What are some considerations that need to be made when doing an extraction of solid sample like soil to water?

Document ID

Document ID TE6734


Version 3.0


Status Published

Published Date

Published Date 08/23/2018
What are some considerations that need to be made when doing an extraction of solid sample like soil to water?
Water extraction considerations when trying to extrapolate results of water analysis to a solid sample.
There are some main concerns that need to be made when considering a water extraction of a solid sample like soil.
1. The bias caused by solvent water. This can be resolved in most cases by using a solvent water that is free of the analyte to be measured, but this may be impossible for some analytes. For example pH or conductivity/resistivity, the solvent water will have it's own natural value that will be a factor in the reading of the extracted sample.
2. Non-aqueous components of the sample. There may be some compounds of the sample that would add to the result of solids test that are not picked up in a water extraction because they are non-aqueous.
3. Recovery of aqueous components. Not all of the aqueous components will be extracted. This can be minimized by appropriate homogenization, but there will always be some aqueous analyte that does not go into solution.
4. Loss from filtration. Many tests that would be done on the extractant will require that the homogenized sample be filtered to remove the remaining solids. This filtration will also cause a loss to the recovery.

With consistent procedure and technique water extraction can often be used as a trending tool when looking for changes, but these factors will make the actual result low accuracy.

If a sample is sent to an outside lab, the lab results can be compared to the results of the extraction of the same sample to calculate a recovery factor.
With several replicates, and a good standard deviation, this recovery factor could be used with extraction results to provide a more accurate estimation of the sample.
If the standard deviation is not good, this shows that there is too much variability in the factors listed above to get reproducibility from an extraction.

Was this answer helpful?

Thank you for your feedback!
There was an error with your submission. Please try again.