Soil Spiking for a Representative ISS Treatability Study



Soil Spiking for a Representative ISS Treatability Study

Presenter:  Paul Jansen (GEI Consultants, Inc.)
Authors:  Paul Jansen, Youness Sharifi, Shu Xu, Christopher Dailey and Currie Mixon (GEI Consultants, Inc.)


A critical aspect of in-situ solidification (ISS) bench scale treatability study is to obtain representative soil samples for bench scale treatment.  A common issue encountered during bench scale ISS treatability study is the loss of volatile organic compounds (VOCs) due to volatilization during treatability laboratory soil mixing.  Literature studies have shown that as much as 90% of VOCs are lost with less than two minutes of soil mixing.  Volatilization of VOCs may bias the soil contaminant concentration low which results in false-positive in the mix design selection. Since in most former Manufactured Gas Plant (MGP) sites, benzene and naphthalene are the main contaminants of concern (COC); it is important to retain these compounds for leachability studies. For the bench scale treatability study, GEI collected soil samples from the site at proposed treatment intervals and homogenized the soil samples.  However, these homogenized samples were only able to retain BTEX and naphthalene (BTEXN) concentrations a few orders of magnitudes lower than the theoretical upper confidence limit (UCL 95) calculated based on the soil analytical results. In order to evaluate leaching, GEI spiked the soil with laboratory prepared concentrated BTEX solution and naphthalene-methanol solution to 200%, 300% and 400% of targeted concentrations. GEI initially introduced BTEXN prior to adding treatment reagents to the soil mix. The result of this spiking method reported a good degree of homogenization with low BTEXN retention rate.  GEI modified this spiking method by introducing BTEXN after the soils have already been homogenized. The modified spiking method reduced the volatilization due to the mixing time.  However, the low Relative Standard Deviation (RSD) reported indicated poor homogenization.  In conclusion, spiking the soil with BTEXN solution did not produced a reliable soil composite that is representative of the upper level soil impact.  To obtain a soil sample that is representative of the upper impact the sampling of ISS soil could consider other methods, e.g., sampling biased towards the source area.  This approach could provide more realistic results for a bench-scale study though using worst case source material   tends to overestimate the concentration of grout reagents needed during full scale to meet physical performance criteria (hydraulic conductivity and strength) and groundwater impacts downgradient of the site.  This presentation discusses the delicate balancing act between creating a representative treatability study and meeting long term performance goals of the full scale remedy.


To view the full Poster Presentation List for MGP 2017 view this link:

Developed by Autograph Media Group