SAMPLE PROPERTIES

Not all processes or tanks contain simple homogenous material, very often they are a mixture of phases (e.g., liquid-gas, liquid-solids). If the material is left for a significant time in storage or is in low velocity pipework, there may be a risk of separation of phases, precipitation, deposition, crystallisation or the release of dissolved gases.
 
It should be recognised that the objective for process plants is to create larger size water droplets either through use of chemical treatments or process and flowline design to encourage coalescence and the dropout of produced water. The production of finely dispersed water droplets for the purposes of representative sampling is at variance with the objectives of good process design and separation.
 
Good sample point installation will assist in the mixing of phases by locating the sample point after a flow disturbing event (e.g., a bend in the pipework). Sampling probes will also improve sampling accuracy if installed and maintained correctly.

SAMPLE POINT DESIGN

Chemists are very rarely consulted on the design and installation of sample points and are only recruited when operations are about to commence. Therefore, if reference has not been made to relevant International Standards during the design and build, there may be a risk that (i) the samples taken will not be representative, and (ii) the analytical results will not be an accurate indication of the composition of the bulk material being produced. Any actions taken based upon incorrect results may cause further inaccuracies in the process.
 
Once a plant or process is installed, it will be extremely expensive and time consuming to redesign or rebuild, so it’s better to get it right first time!

CRITERIA FOR SAMPLING REPRESENTATIVITY

  • The distribution of liquids should be uniform across the whole cross section of the line and the diameter of the entry port into the sample probe should be significantly larger than the maximum water droplet size that it is trying to capture.
  • The flow should remain representative throughout the sampling period.
  • To obtain an analysis result that is representative of the flowline, the sample must be maintained in the same condition as it was when taken from the sample point without loss of liquid, solids or gases.
  • If a sample is to be sub-divided, this has to be carried out in such a way as to ensure each has exactly the same composition as the original sample. A sample must be totally homogeneous before sub samples are drawn from it, otherwise the analysis result cannot accurately reflect the characteristics of the point from which it is taken.

 
If any one of the four main criteria for representative samples is not met, it is very likely that the sample cannot be classed as representative.

Oleochem Nulab Training have been running sampling and fiscal analysis courses for over twenty years both internationally, and now at our new specialised training centre in Bridge of Don. Our instructors have many years’ experience working in laboratories not only in the oil industry, but various chemical plants and industrial laboratories.