Carbolite Muffle Furnace vs. CEM Microwave Furnace for LOI

Because loss-on-ignition (L.O.I) can be so time consuming in routine analysis of soils, peats, lake or marine sediments, we’ve been testing a CEM Pheonix microwave furnace as a substitute to the old muffle furnace methods.

Determining Ramp, Cooling Times and Batch Sizes

The muffle furnace takes around 45mins to get to 550deg.C, and cooling is a three-step process, where the furnace cools naturally to around 250deg.C, then transferring the crucibles to a heat sink plate to cool to 50deg.C, then transferring to a desiccator until room temperature for weighing. This three-step process takes about an hour and a half to two hours. A single batch in the muffle furnace consists of 64 samples.

The microwave furnace takes around 20mins to get to 550deg.C, and cooling is pretty much instant because of the quartz fibre crucibles. A single batch is up to 15 samples.

Determining Ignition Time

For loss-on-ignition, the standard ignition time is 2 hours. For the microwave furnace, I performed a test to determine the most appropriate time, using a mixed peat that had been powdered and dried overnight at 105deg.C. Sample sizes were around 0.3g dry weight.

5mins = 95.4% L.O.I @ 550deg.C – the sample was observed to be igniting when removed from the furnace

10mins = 95.8% L.O.I @ 550deg.C

20mins = 96.4% L.O.I @ 550deg.C

30mins = 95.4% L.O.I @ 550deg.C

60mins = 96.5% L.O.I @ 550deg.C

Testing Comparability

As an initial test I tested the same peat used in the test above, with three repeats. Sample sizes were around 0.7g dry weight.

Muffle furnace = 95.165+/-0.09g L.O.I @ 550deg.C

Microwave furnace = 95.367+/-0.69g L.O.I @ 550deg.C

I’m planning further analyses to compare with results derived from elemental analysers, muffle and microwave furnaces, but these initial tests indicate the methods are likely to be comparable.


Both units provide comparable results. The muffle furnace has a total batch time of around 4hrs45mins, and a maximum batch size of 64, whereas the microwave furnace has a batch time of under 35mins, and a batch size of 15, so a 64 samples could be analysed in under 3hrs. The microwave furnace has the added advantage of integrated balance and L.O.I calculations, and fewer hot surfaces, particularly when using the recommended quartz fibre crucibles. There is no reason why the microwave method wouldn’t be suitable for ignitions of 950deg.C for estimating inorganic carbon content.