Cost and Performance Report: Soil Vapor Extraction at the Hastings Superfund Site, Well Number 3 Subsite Hastings, Nebraska

Table of Contents

• Executive Summary
• Site Information
• Matrix Description
• Treatment System Description
• Treatment System Performance
• Treatment System Cost
• Observations and Lessons Learned
• References
• Appendix A- Treatability Study

Back to Treatment System Cost | Table of Contents | Forward to References

Observations and Lessons Learned

Cost Observations and Lessons Learned

• Actual costs for installing and performing the SVE application, including disposal costs for the GAC, at the Well Number 3 Subsite were approximately $370,000, which corresponds to $620 per pound of CCl₄ removed (600 pounds CCl₄ removed) and $2.00 per cubic yard of soil treated.
  
• Actual costs were 17% less than originally estimated. According to the RPM, cost savings were realized in the following areas:

1. The SVE system worked better than expected, removed the contamination faster than expected, and was on-line about 2/3 of the time.
   
2. Savings were realized by using local construction contractors to provide oversight during the operation phase of the system. The involvement of the ARCS contractor was limited to phone conversations. Limited site travel was required during remediation phase. Costs were saved by utilizing a local chemist to perform chemical monitoring, and by utilizing a Region VII laboratory to provide off-site analysis.
   
3. The strong partnership, involvement, and commitment, between EPA and the State of Nebraska on this project allowed operating decisions to be based upon system performance and through an interactive decision-making process.
   
4. The ability to use one contract vehicle from design to project completion. One ARCS contractor designed the system and then performed project oversight. Subcontractors procured by the ARCS contractor performed well.
   

Performance Observations and Lessons Learned

• Soil vapor extraction met the remedial action cutoff extraction rate (0.001 lbs/hr) to remove carbon tetrachloride contamination at this operable unit within approximately 6 months of system operation. No CCl₄ rebounding effects were observed after a 2-month shutdown period.
  
• More than half of the contaminant removal occurred during the first 22 days of system operation.
  
• The RPM indicated that it is likely that the mass of VOCs removed by the system was greater than shown by the field results, based on the following information:

1. The EE/CA determined that 400 pounds of CCl₄ was estimated to be present;
   
2. Results from Westates Carbon determined that the three GAC canisters contained 19% VOCs (approximately 570 pounds of VOCs);
   
3. The use of 10-mL syringes to collect gas samples from the vacuum side of the system during operation (approximately 7 in. of Hg); and
   
4. The off-site confirmation testing which indicated that the on-site sampling had a negative bias of approximately 50%.
   

In addition, the RPM indicated that the on-site gas collection method, while quick and inexpensive, likely resulted in the dilution of the gas sample during sample collection. The RPM estimated that a total of approximately 600 pounds of CCl4 were removed by the SVE system at the Well Number 3 Subsite during both the treatability study phase and the remedial action phase.

Other Observations and Lessons Learned

• The full-scale system, designed based on the results of the treatability study, was implemented without modification. The treatability study results predicted that a 2-year operation period would be required to remediate the site. The full-scale system achieved the cleanup goals in less than a year.
  
• For the soils at Hastings, modelling was not reflective of system performance. The modelling predicted a radius of influence of 300 feet for the SVE wells; however, the actual radius of influence was found to be at least 1,500 feet for the system (based on an analysis of the source for additional contaminants removed by the SVE system).
  
• The on-site analytical protocol was not reflective of actual contaminant concentrations in the extracted soil vapors towards the end of the remediation (i.e., for lower concentrations of VOCs). For lower VOC concentrations, a larger sample volume is needed (e.g., a 6-liter SUMMA^TM canister).
  
• According to the RPM, the SVE system was sufficiently flexible to allow the sequential pumping of the system. Sequential pumping was desirable for the following reasons:
  
  1. When two or more wells are located close enough to each other that their areas of vacuum influence overlap, a small "dead zone" will occur where soil gas will not move toward either well; and
     
  2. After an extended pumping period, the rate of VOC diffusion from the soil or soil pore water matrix to the soil gas may become the limiting factor in the ability to remove VOCs from the vadose zone (the system becomes "diffusion limited"). In this case it is usually beneficial to stop pumping to allow time for equilibrium to be established between the VOCs in the soil/pore water matrix and the surrounding soil gas (i.e., rebounding).
     
• EPA issued the ROD for groundwater Operable Unit 13 on June 30, 1993, which required groundwater extraction and treatment. EPA initiated a groundwater 30-day treatability study in April 1994 utilizing monitoring well CW-1. Groundwater monitoring results indicated that the levels of CCl₄ found in monitoring well CW-1 varied from a high of 1,400 µg/L (one time), to levels between 100-150 µg/L prior to EPA’s treatability SVE action. Levels continued to drop and in June 1993 were approximately 20 µg/L. During EPA’s 39-day treatability study/pump test, using monitoring well CW-1, the levels of CCl₄ in the well averaged less than 5 µg/L. [22]

Back to Treatment System Cost | Table of Contents |Forward to References

References

22. Comments submitted by Diane Easley of EPA Region VII on February 9, 1995.