Lecture 6: Evolution of the Infant Moon (Our Moon Course)

Lecture 6: Evolution of the Infant Moon (Our Moon Course)


Today I would like to continue our discussion
about the early Moon. Last time we discussed how the Moon likely formed in the aftermath
of a giant impact and that it likely formed very hot. We think most of the Moon may have
been molten rock. This molten rock layer is called the Lunar Magma Ocean and today we’ll
consider how it cooled over time. We celebrated the 50th anniversary of the
Apollo 11 landing this past July and another 50th anniversary is coming up. In January
2020 we will be celebrating the 50th anniversary of the Lunar Magma Ocean idea, which was initially
proposed by two groups (their papers shown above) after they had analyzed anorthositic
fragments in the Apollo 11 sample collection. Last time we saw the standard or “canonical”
Giant Impact Model. In this model, the early Earth is impacted by a Mars-sized object (often
referred to as Theia). That giant impact produced a lot of hot debris and put it into orbit
around the Earth. Some of that debris then accreted to form the Moon. Notice again that
the figure on the right shows an edge view of the disk with the colors indicating the
temperatures in Kelvin. Not only did the Moon form out of material that was very hot (in
the range of 1000s of Kelvin) but the Moon was additionally heated in the formation process
itself as we will see in a moment. Before that I’d like to make a brief statement
about math. Since this course is intended for a general audience, I’m not going to
talk about a lot of math but I’ll occasionally show an equation. I know some folks don’t
like math and many may be turned away from science because of it. I mainly blame this
on the nonsense that people are told when they are young. That they “are not a math
person” or ”are not good at math.” My philosophy about things like this is similar
to the moto in the Pixar movie Ratatouille: “Anyone can cook.” I think anyone can
do math but you may not want to and that’s ok! Math is a language that helps us describe
things in a very compact way. You can write the equation F=ma completely in English or
any other language but it’ll likely be a paragraph of text instead of three variables
and one symbol. So if I show you an equation and you don’t understand it, it’s ok.
I’m happy to help. You can understand it. Much like learning a foreign language it just
takes practice. Ok now let’s look at an equation. This is
the equation for the gravitational binding energy of an object. There are two ways to
think of gravitational binding energy. One is to think about a planet. What if you wanted
to completely break it apart? By that I mean you want to take small pieces of the planet
and move them very large distances away from each other. Obviously planets don’t just
break apart in this manner naturally. You would need to add a considerable amount of
energy for a planet to break apart in this way. The energy required to do this is the
gravitational binding energy (U). It depends on the universal gravitational constant G,
the mass of the planet M, and the radius of the planet R. Note that technically this is
a minimum energy requirement since you’ll likely need to add additional energy to break
the bonds between solid objects. The other way of thinking about the gravitational binding
energy is to think about the opposite process, the formation of a planet. Imagine that small
pieces of material are very far away and slowly they collapsed due to gravity to form a planet.
In this case, gravity is doing the work. But all the small pieces of material had some
gravitational potential energy when they were far away. Energy doesn’t just disappear
so all that energy needs to go somewhere. That energy, which is the gravitational binding
energy, is converted to heat. Now there is a subtle point about this heating process.
Yes, there is a large quantity of energy that is release when a planet forms but whether
or not that melts the planet has to do with the speed at which the planet forms. If the
planet forms rather slowly, say over a billion years, then there would be plenty of time
for that energy to be dissipated away. However, if the planet forms very quickly, then there
isn’t enough time for that energy to be dissipated. As such, that energy goes into
melting the planet. So, going back to the Moon, here’s another reason why the Moon
likely started in a molten state. Not only were the materials that made the Moon starting
off at 1000s of Kelvin in temperature after the giant impact, but since the Moon likely
formed very quickly (i.e., in about 1 year) that gravitational energy would have gone
into melting the Moon. We saw last time that Moon rocks seem to agree
with this notion that the Moon was molten in the past. We already saw this sample from
Apollo 15, which is colloquially called the “Genesis Rock” since it was thought to
be a piece of the original crust of the Moon. This rock is unique in that it’s basically
made of one mineral (that is to say plagioclase). If you picked up a rock outside, that rock
will likely be made of a number of minerals. Rocks like this one from Apollo 15 are called
anorthositic rocks and finding them on the Moon led scientists to propose that a large
part of the Moon was molten rock in the past. Here’s another anorthosite sample but this
time from Apollo 16. Notice how you can see the lighter colored anorthosite once they
cut open the rock. Anorthositic rocks like these likely formed from the solidification
of the molten Moon in the past. This is an overview of how the Moon solidified
over time. There are various estimates of Lunar Magma Ocean (LMO) depth but we can assume
that it was about 1,000 km deep. Note that the radius of the Moon is about 1737 km, that
means that the early Moon largely consisted of the Lunar Magma Ocean. The Lunar Magma
Ocean would have solidified in two stages. First, as it cooled it would form minerals
like olivine, which being more dense than the surrounding liquid would have fallen to
the bottom. As such, the Lunar Magma Ocean would have solidified from the bottom up.
Once about 80% (by volume) of the Lunar Magma Ocean had solidified, there was a change.
The mineral plagioclase became stable and started to form anorthositic rocks (rocks
made mostly of the mineral plagioclase). Anorthositic rocks that formed would have been less dense
than the surrounding magma and as such they would have floated to the surface. The rocks
that floated to the surface would have formed the primordial crust of the Moon. Fragments
of that early crust were picked up by the Apollo astronauts and brought back to Earth. Note that the first stage (going from the
initial Lunar Magma Ocean to the point when 80% of it has solidified) would have occurred
rather quickly (in around 10,000 years). On the other hand, the second stage could have
taken 10s (or even 100s) of millions of years. The difference is because the floatation crust
acted as a thermal blanket and reduced the amount of heat through the surface. Besides Apollo samples, more recently, spacecraft
data have helped researchers identify areas of the Moon where there are pure anorthosite
(show in white circles above). Note that the center of the map consists of the nearside
of the Moon, the side we see from the Earth and the left and right sides of the map are
showing the farside, the side we cannot see from the Earth. This particular map is plotting
surface elevations along with locations where there are pure anorthosite material and locations
where there are olivine-rich material. The distribution of these pure anorthosite locations
across the lunar surface likely indicate that there is a global layer of anorthosite on
the Moon. Are there other pieces of evidence to support
the idea that the early Moon was mostly molten? Yes! The abundance (or lack) of the element
europium is likely the result of the crystallization of the Lunar Magma Ocean. Europium is interesting
because it is attracted to plagioclase. As such, when the lunar crust was forming, europium
in the Lunar Magma Ocean went into the early lunar crust and was removed from the Lunar
Magma Ocean. Notice in the figure above how the crust tends to have more europium, as
compared to KREEP (potassium [K], rare earth elements [REE], and phosphorus [P]). KREEP
material is representative of the last magma of the Lunar Magma Ocean, so it makes sense
that it has less europium (because it has gone into the crust). Now I want to briefly look at how we model
the Lunar Magma Ocean in computers. To model the Lunar Magma Ocean, we first need
to estimate what the chemical composition of it was after the Moon formed. The simple
explanation of what people have done is that they estimated the chemical composition of
the Lunar Magma Ocean by considering the chemical composition of the Earth’s mantle and adjusting
it from there. As you probably remember, the Moon likely formed mostly from Earth’s mantle
material, so this is a reasonable way of doing this. Various groups have different estimates
for the Lunar Magma Ocean’s chemical composition. I’ve shown here seven estimates from previous
works. The next step in the modeling process is to figure out which minerals would crystallize
out of the Lunar Magma Ocean. Let’s say we chose chemical composition “D.” In
that case, the first stacked graph on the right shows the expected sequence of minerals.
This is based on numerous experiments that showed the types of minerals that crystallize
out of different types of magma. Obviously, the minerals that form are derived from molecules
that are present in the Lunar Magma Ocean. From the bottom to the top, the graph shows
that the mineral olivine would form first, followed by other minerals. Towards the later
stages of crystallization, the low density mineral plagioclase would form. Note that
in the graph plagioclase is shown as “An” because it’s a specific type of plagioclase
called anorthite. Not to be confused with anorthosite. While anorthite is the calcium-rich
type of the mineral plagioclase, anorthosites are rocks mostly made of plagioclase. In summary,
the minerals that crystallize out of the Lunar Magma Ocean are dependent on the initial composition
of the Lunar Magma Ocean and the pressures and temperatures that are involved. If the
Lunar Magma Ocean was much shallower than 1,000 km, then even with the same initial
chemical composition, we would expect a different sequence of minerals to form since the pressures
involved in the cooling process would be different. The process of Lunar Magma Ocean solidification
involves many interacting factors. This is a sketch of many of the things that should
be considered. For example, we expect the Lunar Magma Ocean to have been convecting
and the surface to be radiating. However, we should consider if quench crust formed
on the surface. Formation of quench crust would have placed a conductive lid that slowed
the cooling process right from the start. Another important aspect is the detailed crystallization
process. Additionally, impacts may have played a role. Impacts could have punctured holes
in the floatation crust, increased the heat output, and thereby quickened the cooling
of the Moon. So far no computer model incorporates all of these factors at the same time, but
future works will consider how all of these factors interact to produce the solidified
Moon we see today. We saw this timeline of important early lunar
events last time. The age of the solar system is estimated by the oldest calcium-aluminum-rich
inclusion (CAI), shown by the dashed black line. The Moon is thought to have formed between
52 to 152 million years after the beginning of the solar system. Thus, the earliest and
the latest Moon formation times are shown in blue and orange lines respectively. One
recent work used computer modeling to calculate that the Lunar Magma Ocean took 30 million
years to solidify, without the influence of any additional cooling or heating mechanisms
such as impacts and tidal heating. I’ve added lighter colored rectangles next to the
solid blue and orange lines to represent the time required for the Moon to solidify. Now
let’s take a look at the ages of Moon rocks. Particularly, let’s consider radiometric
ages of ferroan anorthosite or FAN samples. I’m showing here the oldest and youngest
ferroan anorthosite ages along with their respective error bars. There are many other
age dates in between but for this discussion it’s useful to only look at the oldest and
youngest dates. First note that there are large error bars and that the oldest ferroan
anorthosite sample seems to be as old as the solar system. That’s unlikely. Also, notice
that the time difference between the oldest and youngest ages is about 270 million years.
If these are pieces of the original crust of the Moon, then the difference in ages should
indicate to us the solidification time of the Lunar Magma Ocean. Recall that one calculated
estimate for the Lunar Magma Ocean solidification time is 30 million years. 30 million years
is much shorter than 270 million years, so this could mean that the Lunar Magma Ocean
had an additional heat source to keep it hot. For example, tidal heating has been proposed
as a mechanism for heating the Lunar Magma Ocean. This is a possibility. However, how
reliable are the ages of the ferroan anorthosites? Recently, Borg and others graded the ages
of the various age estimates using five criteria. Only one ferroan anorthosite sample met all
five of their criteria. I’m showing that sample’s age here as the “Best FAN Age.”
I’ve also included the error bars but the error bars are much smaller than the red square
marker. The Borg paper also estimated the age of that incompatible element or KREEP
layer. Note that ‘ur’ is the German prefix for ‘original.’ Notice that the best ferroan
anorthosite age and the estimated age of the original KREEP layer basically overlap. That
makes sense that an anorthosite that formed towards the end of the Lunar Magma Ocean crystallization
process, would have the same age as the last bits of the Lunar Magma Ocean. However, we
don’t really have a very reliable estimate for the age of the oldest anorthosites on
the Moon. Finding a very reliable age for the oldest anorthosite would help make the
ages of the lunar crust samples consistent with the computer models of the Lunar Magma
Ocean. That would mean we have a more complete understanding of how the Moon solidified than
we do today. I now want to talk about a possible issue
with the Lunar Magma Ocean model. To do that let’s consider the different types of lunar
anorthosites. We have already seen ferroan anorthosites or FANs, which are rocks that
are mostly made of one mineral, plagioclase. More specifically, ferroan anorthosites are
made of the calcium-rich variety of plagioclase. Recall that version of plagioclase is called
anorthite and is shown above as “An.” Another type of lunar anorthosite has a bit
more variation in terms of the amount of anorthite in the rocks and has more magnesium than ferroan
anorthosites. That group of anorthosites is called the Mg-suite (for magnesium). There
is also another group of anorthosites called the alkali-suite but we won’t be discussing
them today. When these different types of anorthosites were discovered in the Apollo
sample collection, it was proposed that the ferroan anorthosites were pieces of the original
floatation crust of the Moon and the other anorthosites, like the Mg-suite of rocks were
produced later in the Moon’s history. As such, when these samples’ ages were measured,
the expectation was that ferroan anorthosites would be old and Mg-suite rocks would be younger. However, this isn’t the case. This figure
shows the ages of Mg-suite rock along with ferroan anorthosites. Please note that in
this figure ferroan anorthosites are referred to as FAS for ferroan anorthosite suite, but
the more common acronym is FAN. The plot is a bit backwards since the beginning of the
solar system is in the middle at about 4567 million years ago or equivalently 4.567 billion
years ago. Then time progresses to your left. As you can see, there is significant overlap
in the ages of ferroan anorthosite and Mg-suite rocks. This issue has been recently raised
and is part of ongoing research. While the Lunar Magma Ocean is the standard
idea of how the Moon solidified over time, there are other ideas that have been proposed
as well. One idea is serial magmatism. In this model, plagioclase-rich diapirs (or plumes)
rise much like the Lunar Magma Ocean model but they get lodged into already existing
crust on the Moon. One difference between the two models is that a plagioclase-rich
crust should be global in the Lunar Magma Ocean model but in the serial magmatism model
it’s more sporadic. Future work may be able to distinguish between these two models by
understanding how widespread plagioclase material is throughout the lunar crust.

Leave a Reply

Your email address will not be published. Required fields are marked *