[thedoc (OP)]

Wilf James asked the Naked Scientists:
   
One of my sort of hobbies is demolishing astronomical howlers. I often hear about the Earth's magnetic field being produced by moving metal within the Earth. I don't think that the moving metal theory for the Earth's magnetism is valid for two reasons:

Firstly, the metal concerned is NOT magnetic because it is too hot at a temperature well above the Curie Point for all known magnetic materials. This can be checked with a kitchen science experiment. The items needed are a strong button magnet of the sort used at the end of a telescopic pickup tool, and a small selection of small iron objects Medium sized screws and nails will do, preferably with an ordinary     paper clip. A screw or a medium sized nail is attatched to the magnet by     magnetism. The paper clip is attached to the screw or nail in the same way.
Then, as many nails as will stick to the paper clip are attached in a chain to the paper clip.

Then using a cigarette lighter, the paper clip is heated until it is red hot. The nails hanging from the paper clip will fall off. They can be easily re-attached when the paper clip has cooled.

This experiment shows that iron loses its magnetic properties when it is red hot. The metal within the Earth is very much hotter. The paper clip is used because it will heat up quickly in a lighter flame.

This experiment shows that the metal within the Earth is too hot to  be magnetic.

We now come to the second reason why the metal within the Earth does not create a magnetic field. A person at the equator is going around the Earth at just over 1,000 miles an hour. The metal at 2,000 miles depth below the equator is thought to be more or less fluid and travelling around at 500 miles an hour. The axis of the Earth's magnetic field is very near the spin axis. The metal is spinning around the centre of the magnetic field. At any given radius from the magnetic axis the strength of the magnetic field is more or less constant. Only changes in magnetic strength can induce currents in conductors. As the magnetic field will be unchanging for most of the metal, no significant electric currents will be induced in it. The molten metal at 2,000 miles depth will only be convecting at a very low speed when compared with its rotational speed. The electric currents induced by convection will be small and partially counteracted by Lenz's Law. An indirect way to test this is with a button magnet and a metal spinning top. The magnet is attached to the centre of the top.
The top is spun at a known rate and timed to find out when it topples. Then a piece of unmagnetised metal the same size and weight of the button magnet is attached to the top. Then the top is spun and timed as before. There should be no significant difference in the time taken for the top to topple. This will show that the magnet does not slow down the top's rotation.

The next tests are with a toy gyroscope with a non magnetic rotor. The gyro is spun up to a known rotation speed with and without the button magnet held near its rotor.
The magnet will slow down the rotor quickly, showing that an off-centre magnet is generating a current in the rotor.

The two experiments with the top and the gyroscope show that a centred magnet does not generate a current that causes drag. The magnetic field at the centre of the Earth does not cause any significant currents to be induced in the molten metal spinning around it.

There is a magnetic field caused by the ring current around the Earth. The ring current is made of of positively charged particles from the solar wind that are dragged around the Earth eastwards by the Earth's rotation. The particles must be positively charged because only a positive current can produce a magnetic south pole in the vicinity of the Earth's geographic north pole. A south magnetic pole is needed to attract the north ends of navigational compasses. This explanation for the Earth's magnetic field provides a possible answer for the repeating changes in the polarity of the Earth's magnetic field. The sun is presumably emitting predominantly positively charged particles at the moment. Over time this will cause the sun to develop a very strong negative charge that will inhibit the emission of positively charged particles. Then it could begin to emilt predominantly negatively charged particles and cause the ring current to be negative. Then the Earth's magnetic field
will be reversed. The magnetism of the rocks on either side of the mid atlantic ridge provides a sort of record of the changes in the Earth's magnetic field. Presumably the sun emits charged particles of one polarity or the other at irregular intervals on a geological scale.

I hope that you will find what I have written interesting

Wilf James BSc
What do you think?

[evan_au]

Studying the source of Earth's magnetic field is a complex problem in magnetohydrodynamics - a chaotic domain requiring complex calculations on the fastest computers - and even small deviations in the conditions can result in large changes in the result.

I think that one factor ignored in the above analysis is that fluid flow in the molten core is driven by randomly changing convection cells. This induces varying changes in currents in Earth's core, which causes random changes in the Earth's magnetic field.

The solar wind does cause currents in the ionosphere and in the Earth's rocks. Because the electromagnetic force is so incredibly strong compared to gravitational forces, every cubic kilometer of solar wind will be approximately electrically neutral. So I don't think it is correct to assume that the Sun can spend long periods emitting mainly positive particles (protons), followed by other long periods emitting mainly electrons.

Laboratory experiments to emulate Earth's dynamo have produced oscillating magnetic fields in molten sodium (as mentioned in the OP, the liquid sodium is not ferromagnetic, merely conductive). This is initiated by a small "seed" magnetic field, and sustained by rotational cells in the molten sodium. It does not rely on patterns of particles emitted by the Sun.

The sun's oscillating magnetic field is thought to be generated by a similar mechanism to Earth's dynamo - only in the Sun, the conductive fluid is a hydrogen/helium plasma, not liquid nickel/iron. As far as we know, there is no influx of charged particles from the galactic center which is driving the Sun's approximately 22-year magnetic reversal cycle.

[alancalverd]

The liquid core doesn't need to be magnetic to generate a magnetic field. What matters is that it is electrically conductive. Moving charge (i.e. an electric current) deep inside the planet produces a magnetic field that extends through the crust and into space.

[AndroidNeox]

Electrical conductivity isn't necessary to generation of magnetic fields. To produce a field one needs to have net charge flow. If the magnetic field is generated by core rotation then there must be a mechanism separating charges, in bulk, at the core.

[guest39538]

My opinion from an amateur scientist perspective and looking at it theoretical, it is believed it is something to do with earth's inner/outer core, I would have to agree this is a starting point , I would also have to consider several other things though, light, thermodynamics, photon electrical effect, entropy, the mantle and electricity. 

If I remember correctly , if we electrified something, such as a pin, the pin becomes magnetised, so on that basis, if the earth's core produces an electrical field, or the ''ground'' by the suns rays internally produces a photon electrical effect. then any metal content of the earth would become magnetised to a monopole or dipole. The mantle could be the ''magnet''. 



And a weird thing, even magnets have mass and are attracted to the ground.

[chiralSPO]

And a weird thing, even magnets have mass and are attracted to the ground.

That's not weird. Very simple rule that is always true: If it's made of atoms, it has mass.

As far as the source of Earth's magnetic field:

As pointed out above, the temperature of the core doesn't matter because it is behaving as an electromagnet, not a ferromagnet. The fact that it is iron has nothing to do with the magnetism, other than its conductivity. Jupiter's magnetic field dwarfs the Earth's (and Jupiter is so much further from the sun that if solar winds were responsible, it would be much, much less). It is theorized (though I do not believe proven) that a giant liquid metallic hydrogen core is responsible.

[guest39538]

And a weird thing, even magnets have mass and are attracted to the ground.

That's not weird. Very simple rule that is always true: If it's made of atoms, it has mass.

As far as the source of Earth's magnetic field:

As pointed out above, the temperature of the core doesn't matter because it is behaving as an electromagnet, not a ferromagnet. The fact that it is iron has nothing to do with the magnetism, other than its conductivity. Jupiter's magnetic field dwarfs the Earth's (and Jupiter is so much further from the sun that if solar winds were responsible, it would be much, much less). It is theorized (though I do not believe proven) that a giant liquid metallic hydrogen core is responsible.

I thought it weird because a person would think that if a dipole magnet has two polarities , pos and neq, that the Earth's magnetic field would repel one or the other with being magnetic fields. Clearly to me, it seems like the earth's ''electromagnetic'' field, is not really very magnetic compared to magnets, the two seemingly two different things altogether?

[Colin2B]

I thought it weird because a person would think that if a dipole magnet has two polarities , pos and neq, that the Earth's magnetic field would repel one or the other with being magnetic fields.

It does. This is how a compass works, the compass pointer is a bar magnet.

Clearly to me, it seems like the earth's ''electromagnetic'' field, is not really very magnetic compared to magnets, the two seemingly two different things altogether?

You can't decide 2 things are different just based on relative strength.
In fact the earth's field is quite strong, try this.
Put a compass on a table. Place the strongest magnet you can find on the opposite side of the room to east or west of the compass. Go see whether the compass still points north.
You will find most diy magnets have to be quite close to make the compass swing away from north, but the earth's magnetic pole affects it from 1000s of miles away.

[guest39538]

I thought it weird because a person would think that if a dipole magnet has two polarities , pos and neq, that the Earth's magnetic field would repel one or the other with being magnetic fields.

It does. This is how a compass works, the compass pointer is a bar magnet.

Clearly to me, it seems like the earth's ''electromagnetic'' field, is not really very magnetic compared to magnets, the two seemingly two different things altogether?

You can't decide 2 things are different just based on relative strength.
In fact the earth's field is quite strong, try this.
Put a compass on a table. Place the strongest magnet you can find on the opposite side of the room to east or west of the compass. Go see whether the compass still points north.
You will find most diy magnets have to be quite close to make the compass swing away from north, but the earth's magnetic pole affects it from 1000s of miles away.

Thank you Colin, I understand about  a compass and how a compass works, I have experimented with a compass and compass needle, my problem is yes the compass will turn to direct north and a corresponding direct south, but the compass does not turn to the ground, if there is a magnetic field being emitted from the ground, then why doesn't  my needle bend towards the ground, rather than pointing North or South?

[chiralSPO]

I thought it weird because a person would think that if a dipole magnet has two polarities , pos and neq, that the Earth's magnetic field would repel one or the other with being magnetic fields.

It does. This is how a compass works, the compass pointer is a bar magnet.

Clearly to me, it seems like the earth's ''electromagnetic'' field, is not really very magnetic compared to magnets, the two seemingly two different things altogether?

You can't decide 2 things are different just based on relative strength.
In fact the earth's field is quite strong, try this.
Put a compass on a table. Place the strongest magnet you can find on the opposite side of the room to east or west of the compass. Go see whether the compass still points north.
You will find most diy magnets have to be quite close to make the compass swing away from north, but the earth's magnetic pole affects it from 1000s of miles away.

Thank you Colin, I understand about  a compass and how a compass works, I have experimented with a compass and compass needle, my problem is yes the compass will turn to direct north and a corresponding direct south, but the compass does not turn to the ground, if there is a magnetic field being emitted from the ground, then why doesn't  my needle bend towards the ground, rather than pointing North or South?

The compass needle aligns itself in the N-S direction. Unless you are standing at one of the poles, this will not involve pointing at the ground.

[guest39538]

The compass needle aligns itself in the N-S direction. Unless you are standing at one of the poles, this will not involve pointing at the ground.

So it is only the poles where there is a magnetic field then ? 

[chiralSPO]

The compass needle aligns itself in the N-S direction. Unless you are standing at one of the poles, this will not involve pointing at the ground.

So it is only the poles where there is a magnetic field then ?

No. The field is everywhere. It is only at the poles where the field points at the ground.

[guest39538]

No. The field is everywhere. It is only at the poles where the field points at the ground.

Yes you said that one once, if the field is everywhere then surely the needle points everywhere? 



if the ground is emitting a magnetic field every where,  then surely the compass needle should point to the ground, I don't understand why it wouldn't , can you explain please?

[Colin2B]

Yes you said that one once, if the field is everywhere then surely the needle points everywhere? 

if the ground is emitting a magnetic field every where, 

no, the needle will point towards the poles along the lines of magnetic force. However, because the poles are created by the molten core they are under the surface. So at the magnetic poles the compass will point directly down. As you move away from the poles the compass will still point towards the poles so will be inclined downwards. Most compasses are balanced so they don't point downwards so you need an instrument like a compass on it's side to measure the inclination.
I remember measuring it at school and it was somewhere around 60 deg, but my memory fails so don't quote me.
At the equator it should be horizontal.

[DavidWoodhams]

Curie Temperature decreases linearly with pressure and with an inner core pressure of approx. 360 GPa it would be necessary to determine the Tc at that pressure.

The relative motion between the inner core and the liquid outer core would produce either a lead-synchronicity-lag, resulting in a field in one direction-no field-reversed field.

Taking the inner core rotation to be constant, the cause of relative variation in speed of rotation of outer core to inner core could be due to a variable balance of the earths rotation as a result of plate tectonics. This would result a variation of rotation of the outer core....but what force could increase its rotational speed? Could be our movement around the Sun, Moons orbit.....

[evan_au]

relative variation in speed of rotation of outer core to inner core could be due to a variable balance of the earths rotation as a result of plate tectonics.

There have been suggestions from seismic measurements that the solid core is rotating with respect to the Earth's surface - by about 1 degree per year. This could lead to a magnetic dynamo effect.
See: https://en.wikipedia.org/wiki/Inner_core#Dynamics

could be due to a variable balance of the earths rotation as a result of plate tectonics.

I am sure that convection plays a big part in the dynamics of the Earth's core, as does crystallisation, concentration gradients  and melting.

These same forces play a part in plate tectonics.

It depends how closely coupled convection in the outer core is to convection in the Earth's mantle.
 

Could be our movement around the Sun, Moons orbit.

This I find less likely. The Earth is in free fall around the Earth/Sun/Moon barycenter, so the actual gravitational attraction won't affect the Earth's core.

Tidal effects will impact the core, but the core has smaller diameter than the Earth's crust, so tidal influences will have smaller amplitude than they do on the oceans. More importantly, tidal effects will produce alternate tugs in different directions every 12-13 hours. These will cancel out when the viscosity and momentum of the Earth's outer core suggests convection patterns that may last for many millions of years.

[syhprum]

Not all studies of the Earths magnetic field use computer simulations some analogue devices have been built for the purpose.

https://www.universetoday.com/14664/how-do-you-model-the-earths-magnetic-field-build-your-own-baby-planet/

[evan_au]

This scientist suggests that the Earth's solid core may have started forming "only" 500 million years ago...
....around the time of the Cambrian explosion, although he hesitates to suggest that the magnetic field was a contributor.

He says that there was a magnetic field prior to that - they can tell from magnetism frozen in rocks. But the field got rather weak around 500 million years ago.
Listen (11 minutes):  https://www.sciencefriday.com/segments/the-earths-core-might-be-younger-than-scientists-thought/

[yor_on]

Actually the Russians made 'analogue' computers to study fluid systems at the time they were embargoed from getting  'new fangled' computers. They worked very well. I'll see if I can find it later.
=
A little too early perhaps :)
Still.  http://www.digitaljournal.com/article/338106

[Bored chemist]

...it would be necessary to determine the Tc at that pressure.

Not really.
The Curie temperature has to be below the melting point, and we know that the deep core is molten.
However the Earth isn't a permanent "bar magnet" made of iron.
It's an electromagnet; so it doesn't matter what the Curie temperature is.

Actually the Russians made 'analogue' computers to study fluid systems at the time they were embargoed from getting  'new fangled' computers.

The oldest analog computers I know of were water clocks.
It's not clear how long ago they were invented but they certainly predate the existence of Russia as a country.