What is the evidence for geomagnetic reversal ?

[rogerharris]

Apologies if this has been raised. Could not find what i was looking for.

We are being given paleomagnetics as part of a basic science degree, and i was having problems getting what the evidence is for geomagnetic reversal. My tutor could not answer this question and tells me to move on to the current assignments. I keep coming back to this. Damn curiosity !

the idea for the reversal theory is that magnetite aggregations have Subchronozone distributions clearly reversing their domain orientations across the planet within the same time frame.

However when i dig out papers i only find one section of the Atlantic where this is being explored and some mountain areas where domains went the opposite direction. at the same time we are being told that in these active mountain building areas the rock recycling could flip a crustal section 180 degrees anyway.

Ok my point is this, i originally thought that the magnetic bands on Earth wide sea floor spreading were marked black and white because the magnetite domain ordering had flipped 180 degrees. i.e. north becomes south and vice versa. And that was our Earth wide time synchronized evidence. However it seems that the bands are really areas where the magnetic field has become weak or the domains shift by slight degrees not 180 degrees ! Is this correct ? If so this just tells us that there has been Earth wide pole wander and not reversal.

Am i missing the obvious or did i misinterpret what i read ?

what is the evidence that clearly distinguishes regular earth-wide pole wander from actual regular 180 degree reversal ?

 

[Evo]

This post explains both pole shifts and pole reversals (magnetic field).

https://www.physicsforums.com/showpost.php?p=2753985&postcount=9

The entire thread https://www.physicsforums.com/showthread.php?t=406653

 

[Integral]

The evidence lies in the Mid Atlantic ridge. The Mid Atlantic Ridge is formed by an up welling of new rock. The magnetic field of the upwelling material is fixed when the molten rock solidifies. Thus as the ridge grows it leaves a time map of the magnetic field. It is in these rocks that they find periodic reversals of the magnetic field.

 

[Evo]

The evidence lies in the Mid Atlantic ridge. The Mid Atlantic Ridge is formed by an up welling of new rock. The magnetic field of the upwelling material is fixed when the molten rock solidifies. Thus as the ridge grows it leaves a time map of the magnetic field. It is in these rocks that they find periodic reversals of the magnetic field.

They find evidence many places.

http://www3.geosc.psu.edu/~jte2/geosc20/lect26.html

 

[Andre]

Much of the sediment cores drilled in the ocean in the Ocean Drilling Program are magnetized very weakly, yet detectable in labs like this.

Since the magnetic polarization proves to be very consistent in time/depth, it's one of the more popular methods to date the cores back hundreds of million years, using the geomagnetic pole reversals.

Here is an example of correlating geomagnetic dating of a lot of oceanic sediment cores (fig 4 on page 6):

Notice the black/white bar above the time scales with the names like Brunhes, Matuyama, jaramillo, Olduvai, etc, those are the geomagnetic "chrons", between geomagnetic pole reversals, which are dated very accurately, over and over again.

 

[geo101]

As folk have pointed out above there is a range of evidence from different sources that support geomagnetic reversals. I think some of your confusion lies in the mechanisms by which rocks can record the geomagnetic field. In particular the confusion between reversals of the geomagnetic field and the phenomenon of self-reversal behaviour, which is seen in some rocks.

I don’t have time to go into a detailed description right now, but, if your library has it, I would recommend Paleomagnetism: Continents and Oceans (2000) by McElhinny and McFadden (ISBN 0-12-483355-1). Chapter 4 explains the background, self-reversals and the evidence supporting geomagnetic field reversals.

And by the sounds of it, your tutor should read it too. It is an excellent all round paleomag book.

 

[rogerharris]

Much of the sediment cores drilled in the ocean in the Ocean Drilling Program are magnetized very weakly, yet detectable in labs like this.

Since the magnetic polarization proves to be very consistent in time/depth, it's one of the more popular methods to date the cores back hundreds of million years, using the geomagnetic pole reversals.

Here is an example of correlating geomagnetic dating of a lot of oceanic sediment cores (fig 4 on page 6):

Notice the black/white bar above the time scales with the names like Brunhes, Matuyama, jaramillo, Olduvai, etc, those are the geomagnetic "chrons", between geomagnetic pole reversals, which are dated very accurately, over and over again.

thanks, that's the kind of in depth stuff i was trying to find.

i will need to take some time out to get through that, as i can't see from this data where the field reverses rather than weakens, but i presume it just needs more reading and what this papers does is this.

1. plots distributions of the direction of the Earth's magnetic domains at each period in time.

2. Finds these are consistent with a single direction across the entire planet at a resolution that is reasonable to provide an overall picture and that this overal picture reguarly flips orientation domains.

3. seeks to falsify the data by seeking out local heat related domain flips as well as crustal overturn which both which can confound the data.

 

[rogerharris]

As folk have pointed out above there is a range of evidence from different sources that support geomagnetic reversals. I think some of your confusion lies in the mechanisms by which rocks can record the geomagnetic field. In particular the confusion between reversals of the geomagnetic field and the phenomenon of self-reversal behaviour, which is seen in some rocks.

I don’t have time to go into a detailed description right now, but, if your library has it, I would recommend Paleomagnetism: Continents and Oceans (2000) by McElhinny and McFadden (ISBN 0-12-483355-1). Chapter 4 explains the background, self-reversals and the evidence supporting geomagnetic field reversals.

And by the sounds of it, your tutor should read it too. It is an excellent all round paleomag book.

annoying. google books has all but chapter 4 and the uni library only has reviews. However i recognize the name McElhinny and think they put out a seminal paper which might provide a brief overview.

I do recall somebody pointing out that McElhinny can not so confidently extract latitude of the orientation of magnetic domains from the data as has been carried out. finals on other stuff so a come back later for project.

 

[rogerharris]

thanks for other replies. Primarily trying extract the in depth technical data for this subject. Not easy, there appears to be more geology books than journal publications for some reason. i guess that's due to them being out on the field most of the time.

 

[billiards]

The evidence lies in the Mid Atlantic ridge. The Mid Atlantic Ridge is formed by an up welling of new rock. The magnetic field of the upwelling material is fixed when the molten rock solidifies. Thus as the ridge grows it leaves a time map of the magnetic field. It is in these rocks that they find periodic reversals of the magnetic field.

The surveying of the Earth's magnetic field over the oceans provided the most compelling evidence for field reversals in conjunction with evidence for sea floor spreading. It is important to note that the anomalies are formed into the newly formed rock as it is cooled below its Curie temperature at the mid-ocean ridge. However, these anomalies migrate away from the ridge as the ocean floor physically moves away from the spreading centre. The point being that the evidence is not only found at the ridge; the evidence is found over the entire ocean floor. The compelling thing about it is that the same pattern of magnetic reversals is found either side of the spreading ridge. It really is beautiful.
http://www.moorlandschool.co.uk/earth/earth_science/Diverging_ridge_magnetism.gif

They find evidence many places.

http://www3.geosc.psu.edu/~jte2/geosc20/lect26.html

True. But the link you have provided does not help us to find out where those other places are.

Much of the sediment cores drilled in the ocean in the Ocean Drilling Program are magnetized very weakly, yet detectable in labs like this.

Since the magnetic polarization proves to be very consistent in time/depth, it's one of the more popular methods to date the cores back hundreds of million years, using the geomagnetic pole reversals.

Here is an example of correlating geomagnetic dating of a lot of oceanic sediment cores (fig 4 on page 6):

Notice the black/white bar above the time scales with the names like Brunhes, Matuyama, jaramillo, Olduvai, etc, those are the geomagnetic "chrons", between geomagnetic pole reversals, which are dated very accurately, over and over again.

I wouldn't spend too much time worrying about this. The plots are not evidence for magnetic reversals. They are using the magnetic reversals as a dating mechanism for a geochemical property of the rocks. This does not get at the heart of the question you originally asked.

 

[billiards]

i can't see from this data where the field reverses rather than weakens, but i presume it just needs more reading and what this papers does is this.

The issue is that when we are making a measurement of the magnetic field we are measuring everything. Whatever sources of magnetic field are out there, be they the Earth's magnetic field today or the rocks of 50 million years ago, we are measuring it all together in one bulk measurement (not to mention all the sources of "noise").

Now we can isolate bits of rock and analyse the orientation of their magnetic field in the lab. But that is not what was routinely done by the US military when they first started surveying the oceans. They just measured the whole field, and what they found was that the field got a bit stronger here, then a bit weaker over there, and then a bit stronger again, and when they plotted it all up it looked kind of stripy and they thought that that was kind of interesting. Why would the magnetic field look like that?

It turns out that it is because of the alignment of "magnetic domains" in the rocks. When the rocks were created they were very hot, they were above their http://en.wikipedia.org/wiki/Curie_temperature" and as the rock cooled the crystal in the rock formed little magnetic domains which lined up with the prevailing geomagnetic field and were frozen solid in place. Then the field reversed, and the new rocks that were being created all lined up their magnetic domains in the opposite direction.

So now when we measure the magnetic field we find that the magnetic field is stronger if the rocks were formed at a time when the magnetic field was in the same direction as it is today -- because it adds to the Earth's magnetic field. When we run our magnetometer over some rocks that were formed at a time when the magnetic field was in the opposite direction to what it is today we find that the magnetic field is weaker, because the magnetic field of the rocks is "fighting" against the Earth's present magnetic field. So what we measure is a strengthening and weakening of the magnetic field today, but this is caused by complete field reversals in the past. To confirm this you would need to look closely at samples of the rock in the lab and to measure the direction of the magnetic field -- these studies have been done.

 

[OmCheeto]

The issue is that when we are making a measurement of the magnetic field we are measuring everything. Whatever sources of magnetic field are out there, be they the Earth's magnetic field today or the rocks of 50 million years ago, we are measuring it all together in one bulk measurement (not to mention all the sources of "noise").

Doesn't one of the theories regarding field reversal say that the main field will start to wander and break up?

I was just looking at the following field map:

Figure 2. Total magnetic field presented in nanoTeslas
(please click on the image for the reference.)

And it looks as though we currently have two magnetic north poles, and one magnetic south pole.

Is this correct?

disclaimer: I am not a geophysicist, have never been one, and doubt I ever will be. But I do like playing with magnets.

 

[Evo]

What the Earth's magnetic fields are supposed to look like during a reversal.

http://en.wikipedia.org/wiki/Geomagnetic_reversal#Character_of_transitions

A good article from NASA.

http://www.nasa.gov/vision/earth/lookingatearth/29dec_magneticfield.html

 

[rogerharris]

The issue is that when we are making a measurement of the magnetic field we are measuring everything. Whatever sources of magnetic field are out there, be they the Earth's magnetic field today or the rocks of 50 million years ago, we are measuring it all together in one bulk measurement (not to mention all the sources of "noise").

Now we can isolate bits of rock and analyse the orientation of their magnetic field in the lab. But that is not what was routinely done by the US military when they first started surveying the oceans. They just measured the whole field, and what they found was that the field got a bit stronger here, then a bit weaker over there, and then a bit stronger again, and when they plotted it all up it looked kind of stripy and they thought that that was kind of interesting. Why would the magnetic field look like that?

It turns out that it is because of the alignment of "magnetic domains" in the rocks. When the rocks were created they were very hot, they were above their http://en.wikipedia.org/wiki/Curie_temperature" and as the rock cooled the crystal in the rock formed little magnetic domains which lined up with the prevailing geomagnetic field and were frozen solid in place. Then the field reversed, and the new rocks that were being created all lined up their magnetic domains in the opposite direction.

So now when we measure the magnetic field we find that the magnetic field is stronger if the rocks were formed at a time when the magnetic field was in the same direction as it is today -- because it adds to the Earth's magnetic field. When we run our magnetometer over some rocks that were formed at a time when the magnetic field was in the opposite direction to what it is today we find that the magnetic field is weaker, because the magnetic field of the rocks is "fighting" against the Earth's present magnetic field. So what we measure is a strengthening and weakening of the magnetic field today, but this is caused by complete field reversals in the past. To confirm this you would need to look closely at samples of the rock in the lab and to measure the direction of the magnetic field -- these studies have been done.

Ok thanks, well explained. so basically where the field is weaker the magnetic domains are actually completely flipped but this is hard to tell in the ground. so in the ground we really can't know if its flipped or just wandered.

A piece of this weak rock has to be cut out, with its north pole labeled, and another piece cut out where the field is strong and similarly labeled, there would be complete dipole reversals of each other under lab conditions with magnetic shielding from the Earth's background field(presuming the magnetic oxides are similar) ?

If anybody knows the best papers which summarize that evidence where its completely clear that we have flipping and not just wander or fluctuation that would be useful thanks.

 

[rogerharris]

What the Earth's magnetic fields are supposed to look like during a reversal.

http://en.wikipedia.org/wiki/Geomagnetic_reversal#Character_of_transitions

A good article from NASA.

http://www.nasa.gov/vision/earth/lookingatearth/29dec_magneticfield.html

yeh i have seen these simulations, but had the problem they may have modeled presumptions.

another aspect i don't get is that the hard ferromagnetic core where the bulk of the magnetic field comes from, how does this undergo a domain flip due to paramagnetic fluids sloshing around it. These fluids are kind of random in their turbulence and they put out a weaker field than the core, so where do they get the consistency and magnetic power to re-orientate the hardcores magnetic dipole.

Especially as the hard core itself remains cooler due to convection so its curie temp is not being raised. So there is another question, what is the actual reversal mechanism. These new models point more to the idea that the soft outer core can really just modulate and not flip the hard cores dipole.


these are the recent articles on the hard core of the Earth being a lower temperature dipole BTW. It even says "A cross-section of the Earth's interior shows the outer crust, the hot gooey mantle, the liquid outer core and the solid, frozen inner core (gray). (Calvin J. Hamilton graphic)"

Although i don't know what they mean by frozen as in it was frozen and its temp is higher now.

http://newscenter.berkeley.edu/2010/12/16/earth-magnetic-field/"

https://docs.google.com/viewer?a=v&...LGGMlJs45g93JLU8CsB0dHOc3CVgvrjm-2c&hl=en_US"

https://www.google.com/accounts/Ser...MyLTk0N2QtYzcwZmQxOGIyOTZj&hl=en_US&hl=en_US"

it says in the news report

"The Earth’s magnetic field is produced in the outer two-thirds of the planet’s iron/nickel core. This outer core, about 1,400 miles thick, is liquid, while the inner core is a frozen iron and nickel wrecking ball with a radius of about 800 miles – roughly the size of the moon. The core is surrounded by a hot, gooey mantle and a rigid surface crust."

So if the inner core is 800 miles radius of compressed and cool iron/ nickel would this not be a pretty strong magnetic dipole structure on its own.

 

[Evo]

Have you read any of the papers suggested at the bottom of the NASA article?

Like this one?

http://es.ucsc.edu/~glatz/geodynamo.html

 

[billiards]

Ok thanks, well explained. so basically where the field is weaker the magnetic domains are actually completely flipped but this is hard to tell in the ground. so in the ground we really can't know if its flipped or just wandered.

A piece of this weak rock has to be cut out, with its north pole labeled, and another piece cut out where the field is strong and similarly labeled, there would be complete dipole reversals of each other under lab conditions with magnetic shielding from the Earth's background field(presuming the magnetic oxides are similar) ?

So the theory is that whatever field was prevalent at the time the rock cooled below its Curie temperature will be locked into the rock. Actually you'll find that it can alter a bit with weathering, and there is alway the distinct possibility that the rock may be tilted -- even completely overturned -- by later geotectonic activity. It's never quite THAT simple.

But yes, if you wish to find the magnetic field of the rock (and then infer that this represents the geomagnetic field at the time of its formation) then you need to isolate a sample of it.

If anybody knows the best papers which summarize that evidence where its completely clear that we have flipping and not just wander or fluctuation that would be useful thanks.

This is not my area and so I do not know the BEST papers. You could try this one (http://www.springerlink.com/content/c5m8212702058nw8/) to see a study of the rocks in the Deccan Traps -- which themselves are extremely interesting for other geological reasons.

 

[billiards]

So if the inner core is 800 miles radius of compressed and cool iron/ nickel would this not be a pretty strong magnetic dipole structure on its own.

The inner core is not permanently magnetised. This is quite simple to see. Curie temperature gets lower with increasing pressure -- that means that at the centre of the Earth the curie temperature will be quite low (the pressure being quite high!). Temperature increases with depth -- that is another way of saying that the temperature of the inner core is really very high. The conclusion of this, if you think about it (and perhaps if necessary check a few numbers,) is that it is impossible for the inner core to be permanently magnetised.

EDIT: I should also mention that above the Curie temperature it is not possible for a body to have a permanent magnetic field.

The geomagnetic field is generated by the geodynamo as modeled by Glatzmeier and others.

 

[rogerharris]

Have you read any of the papers suggested at the bottom of the NASA article?

Like this one?

http://es.ucsc.edu/~glatz/geodynamo.html

thanks. I've seen this material, but problem is every few years the onions sway back and forwards between scientists as to various parameters and models, and i am curious as to whether there is ferromagnetic order at the inner core despite the high temperatures.

e.g. a while back this summary explains opinion of the inner core structure

http://www.psc.edu/science/Cohen_Stix/cohen_stix.html

Prevalent opinion before these calculations held that iron's crystal structure in the inner core was bcc. To the contrary, the calculations showed, bcc iron is unstable at high pressure and not likely to exist in the inner core.

now we have this from japan last year.

http://www.pnas.org/content/107/21/9507.shor

The detected travel-time anomalies can only be disclosed by a lattice-preferred orientation of a body-centered-cubic iron aggregate, having a fraction of their [111] crystal axes parallel to the Earth’s rotation axis. This is compelling evidence for the presence of a body-centered-cubic Fe phase at the top of the Earth’s inner core.

At the top of the inner core as pointed out by other recent research there is supposed to be freezing occurring due massive latent convection.

http://newscenter.berkeley.edu/2010/12/16/earth-magnetic-field/

About 60 percent of the power generated inside the Earth likely comes from the exclusion of light elements from the solid inner core as it freezes and grows, he said. This constantly builds up crud in the outer core.

The Earth’s magnetic field is produced in the outer two-thirds of the planet’s iron/nickel core. This outer core, about 1,400 miles thick, is liquid, while the inner core is a frozen iron and nickel wrecking ball with a radius of about 800 miles – roughly the size of the moon. The core is surrounded by a hot, gooey mantle and a rigid surface crust.

BTW anybody have any idea what this crud is ?

Although papers from the 1950's negate the possibility of high pressure overcoming the curie temps at inner core pressure using calculation from the models of that time, curie temps appear to rise in line with pressure when an apparatus is actually devised, as more recently.

http://iris.elf.stuba.sk/JEEEC/data/pdf/8s_106-6.pdf

MAGNETIC MEASUREMENTS UNDER PRESSURE
Mária Zentková* – Zdenĕk Arnold** - Matúš Mihalik*** - Marián Mihalik* – Anton Zentko * -
Jiří Kamarád** - Zuzana Mitróová* - Slavomír Maťaš*
Two different methods were used to demonstrate that high pressure is a useful tool for investigation of magnetic properties. We report
on the effect of high pressure on the ferromagnetic transition in PrNi single crystal. The Curie temperature was found to increase
under pressure up to 0.9 GPa with a positive pressure coefficient Tc/p = 1 K/GPa. Such a behavior has been attributed to
enhancement of ferromagnetic coupling between Pr ions in PrNi due to pressure induced instabilities of the crystal field singlet
ground state of PrNi. The measurement was realized by transformer method. Additionally, the effect of pressure on magnetic properties
of Cr3[Cr(CN)6]2 x 15 H2O has been studied by means of SQUID magnetometry. Observed increase of Curie temperature
with the pressure coefficient Tc/p = 26 K/GPa can be explained by pressure induced increased overlapping of magnetic orbitals.

However i have not found any recent calculation of this effect for Earth's core to explain how it can have the Bcc structure.

i think this equation is used for this.

http://en.wikipedia.org/wiki/Clausius–Clapeyron_relation

So I am just wondering if there can be ferromagnetism at the surface of inner core. The new models are saying it has a Bcc structure and at this layer temperatures are freezing. Are these linked ?

 

[rogerharris]

This is not my area and so I do not know the BEST papers. You could try this one (http://www.springerlink.com/content/c5m8212702058nw8/) to see a study of the rocks in the Deccan Traps -- which themselves are extremely interesting for other geological reasons.

thanks for that, looks like i might have to hassle a geologist if i read a load of stuff and its still not clear. Luckily got other exams to to distract me from this momentary obsession for a month or so.

 

[rogerharris]

there are now two recent studies which are saying that the inner core has either a Bcc or Hcp structure which are ferromagnetic.

according to this team from 2010

http://www.pnas.org/content/107/21/9507.shor


Hemispherical anisotropic patterns of the Earth’s inner core
Maurizio Mattesinia,1, Anatoly B. Belonoshkob, Elisa Buforna, María Ramíreza, Sergei I. Simakc, Agustín Udíasa, Ho-Kwang Maod, and Rajeev Ahujae,f

+
Author Affiliations
aDepartamento de Física de la Tierra, Astronomía y Astrofísica I, Universidad Complutense de Madrid, E-28040 Madrid, Spain;
bCondensed Matter Theory, Department of Theoretical Physics, AlbaNova University Center, Royal Institute of Technology, SE-10691 Stockholm, Sweden;
cDivision of Theory and Modeling, Linköping University, SE-581 83 Linköping, Sweden;
dGeophysical Laboratory, Carnegie Institution of Washington, Washington DC 20015;
eCondensed Matter Theory Group, Physics Department, Uppsala University, Box 530, SE-75121 Uppsala, Sweden; and
fApplied Materials Physics, Department of Materials Science and Engineering, Royal Institute of Technology, SE-10044 Stockholm, Sweden

Contributed by Ho-Kwang Mao, April 12, 2010 (sent for review March 20, 2010)
Abstract

It has been shown that the Earth’s inner core has an axisymmetric anisotropic structure with seismic waves traveling ∼3% faster along polar paths than along equatorial directions. Hemispherical anisotropic patterns of the solid Earth’s core are rather complex, and the commonly used hexagonal-close-packed iron phase might be insufficient to account for seismological observations. We show that the data we collected are in good agreement with the presence of two anisotropically specular east and west core hemispheres. The detected travel-time anomalies can only be disclosed by a lattice-preferred orientation of a body-centered-cubic iron aggregate, having a fraction of their [111] crystal axes parallel to the Earth’s rotation axis. This is compelling evidence for the presence of a body-centered-cubic Fe phase at the top of the Earth’s inner core.

here bcc is ferromagnetic but not fcc, yet bcc is indicative of the iron being below the curie point.

http://www.iap.tuwien.ac.at/www/sur...stallography#the_fcc-bcc_phase_transformation

At ambient conditions, pure iron is body-centered cubic (bcc) and ferromagnetic (i.e., it can be magnetized and is strongly attracted by magnets). Above approx. 920 °C, it becomes face centered cubic (fcc). Whereas the bcc phase gains its stability from magnetism (even though it becomes paramagnetic above its Curie temperature of 770 °C), the high-temperature fcc phase is paramagnetic

recently on tv last week.

http://www.bbc.co.uk/news/uk-14678002

Kei Hirose has created an incredibly powerful vice using the tips of two diamonds. Between them he has pressurised a sample of iron-nickel to three million times atmospheric pressure and heated the sample to about 4,500C.

Under these extraordinary conditions, the crystal structure of iron-nickel alloy changed and the crystals rapidly grew in size. "We may have very big crystals at the centre of the Earth, maybe up to 10km," says Hirose.

These crystals would all align "like a forest", says Hirose, pointing at the poles.

from the research institute

http://www.spring8.or.jp/en/news_publications/press_release/2010/101015

and the paper

http://www.sciencemag.org/content/330/6002/359

So even in this Hcp phase nickel is still ferromagnetic.

http://prb.aps.org/abstract/PRB/v39/i4/p2526_1

The Stoner theory of ferromagnetism has been applied to 3d transition metals in the hexagonal-close-packed (hcp) phase. The elements Co and Ni (and possibly Cr) are found to be ferromagnetic. A self-consistent calculation of the band structure of paramagnetic hcp Ni revealed the highest-known density of states at the Fermi level of any transition metal in any structure, providing strong evidence for ferromagnetism


so there is now two models that propose the iron and nickel could be in ferromagnetic states
even at these incredible temps. the first proposing the Bcc state for core surface, the next the hcp state.

Another recent article points out chromium could be part of the Earth's core. Dont know if it sill is. chromium is also ferromagnetic in the hcp state..

http://www.sciencemag.org/content/331/6023/1417.abstract

So... looks like no current consensus as to what's even in there, never mind what state it is ferromagnetic or not.

 

[billiards]

there are now two recent studies which are saying that the inner core has either a Bcc or Hcp structure which are ferromagnetic.

according to this team from 2010

http://www.pnas.org/content/107/21/9507.shor


Hemispherical anisotropic patterns of the Earth’s inner core
Maurizio Mattesinia,1, Anatoly B. Belonoshkob, Elisa Buforna, María Ramíreza, Sergei I. Simakc, Agustín Udíasa, Ho-Kwang Maod, and Rajeev Ahujae,f

+
Author Affiliations
aDepartamento de Física de la Tierra, Astronomía y Astrofísica I, Universidad Complutense de Madrid, E-28040 Madrid, Spain;
bCondensed Matter Theory, Department of Theoretical Physics, AlbaNova University Center, Royal Institute of Technology, SE-10691 Stockholm, Sweden;
cDivision of Theory and Modeling, Linköping University, SE-581 83 Linköping, Sweden;
dGeophysical Laboratory, Carnegie Institution of Washington, Washington DC 20015;
eCondensed Matter Theory Group, Physics Department, Uppsala University, Box 530, SE-75121 Uppsala, Sweden; and
fApplied Materials Physics, Department of Materials Science and Engineering, Royal Institute of Technology, SE-10044 Stockholm, Sweden

Contributed by Ho-Kwang Mao, April 12, 2010 (sent for review March 20, 2010)
Abstract

It has been shown that the Earth’s inner core has an axisymmetric anisotropic structure with seismic waves traveling ∼3% faster along polar paths than along equatorial directions. Hemispherical anisotropic patterns of the solid Earth’s core are rather complex, and the commonly used hexagonal-close-packed iron phase might be insufficient to account for seismological observations. We show that the data we collected are in good agreement with the presence of two anisotropically specular east and west core hemispheres. The detected travel-time anomalies can only be disclosed by a lattice-preferred orientation of a body-centered-cubic iron aggregate, having a fraction of their [111] crystal axes parallel to the Earth’s rotation axis. This is compelling evidence for the presence of a body-centered-cubic Fe phase at the top of the Earth’s inner core.

here bcc is ferromagnetic but not fcc, yet bcc is indicative of the iron being below the curie point.

http://www.iap.tuwien.ac.at/www/sur...stallography#the_fcc-bcc_phase_transformation

At ambient conditions, pure iron is body-centered cubic (bcc) and ferromagnetic (i.e., it can be magnetized and is strongly attracted by magnets). Above approx. 920 °C, it becomes face centered cubic (fcc). Whereas the bcc phase gains its stability from magnetism (even though it becomes paramagnetic above its Curie temperature of 770 °C), the high-temperature fcc phase is paramagnetic

recently on tv last week.

http://www.bbc.co.uk/news/uk-14678002

Kei Hirose has created an incredibly powerful vice using the tips of two diamonds. Between them he has pressurised a sample of iron-nickel to three million times atmospheric pressure and heated the sample to about 4,500C.

Under these extraordinary conditions, the crystal structure of iron-nickel alloy changed and the crystals rapidly grew in size. "We may have very big crystals at the centre of the Earth, maybe up to 10km," says Hirose.

These crystals would all align "like a forest", says Hirose, pointing at the poles.

from the research institute

http://www.spring8.or.jp/en/news_publications/press_release/2010/101015

and the paper

http://www.sciencemag.org/content/330/6002/359

So even in this Hcp phase nickel is still ferromagnetic.

http://prb.aps.org/abstract/PRB/v39/i4/p2526_1

The Stoner theory of ferromagnetism has been applied to 3d transition metals in the hexagonal-close-packed (hcp) phase. The elements Co and Ni (and possibly Cr) are found to be ferromagnetic. A self-consistent calculation of the band structure of paramagnetic hcp Ni revealed the highest-known density of states at the Fermi level of any transition metal in any structure, providing strong evidence for ferromagnetismso there is now two models that propose the iron and nickel could be in ferromagnetic states
even at these incredible temps. the first proposing the Bcc state for core surface, the next the hcp state.

Another recent article points out chromium could be part of the Earth's core. Dont know if it sill is. chromium is also ferromagnetic in the hcp state..

http://www.sciencemag.org/content/331/6023/1417.abstract

So... looks like no current consensus as to what's even in there, never mind what state it is ferromagnetic or not.

I think you have taken some time to make your post, and it is a good post. You are clearly thinking on a higher level than I had given you credit for.

One thing that I had not fully appreciated until I looked into it further (and really it should have been obvious) is that the Curie temperature as a function of pressure is not always the same simple relationship: The relationship between Curie temperature and pressure is different for different materials.

I think that you have slipped up a bit in that you assume that BCC iron is always ferromagnetic -- you seem to think that BCC iron does not have a Curie temperature. I could understand this if BCC iron was not stable unless it was ferromagnetic, however, a little digging reveals that BCC iron does have a Curie temperature. Furthermore the relationship between BCC iron and Curie temperature has been studies, and is known experimentally and from first principle calculations.

The pressure dependence of the Curie temperature in bcc iron has been studied employing spin-density-functional theory in combination with the Heisenberg model. We show that the results correctly predict an essential independence of the Curie temperature of bcc iron on the external pressure, in agreement with the experimental findings. This behavior is explained as a result of a competition between the decrease in the local magnetic moments and the increase in the magnetic coupling as function of rising pressure.

http://prb.aps.org/abstract/PRB/v79/i18/e184406

The upshot is that the Curie temperature for BCC iron is the same for all pressures! What's the Curie temperature for BCC iron then? Certainly less than 2000 K. What the temperature of the inner core? Certainly more than 4000 K. Therefore we can be certain that if the core is composed of BCC iron it is NOT permanently magnetised.

Now I believe the same conclusion holds for HCP iron, but would be interested to see if you can prove otherwise.

 

[geo101]

annoying. google books has all but chapter 4 and the uni library only has reviews. However i recognize the name McElhinny and think they put out a seminal paper which might provide a brief overview.

I do recall somebody pointing out that McElhinny can not so confidently extract latitude of the orientation of magnetic domains from the data as has been carried out. finals on other stuff so a come back later for project.

There are a couple of other books available on Google that should answer the original questions about reversals

The magnetic field of the earth: paleomagnetism, the core, and the deep mantle
By Ronald T. Merrill, M. W. McElhinny, Phillip L. McFadden
Chapter 5, but not all is available.
http://books.google.co.uk/books?id=96APl4nK9lIC&printsec=frontcover#v=onepage&q&f=false"

or

Reversals of the Earth's magnetic field
By John Arthur Jacobs
Older, but Chapter 2 should give you the back ground.
http://books.google.co.uk/books?id=MUckFaE5hfYC&printsec=frontcover#v=onepage&q&f=false"


Some clarification on the core papers...

Kei Hirose has created an incredibly powerful vice using the tips of two diamonds. Between them he has pressurised a sample of iron-nickel to three million times atmospheric pressure and heated the sample to about 4,500C.

Under these extraordinary conditions, the crystal structure of iron-nickel alloy changed and the crystals rapidly grew in size. "We may have very big crystals at the centre of the Earth, maybe up to 10km," says Hirose.

These crystals would all align "like a forest", says Hirose, pointing at the poles.

So the most likely Fe phase of the inner core is hcp Fe ($\epsilon$-Fe).
To avoid confusion (in case there is any, maybe not) the poles that Hirose is pointing to are geographic poles they are not related in any way to magnetic poles.

The Stoner theory of ferromagnetism has been applied to 3d transition metals in the hexagonal-close-packed (hcp) phase. The elements Co and Ni (and possibly Cr) are found to be ferromagnetic. A self-consistent calculation of the band structure of paramagnetic hcp Ni revealed the highest-known density of states at the Fermi level of any transition metal in any structure, providing strong evidence for ferromagnetism

Firstly, here there are no reports of the temperatures at which these elements are ferromagnetic (I haven't read the paper). I would argue that the key point here is that this is referring to Ni, Cr and Co. By wt% they constitute ~5%, ~1% and so small it is not listed in the table! Fe is ~85% [1]. There are relatively minor constituents of the core.


[1] McDonough, W. F. (2007), Core composition, in Encyclopedia of Geomagnetism and Paleomagnetism, edited by D. Gubbins and E. Herrero-Bervera, pp. 77-80, Springer, Dordrecht.


Curie temperatures (Tc)
The temperature of inner core at the inner core boundary (ICB) is estimated to be 5000 K ($\pm$500 k) so let's use that.

The upshot is that the Curie temperature for BCC iron is the same for all pressures! What's the Curie temperature for BCC iron then? Certainly less than 2000 K. What the temperature of the inner core? Certainly more than 4000 K. Therefore we can be certain that if the core is composed of BCC iron it is NOT permanently magnetised.

While the Tc of Fe in invariant of pressure, there is some evidence to suggest that for Fe-Ni alloys Tc decreases with pressure at a rate which depends on the Fe-Ni ratio (http://prb.aps.org/abstract/PRB/v6/i11/p4250_1) This an old reference, I'll see if I can dig out a more recent one.

Regardless, the Tc of bcc Fe is ~1040 K so at core temperatures it is paramagnetic. So even at depths well above the ICB Fe cannot hold a permanent magnetization.

As for hcp Fe the main magnetic ordering is reported to be antiferromagnetic, which has a Tc of 69 K (http://prb.aps.org/abstract/PRB/v67/i18/e180405). This is model based.

In addition to all of this there are observation that indicate that as the structure of Fe changes to hcp magnetic ordering is lost.

The single Mossbauer line indicates that Fe57 may not be aligned in the hexagonal phase. This might indicate a paramagnetic medium...

http://www.sciencemag.org/content/141/3585/1035.full.pdf

I hope this all helps.

 

[rogerharris]

The upshot is that the Curie temperature for BCC iron is the same for all pressures! What's the Curie temperature for BCC iron then? Certainly less than 2000 K. What the temperature of the inner core? Certainly more than 4000 K. Therefore we can be certain that if the core is composed of BCC iron it is NOT permanently magnetised.

Now I believe the same conclusion holds for HCP iron, but would be interested to see if you can prove otherwise.

well there goes that idea, i guess ill just have to live with the reality of an unstable core !

 

[rogerharris]

There are a couple of other books available on Google that should answer the original questions about reversals

The magnetic field of the earth: paleomagnetism, the core, and the deep mantle
By Ronald T. Merrill, M. W. McElhinny, Phillip L. McFadden
Chapter 5, but not all is available.
http://books.google.co.uk/books?id=96APl4nK9lIC&printsec=frontcover#v=onepage&q&f=false"

or

Reversals of the Earth's magnetic field
By John Arthur Jacobs
Older, but Chapter 2 should give you the back ground.
http://books.google.co.uk/books?id=MUckFaE5hfYC&printsec=frontcover#v=onepage&q&f=false"


Some clarification on the core papers...


So the most likely Fe phase of the inner core is hcp Fe ($\epsilon$-Fe).
To avoid confusion (in case there is any, maybe not) the poles that Hirose is pointing to are geographic poles they are not related in any way to magnetic poles.



Firstly, here there are no reports of the temperatures at which these elements are ferromagnetic (I haven't read the paper). I would argue that the key point here is that this is referring to Ni, Cr and Co. By wt% they constitute ~5%, ~1% and so small it is not listed in the table! Fe is ~85% [1]. There are relatively minor constituents of the core.


[1] McDonough, W. F. (2007), Core composition, in Encyclopedia of Geomagnetism and Paleomagnetism, edited by D. Gubbins and E. Herrero-Bervera, pp. 77-80, Springer, Dordrecht.


Curie temperatures (Tc)
The temperature of inner core at the inner core boundary (ICB) is estimated to be 5000 K ($\pm$500 k) so let's use that.




While the Tc of Fe in invariant of pressure, there is some evidence to suggest that for Fe-Ni alloys Tc decreases with pressure at a rate which depends on the Fe-Ni ratio (http://prb.aps.org/abstract/PRB/v6/i11/p4250_1) This an old reference, I'll see if I can dig out a more recent one.

Regardless, the Tc of bcc Fe is ~1040 K so at core temperatures it is paramagnetic. So even at depths well above the ICB Fe cannot hold a permanent magnetization.

As for hcp Fe the main magnetic ordering is reported to be antiferromagnetic, which has a Tc of 69 K (http://prb.aps.org/abstract/PRB/v67/i18/e180405). This is model based.

In addition to all of this there are observation that indicate that as the structure of Fe changes to hcp magnetic ordering is lost.


http://www.sciencemag.org/content/141/3585/1035.full.pdf

I hope this all helps.

Thanks for these. Its looking unlikely to make a case there could be any ferro or anti-magnetism in the core.

What got me interested in there being a ferro or anti-ferromagnetic order was this freezing layer that is brought up. Ill need to dig it back out. It seemed that what was being said was that there was still such high latent extraction that the surface of the inner core had freezing even today. Hirose talked of drops of frozen crystals here.

I need to extract the papers rather than news articles. I am guessing i must have misunderstood and taken this freezing out of Earth evolution time context ?

 

[geo101]

What got me interested in there being a ferro or anti-ferromagnetic order was this freezing layer that is brought up. Ill need to dig it back out. It seemed that what was being said was that there was still such high latent extraction that the surface of the inner core had freezing even today. Hirose talked of drops of frozen crystals here.

The freezing of the inner core at the inner core boundary (ICB) is completely correct. The inner core is solid and it is growing as heavier elements (mostly Fe) from the liquid outer core solidify at the ICB. Although the core is solid it is at a temperature well above it's Curie temperature and cannot hold any permanent magnetization.

Taking this a little further, the freezing of heavier elements at the ICB creates a compositional gradient in the liquid outer core. As the inner core grows lighter elements are released at the ICB. This drives compositional convection in the outer core and this is one of the fundamental forces that runs the geodynamo and hence creates the geomagnetic field.

 

[rogerharris]

The freezing of the inner core at the inner core boundary (ICB) is completely correct. The inner core is solid and it is growing as heavier elements (mostly Fe) from the liquid outer core solidify at the ICB. Although the core is solid it is at a temperature well above it's Curie temperature and cannot hold any permanent magnetization.

Taking this a little further, the freezing of heavier elements at the ICB creates a compositional gradient in the liquid outer core. As the inner core grows lighter elements are released at the ICB. This drives compositional convection in the outer core and this is one of the fundamental forces that runs the geodynamo and hence creates the geomagnetic field.

So how thick is this ICB layer where it is actually freezing and what is the temperature there ?

Well you see what I am trying to get at. This is really the bit i didnt understand.

 

[geo101]

So how thick is this ICB layer where it is actually freezing and what is the temperature there ?

Well you see what I am trying to get at. This is really the bit i didnt understand.

This layer of crystal and liquid "mush" is on the order of about 150 km, but it has variable thickness. I don't want to complicate things too much, but in some areas the inner core may actually be melting (http://www.nature.com/nature/journal/v473/n7347/full/nature10068.html"). The temperature here is ~ 5000 K. In fact this estimate is based on the coexistence of both liquid and solid Fe phases. Strictly, the melting point of Fe at ICB pressures is ~6200-6350 K, but the addition of lighter elements (e.g., O, Si, S are the more likely candidates) will suppress this.

Is it the temperature structure of Earth that you are not too sure about? I don't have a good diagram handy, but if that is the confusion I can find one.

 

[billiards]

While the Tc of Fe in invariant of pressure, there is some evidence to suggest that for Fe-Ni alloys Tc decreases with pressure at a rate which depends on the Fe-Ni ratio (http://prb.aps.org/abstract/PRB/v6/i11/p4250_1) This an old reference, I'll see if I can dig out a more recent one.

Perhaps this is why I thought the Curie temperature decreased?

As for hcp Fe the main magnetic ordering is reported to be antiferromagnetic, which has a Tc of 69 K (http://prb.aps.org/abstract/PRB/v67/i18/e180405). This is model based.

In addition to all of this there are observation that indicate that as the structure of Fe changes to hcp magnetic ordering is lost.


http://www.sciencemag.org/content/141/3585/1035.full.pdf

I hope this all helps.

Thanks for that. Good information that helps to complete the picture!

It seems there is no way around it. The inner core is emphatically NOT a permanent magnet!

 

[geo101]

As for hcp Fe the main magnetic ordering is reported to be antiferromagnetic, which has a Tc of 69 K (http://prb.aps.org/abstract/PRB/v67/i18/e180405). This is model based.

In addition to all of this there are observation that indicate that as the structure of Fe changes to hcp magnetic ordering is lost.

Just to clarify (I hadn't finished my morning coffee when I wrote this).

I say that hcp Fe loses magnetic ordering as it due to the structural change "in addition to" it's low Tc. This is should really be that as Fe transforms to hcp Fe loses magnetic ordering due to the fact that hcp Fe has a lower Tc (these experiments were at room temperature).

 

[rogerharris]

Just to clarify (I hadn't finished my morning coffee when I wrote this).

I say that hcp Fe loses magnetic ordering as it due to the structural change "in addition to" it's low Tc. This is should really be that as Fe transforms to hcp Fe loses magnetic ordering due to the fact that hcp Fe has a lower Tc (these experiments were at room temperature).

im a believer ! not seeing how there could be a permanent magnet anyway.

here comes the columbo bit tho

whats the core mechanism for reversal ? if these inner cores are massive upward pointing ultra compressed paramagnetic crystals, and the dipole field is created by the liquid sloshing around it, then the crystals inner core still has the dominant paramagnetic order by comparison to the outer core.

how do these highly pressurized crystals get flipped around by outer core fluids which are going to be kind of random, inconsistent and paramagnetically weaker ?

 

[geo101]

ere comes the columbo bit tho

Do that mean I am the killer?

Now we are getting into the territory of another thread...
https://www.physicsforums.com/showthread.php?t=518080"

For the core you can completely ignore any ferro/ferri/antiferro/para/dia magnetic effects. The geomagnetic field is generated by the geodynamo. Because I am lazy, I'll copy paste from the other thread

The geomagnetic field is generated by the fluid motion of the electrically conductive fluid outer core around the solid inner core. The fluid motion of the outer core is driven by both thermal and compositional convection (as Fe freezes out of the liquid lighter elements remain creating compositional buoyancy) and is dominated by large scale flow. The Earth’s rotation also plays a big role; it produces convection columns within the outer core that align along the rotation axis. So to answer your question, thermal and compositional convection along with the Earth’s rotation control fluid flow in the outer core... The flow of fluid is approximately axial-symmetric. The thermal and compositional convection is radial and the Earth’s rotation adds a helical twist to the fluid motion. [So the geomagnetic field aligns with the geographic poles, and not, for example the equator]

I guess what you are really asking is why does the geomagnetic field point north or south? First some background. As I mentioned the outer core is conductive and, in the presence of a magnetic field, electric currents will be produce inducing new magnetic. This is the basic premise of the self-sustaining dynamo, which is a big feedback system of convection to electric currents to magnetic fields, which then modify convection currents (through magnetohydrodynamics; http://en.wikipedia.org/wiki/Magnetohydrodynamics" ), and so on.

Now, imagine you have a convecting Earth-like core system in the absence of any magnetic field, i.e., convection without magnetic induction. Then a seed magnetic field was instantaneous “switched” on; the final stable configuration would depend on the interplay of the strength and direction of the seed field, the configuration of convection before the seed field (and how it changed through time, i.e., magnetohydrodynamics), magnetic diffusion through the solid inner core and its relative scale to that in the outer core (the inner core can act as a breaking system to changes in the magnetic field induced by convection in the outer core), controls on the heat flux through the core mantle boundary, and a host of other details. So those are (some of) the factors that control the orientation of the magnetic field.

So what causes the field to reverse? The short answer is that we don't really know, but basically major changes in the convective regime are likely to cause reversals of the geomagnetic field. The heat flux across the core-mantle boundary, which the main driving force behind the thermal convection, will play a key role in this.

 

[billiards]

So what causes the field to reverse? The short answer is that we don't really know, but basically major changes in the convective regime are likely to cause reversals of the geomagnetic field. The heat flux across the core-mantle boundary, which the main driving force behind the thermal convection, will play a key role in this.

How important is lateral heterogeneity in the mantle above the core mantle boundary to the geomagnetic field?

To me it seems "obvious" that the temperature gradients in the rocks in the lowermost mantle will control the flow of heat by conduction out of the core. One can envisage more heat flowing towards colder regions of the lowermost mantle (which are colder in the first place because they are the sites of downwelling in mantle convection). How important is this effect in controlling outer core convection -- given that this is to first order controlled by the Earth's rotation axis -- would the thermal gradients exert second order eddies? How important are eddy currents to the geomagnetic field?

Also hot and cold regions in the lowermost mantle seem to be quite permanent features, which to me suggests that heat flux across the core mantle boundary should not vary much in a short span of time. Unless a very slight perturbation in the heat flux can cause the field to reverse I would be surprised that this influence alone could be responsible for the flipping.

I seem to remember Glatzmeier modelling reversal as happening spontaneously. The geodynamo is inherently chaotic, I guess the flipping is an emergent property of the system, there is not a simple north/south switch that is being flicked. It is something less tangible to us.

 

[Dotini]

billiards, do you accept that sea-floor striping and the other evidences/anomalies constitute proof that the whole Earth magnetic field has flipped in the past?

Respectfully submitted,
Steve

 

[billiards]

billiards, do you accept that sea-floor striping and the other evidences/anomalies constitute proof that the whole Earth magnetic field has flipped in the past?

Respectfully submitted,
Steve

Yes I do.

Why do you ask?