On the possibility that problem of the stickiness of the soil might be due to the magnetic particles, the Phoenix team has done some experiments with Mars soil simulants that contain some proportion of magnetic particles so you would think this problem would have already been seen in these simulations if that were the reason. Perhaps though there was a higher proportion of magnetic particles than expected in the actual Mars soil.
I did a web search on how you can demagnetize a permanent magnet and found this:

Magnet.
5 Magnetization and demagnetization.
"Permanent magnets can be demagnetized in the following ways:
Heating a magnet past its Curie point will destroy the long range ordering.
Contact through stroking one magnet with another in random fashion will demagnetize the magnet being stroked, in some cases; some materials have a very high coercive field and cannot be demagnetized with other permanent magnets.
Hammering or jarring will destroy the long range ordering within the magnet.
A magnet being placed in a solenoid which has an alternating current being passed through it will have its long range ordering disrupted, in much the same way that direct current can cause ordering."
http://en.wikipedia.org/wiki/Magnet#...emagnetization


The TEGA instrument will heat the sample to high temperature, but of course this can't be used to remove the magnetization if you can't get the sample to the instrument in the first place.
The second method of demagnetizing could conceivably work, but I doubt the Phoenix lander has a magnet that could be passed over a soil sample. A variation on this might be to rub and mix around the sample on itself, then the different magnetic orientations on the particles might tend to cancel each other out.
The third possibility might also be feasible by striking hard on the sample to demagnetize it. The scoop might be used for this purpose. My guess though is that you would have to place the sample on a hard surface to do this. A flat metal surface on the lander would work but there might be a worry that this could jar the landers internal electronics by doing this. Perhaps the sample could be placed on top of a hard rock that the robot arm could reach.
I don't think the fourth method mentioned of passing an alternating current over the sample is feasible since I doubt the lander has the capability of doing this to a sample that is outside the lander.


Bob Clark