Sampling Molten Mixture in Tank

[waynef]

I have a mixture of two chemicals coming off of a column to a holding tank. This is batch process where the stream is predominately Chemical #1 initially and toward the end of the run Chemical #2 predominates. I need to sample the stream so that I know the ratio of the two chemicals in the tank at the end of the run. This material will be fed back into the process and used in the next batch.

The issues are that the material is at 400 F, has a melting point of approx. 350 F, and is under 30 mmHg of vacuum.

I could sample directly out of the piping from the column. I believe I would need at least 3 samples during the batch run. Assuming that the transition from #1 to #2 is non-linear, an averaging of the results may not give an accurate representation of the ratio of the two chemicals.

The material has a short residence time in the 150 gal. tank of about 20 mins. I'm afraid that there may not be sufficient diffusion in this short time frame. There is no ability to agitate this tank. I'd like to sample from the tank because I'd only need one sample.

I'm also concerned that the material will solidify in any sampling apparatus.

Any suggestions would be sincerely appreciated!

Wayne

 

[butelja]

A ram type sampling valve, such as manufactured by Strahman or Fetterolf can be steam jacketed and will work in applications like this without freezing. However, you still have the personnel hazard of handling 400 F material.