the M x 2 - 2.7 (M = mean density) equation. In light of this, I think I will use 10.152 g/cc as my minimum and 12.204 g/cc as my ideal. I don't know what my maximum would be and I don't much care; I am lucky to get anything denser than 10 g/cc as it is.

Yeah, I know I have typed about this whole density and core ratio thing a lot. The problem is I keep thinking that what I've covered is incomplete, then I forget I have already mentioned it or what I did mention, and a better way to phrase or explain it comes into my head. The result is the same thing repeated over and over again in a different way each time. That is likely where I am going with the issue of core size proportionate to planet size and the role of mean density in that comparison. I will probably need to look into the geological effects of a larger proportional core size rather than a smaller proportional core size relative to that of Earth. It will have a big effect on the tectonic plates; no more vertical plate movement. I kind of like the idea of a wading pool planet (albeit not a geologically dead one) as proposed in that ZBB thread about volcanic gases I have brought up numerous times before, so maybe having a larger core will not be so bad after all.

In the meantime, I will assume readers have generally gotten the picture by now so I will stop bringing it up in future entries. There may be further mention of core composition but only so far as to describe where I am at that moment.

Here is something I didn't bring up before: I looked through Wikipedia's Ferromagnetism article again and found the list of metals and alloys that are ferromagnetic. Yttrium is listed as a component of yttrium-iron garnet. Its atomic number, as shown on the periodic table, is 39. This means it is a Period 5 transitional metal. Europium, gadolinium and dysprosium are also listed, with the atomic numbers 63, 64 and 66 respectively. These latter three are all lanthanides, meaning they go in Period 6. All four are lower down in the periodic table than cobalt and nickel are, and yet none are as dense. I was tempted to try adding yttrium to my NiCo alloy because I thought it would increase the alloy's density, but yttrium has a density of only 4.472 g/cc. This would make my NiCo alloy less dense. It isn't something we usually think of when we add two things together, but when two different metals are added, their density mediates to somewhere in-between those of the two component metals instead of increasing beyond either. Ergo, I am left with the tricky task of finding a ferromagnetic metal - or a metal that won't reduce the ferromagnetic properties of the NiCo alloy - that is actually denser than that of whatever alloy I want my planet's core to be composed of.

Anyway, as I had just mentioned, I looked through the Wikipedia article for a better addition or substitute for my NiCo alloy but with unlucky results. Nevertheless I am listing what I found here for referential purposes. Listed by the extent of their Curie Temperature limits and sorted first by pure metals, then by metal alloys, they are:

Un-oxidized Pure Metals Un-oxidized Metal Alloys Density for each alloy was found thanks again to the Alloy Density Calculator. If you want the full list, you can go to the Wikipedia article I linked to earlier on ferromagnetism. I omitted the oxidized metals and alloys because apparently, when oxygen or some other gas is involved, something different happens to the resulting density of the new compound. I have not yet figured out how to determine the density of an oxidized metal so I am leaving all oxidized ferromagnetic metals and alloys out of the picture for now. I may return to them in a later entry. If I need to.
 * Cobalt - 8.9 g/cc
 * Iron - 7.874 g/cc
 * Nickel - 8.908 g/cc
 * Gadolinium - 7.901 g/cc
 * Dysprosium - 8.551 g/cc
 * Manganese-bismuth - 9.18852 g/cc
 * Manganese-antimony - 6.91963 g/cc
 * Manganese-arsenic - 6.35425 g/cc

Also, I feel embarrassed to say that hydrogen, nitrogen, oxygen, fluorine, chlorine, and all of the noble gases, use grams per litre instead of grams per cubic centimetre. One litre is apparently equivalent to 1,000 cubic centimetres, that much I know, but I haven't yet managed to assign a density in grams per cubic centimetre in a way that looks accurate. I tried assigning oxygen a density of 0.001429 g/cc by dividing the 1.429 g/l that is listed for it into 1,000 but the density I got for the resulting oxidized metal still didn't look right. The alloy calculator tool treats gases like metals so an intermediate density was assigned which was far less dense than the original metal. For reference, this site gives hematite a weight of 159.6882 g/mol and a density of 5.26 g/cc, and this site gives yttrium-iron garnet a weight of 737.95 g/mol and a density of 5.17 g/cc. I tried getting the density listed for hematite with the alloy calculator upon assuming a density of 0.001429 g/cc for oxygen but it didn't work. My result didn't even come close. And the yttrium-iron garnet is so complex that I am not going to bother trying to find its density myself until I am sure I know how.

So this is where I am at now. I am left with the nickel-cobalt alloy "NiCo" and that palladium-manganese-tin Heusler alloy I mentioned a couple of entries back. No other possible choices have presented themselves yet. When they do, I will bring them up in another entry. I actually have more to type about but this entry is long enough so I may bring it up in my next entry. Plus it will make the entires more topical that way. Stay tuned for that within the next few days or so.

Thanks for reading.