Analytical Methods:Lab Mag Sus

SASSA Home Page &#8658; Analytical Methods Home Page &#8658; Laboratory Techniques &#8658; Magnetic Susceptibility

Magnetic susceptibility: laboratory analysis
The magnetic susceptibility of a material is a measure of its ability to become magnetised by an external magnetic field (Dearing, 1999).

Questions
The magnetic susceptibility of a soil reflects the presence of magnetic iron-oxide minerals such as maghaematite; just because a soil contains a lot of iron does not mean that it will have high magnetic susceptibility. Magnetic forms of iron can be formed by burning and microbial activity such as occurs in top soils and some anaerobic deposits. Magnetic iron compounds can also be found in igneous and metamorphic rocks.

The relationship between iron and burning means that magnetic susceptibility is often used for:
 * Site prospection, to identify areas of archaeological potential prior to excavation.
 * Identifying hearth areas and the presence of burning residues in deposits (Tite and Mullins, 1971).
 * Explaining whether areas of reddening are due to burning or other natural processes such as gleying (waterlogging).

The relationship between soil formation and magnetic susceptibility means that it can also be used to:
 * Identify buried soils in depositional sequences.
 * Identify redeposited soil materials in peat, lake sediments etc.

Sampling
Each sample should be at least 10 cm3 and if possible plastic, rather than steel, implements should be used for sampling. Samples should be taken from a representative part of the context, being careful not to bulk together different deposits. Due to high background variability replicate samples are also a good idea. Samples should be taken from the 'natural' parent material and from throughout the profile; not just from the context of interest.

Samples can be measured wet or dry. If samples are to be stored before measurement they should be air-dried and sieved to less than 2 mm. It is important that samples are not heated to more than 40oC (Collinson, 1983; Walden, 1999). Typically samples are packed into 10 cm3 plastic pots to standardise the volume measured.

Analysis
The most commonly used instrument is the Bartington MS2, although other systems are available. The methods outlined here assume the Bartington system is being used.

When setting up the meter should be kept away from an metal objects, and the transformer and mains cable; it should be away from all electronic devices and sources of vibration. The temperature of the room should be kept constant and the meter should be switched on at least 10 minutes before the first measurement is made (Dearing, 1999). It is also good practice to make continuous air measurement in the 0.1 range to check for drift and sources of interference.

Low frequency measurement (0.46 kHz)
Low frequency measurements are the most common analysis made and would normally be made on all samples before carrying out any high frequency or fractional conversion measurements (see below).


 * 1) Weigh a 10 cm3 plastic pot, pack it with prepared soil and reweigh to find the mass (weight) of soil used (W).
 * 2) Before starting the measurement, ensure the meter is switched to low frequency (LF) and zero the meter by pressing the Z button.
 * 3) With the meter empty press the M button to take an air measurement (M1).
 * 4) Lower the sample into the meter and press M again to take a sample measurement (sample &kappa;)
 * 5) Remove the sample and with the meter empty take a second air measurement (M2)

Magnetic susceptibility is usually expressed as volume specific (&kappa;), or mass specific (&chi;).

If samples are analysed in pots of known volume the measurement is known as volume specific magnetic susceptibility (&kappa;). Volume specific magnetic susceptibility is unitless (SI units) and are measured as 10-5 SI. &Kappa; should be corrected for drift and background interference as follows:
 * Volume-specific magnetic susceptibily

&kappa; (corrected)= sample &kappa; - ((M1 - M2)/2)

However, mass specific magnetic susceptibility (&chi;) is a better measure as it accounts for differences in sample density and makes measurements more comparable. Bulk density is mass (weight; usually around 10 g) divided by volume (10 cm3 if the standard pots are used). However the SI units for &Chi;LF are usually 10-6 m3 kg-1. To get the right answer when the 10 cm3 pots have been used:
 * Mass-specific magnetic susceptibility

&chi; 10-6 m3 kg-1 = (&kappa; / Mass) / 10

where &kappa; is the corrected value and Mass is the weight of soil in grams.

High frequency measurement (4.6 kHz)
By measuring the magnetic susceptibility of the sample at both low and high frequencies (0.46 and 4.6 kHz) it is possible to detect certain types of grains (superparamagnetic ferrimagnetic crystals) that are created by certain soil processes. Switch the meter to high frequency (HF) and repeat the measurement of the sample in exactly the same way as for the low frequency measurement.

Frequency dependance is the difference of the low frequency (LF) and high frequency (HF) measurements expressed as a percentage of the low frequency measurement. To calculate the percentage frequency dependant susceptibility (&Chi;fd%) of a sample:

&Chi;fd% = ((&Chi;LF - &Chi;HF) / &Chi;LF) x 100

Core measurements
The Bartington meter has a special coil for measuring the magnetic susceptibility of core samples. This can be useful for directly detecting the inwashing of soil materials to peat or lake sediment cores for example.

Cores are fed through the meter, and measurments taken at set measuring points taking care that the core is not touching the sensor whilst the measurement is being made. Air measurements should be taken before inserting the core and again once the core measurements have been made.

Because the mass of the sample is unknown the results are expressed as &Kappa;.

Nb. measurements may be reduced in the first or last 10 mm of the core as both the core and air are being measured (Dearing, 1999).

Fractional conversion
Fractional conversion is the ratio between low frequency magnetic susceptibility (&Chi;LF) and maximim potential suscptibility (&Chi;MAX) and it provides a measure of the magnetic susceptibility enhancement of a sample, which can be very useful in interpreting the low frequency magnetic susceptibility (Crowther, 2003). Because of the time taken to measure fractional conversion only a sub-set of samples are usually analysed (Crowther and Barker, 1995).

To measure &Chi;MAX low frequency magnetic susceptibility is first measured, then samples are mixed with approximately 5% wt flour and placed in lidded crucibles, these are heated in a furnace to 650oC for 1 hour (reducing conditions). The lids are then removed and the samples are heated at 650oC for a further 45 minutes (oxidising conditions). Once the samples have cooled to room temperature the low frequency magnetic susceptibility is remeasured (&Chi;MAX).

Fractional conversion (&Chi;CONV) is calculated as:

&Chi;CONV = &Chi;LF / &Chi;MAX

Data and Interpretation
The magnetic susceptibility of a sample depends on the mixture of minerals it contains and their individual magnetic behaviour. The different types of magnetic behaviour are:

Soils contaminated with iron from sampling tools etc. will have a very high magnetic susceptibility signature becuase of the presence of ferromagnetic minerals.

Igneous rocks contain ferrimagnetic minerals such as magnetite with a high magnetic susceptibility; as a result soils in igneous areas may have high and sometimes noisy background magnetic susceptibility levels, which can make it hard to identify subtle changes due to soil formation for example. Burning soil can result in the formation of maghaematite and other minerals (Le Borgne, 1960), which also give high magnetic susceptibility measurements. You may also get high values associated with the formation of iron sulphide compounds such as greigite (Fe3S4) in waterlogged anaerobic conditions.

Less pronounced enhancements in magnetic susceptibility are also associated with topsoils thought to result form the effects of bacterial and redox processes. Antiferromagnetic minerals such as haematite and goethite are also common in soils.

Paramagnetic and Diamagnetic minerals and materials are also common in soils and have a very low magnetic susceptibility; these minerals may dilute the effects of more magnetically susceptible materials in the sample.

Frequency dependence measures the contribution of very small ferrimagnetic minerals known as superparamagnetic (SP) particles. These SP minerals are able to contribute to magnetic susceptibility at low frequencies, but not at high frequencies. The percentage difference between these two measure (%FD) is related to the proportion of SP minerals in the sample. The magnetic susceptibility of many topsoils are affected by the presence of SP particles (Dearing et al. 1996); hence topsoil may have a high frequency dependance. The burning of soil, though not necessarily fuel combustion, also result in the formation of SP particles and high frequency dependence.

Potential magnetic susceptibility shows the maximum magnetic susceptibility that burning a sample of soil may achieve. Fractional conversion, therefore, is a measure of the actual enhancement resulting from burning (Crowther, 2003).

Related Techniques

 * Field magnetic susceptibility
 * Magnetic remanence