Take a known quantity of sample (approx. 1gm), weigh it accurately, place it in a crucible, heat it for 1 h at a minimum ob 550 degree centigrade, cool it in a desiccator and weigh it again. The percentage difference between initial weight and present weight is the loss on ignition and is expressed in %
Even better: first dry the powdered sample at 110 degrees to get rid of any adsorbed water. And I would think that 550 degrees is a bit low (I typically go for 800-900 degrees), but if there is a possibility of alkali-loss, then you may want to keep the temperature lower than that
If your goal is to link LOI to the amount of organic matter in the sample, be careful not to loose your carbonate as well. They go at a higher temperature. (check literature for recommended temperature /time settings to avoid that).
Christoph's answer is very valid. It completely depends on the goal you want to achieve how you have to do your LOI. I was assuming you were dealing with an igneous rock, and wanted to get rid of the volatiles before doing major element analysis by XRF or ICP-OES. But if you were dealing with a sedimentary rock of which you want to determine the content of organic matter, then you should remain at 550 degrees, rather than frying the material at 800-900 degrees as I suggested.
In LOI experiments, we normally measure the amount of volatile in our samples. Commonly these volatiles are CO2 (e.g. carbonate rocks, organic materials) and H2O (mica, amphiboles, clay minerals). Therefore, depending on the type of rock that you are going to analyze (mineralogical composition), the temperature of ignition is variable. If your target is organic material, you have to burn the sample at about 500 oC for 1 h. If your rock is carbonate in composition you have to burn the sample up to about 900 oC. Note that some carbonates tend to decompose at about 500 oC as well. In each case you can use normal furnace and you need about 1g of powdered sample.
In case of whole rock analyses and major, and trace elemental analyses of mineral separates you can go for XRF (X-Ray Fluorescence analyzer). In XRF analyses you can obtain the LOI by measuring the crucible weight after fused bead preparation (here the temperature is about 1100 oC).
Further, the volatile components (CO2 and H2O) can analyze by high-temperature combustion methods as well.
Thanks all for your contribution. I was asking this question for about two weeks because, I forgot the magical word "LOI" unfortunately.
I am not worrying about organic matter. Only about H2O and or CO2. I am analysing metamorphic rocks. So as I understood, its better to go 110C first and then unto 900C.
Organic carbon and carbonate contents of black shales from the lower Cretaceous Paja Fm. (Colombia) by loss on ignition and CHNS analysis: comparison of methods
1)At first you have to take a sample-container in a weight machine and weight that container and write down what is weight of container. i.e called empty Cruicible in gm.
2) Then powder sample should be taken carefully and put the powder on the sample-container/cruicible.
3) After measuring weight carefully; then note down the what is the weight of sample.
Then add the empty sample container and powder sample weight individually.
After Ignition
1) After Ignition weight them again carefully and write down the weight of the total sample container with powder sample.
2) Then subtract them and then calculate the percentage
I heat at 105 Celsius to remove adsorbed water and 950 Celsius to remove both organic matter and carbonate. Like you now already know, what you want to get rid off determines the temperature of ignition.
Can I further request to provide what is the significance to measure the LOI in rock phosphate in view of using it for manufacturing of phosphoric acid ?
Can someone clarify this for me. I did XRF analysis of oxides and elements and when I sun the oxide percentage it gives me 💯 percent. Then how do I know LOI? I checked many papers and have seen LOI in their table of values