The Measure of Soil Carbon by Loss on Ignition
The measure of organic carbon is determined by loss on ignition. The normal procedure is to prepare a 10g air dry sample by oven drying to 105 degrees C for 24 hours to remove all residual water. After cooling in a desiccator, re-weighed before being heated to 5000C for a further 24 hours to remove all the carbon. Cooled in a desiccator and re-weighed the carbon loss can now be calculated. The procedure however requires the ability to weigh to 3 decimal places and with the exception of laboratory grade fine scales this is beyond the abilities of most affordable scales.
However if a 100g air dried sample is used then to obtain the same degree of accuracy we only need to be able to weigh to 2 decimal places, something that is within the reach of most good quality and moderately priced electronic or spring balance scales. Thus the procedure is to first air dry sufficient soil to provide a 100g sample.
Prepare a crucible or similar ceramic vessel by placing into an electric oven for ½ hour at 105 degrees C (this removes any surface condensation), before placing it in a desiccator to cool (this prevents re-hydration).
Once cooled the crucible should be weighed (C1) and that weight recorded before placing 100g of air dried soil into it and the gross weight again recorded (C2)
The crucible and soil should now be heated in the electric oven at 105 degrees C for 24 hours before being removed and placed into the desiccator to cool.
Once cooled the now oven-dried sample should be removed and immediately weighed, again to avoid re-hydration (if you leave the sample for several hours and then re-weigh it you will find it has gained in mass as a consequence of re-hydration from the atmosphere). Similarly do not try to be smart and weigh the sample hot as it will weigh light! Trapped air within the sample (as much or more than 50% by volume) will have a lower density than the surrounding cooler atmosphere.
We now have a figure for the residual water content and this should confirm our estimates from the particle size distribution.
The sample now needs the organic carbon removed and this is burned off by heating the sample to 500 degrees C for a further 24 hours. Again it is allowed to cool in a desiccator before being reweighed to calculate the carbon loss as a consequence of ignition.
Thus this procedure requires being able to both weigh the sample accurately (to 2 decimal places) and to be able to heat it to 105 degrees C and 500 degrees C respectively. Achieving 105 degrees C should be possible using a domestic electric oven. A gas oven should be avoided as the combustion of natural gas results in the production of water which can affect the result.
Heating to 500 degrees C however is beyond the capabilities of most domestic ovens and a specialist oven (pizza, bread, etc) or pottery kiln is needed. Where this is not available a sand bath can be constructed and the crucible buried into the bath. The sand bath can then be heated over a gas flame and the temperature of the sand monitored and adjusted to keep it at or just above 500 degrees C. It is important to ensure that the temperature reaches 500 degrees C in order to burn off all the organic carbon (which begins to oxidise at 325 degrees C). Similarly the temperature should not be allowed to exceed 550 degrees C as this encourages the removal of inorganic carbon and leads to an over-estimation of the organic. Most of this inorganic carbon is held as CaCO3 and this hydrates completely to CaCO at 770 degrees C.
Furthermore this method is most accurate for sandy soils. Soils which have a high clay or sesquioxide content hold a ‘structural’ water content within that clay. This water is not removed by oven drying to 105 degrees C and is only released above 105 degrees C as the clays become de-hydrated. We are in-effect partially firing the clay and altering it both chemically and in terms of mass. In soils with a low clay content this is insignificant but in clay soils its significance can be considerable and lead to gross over estimation of the organic carbon content.
Thus whilst this procedure is relatively accurate over heating can lead to additional inorganic losses of carbon and high clay content to losses of structural water; both of which lead to over-estimating the organic carbon content.
Similarly great care must be taken to ensure that the sample does not re-hydrate between drying and weighing and a desiccator must be used. It is not however necessary to have an expensive heavy glass laboratory desiccator. One can be simply constructed using a glass or plastic jar or bowl, clean dry sand some cellophane or plastic bag, an elastic band, and a bag of silica gel crystals.
The vessel should be big enough to get your hand (if you use an oven glove) or your tongs into easily and to hold your sample(s) and a pot containing the silica gel crystals.
Fill the vessel with clean washed and oven dried sand (above 105 degrees C, no upper limit) to a level that makes it comfortable to work with and so there is at least 5cm from the top of the crucible (once placed onto the sand) to the lip of the dessicator vessel. The purpose of the sand is to prevent the heat of the crucible melting plastic or cracking glass on contact. However in deep vessels we can also use the sand to raise the bottom to a comfortable height!
Silica gel crystals can be purchased from garden suppliers who sell it as a water retaining gel for inclusion in container mixtures. The silica granules should be placed into a small metal, glass or ceramic bowl and dehydrated in an oven at 105 degrees C for 1 hour prior to being placed in the desiccator. The top should now be loosely covered with the plastic bag and this sealed with the elastic band. Leave the bag loose so that as the gel cools and contracts the air pressure inside the desiccator can balance with the outside. Your desiccator is now ready to use. The silica gel crystals will scavenge and remove any moisture from the sealed atmosphere in the desiccator and thus provide a moisture free environment in which your oven dried and ignition fired sample(s) can cool.
You should now have the mass of the crucible plus 100g of air dried soil (C1), the mass of this oven dried (C2) and the mass following ignition (C3)
The mass (and volume; 1g = 1cm3) of residual water = C1 – C2
The mass of carbon lost = C2 – C3
The mean average density or organic carbon is gcm3 and so the volume of carbon in the sample = (C1-C2) x density.
The density of soil is 1.6g cm3 so 100g occupies 62.5cm
Therefore the %v of organic carbon = volume of carbon / 62.5
we can confirm the density for the specific soil by taking a 100ml measuring cylinder, weigh and record the weight then fill it to just above the 100ml mark (approx 110ml) with air dried 2ml sieved soil. Tap the cylinder three times on a firm surface to settle the contents. This is a ‘precise method’ where it is more normal to use a 1.5ltr measuring cylinder however it can be adapted as long as the settlement method is maintained.
This method requires lifting the cylinder by approximately 3cm and allowing it to ‘drop’ onto the surface under force of gravity. It is repeated 3 times and then the level and mass are recorded. Subtract the mass of the empty cylinder from the mass of the filled and divide the product of this sum with the volume recorded. It should equal or be close to 1.6gcm3.