Perchlorate is a concern in drinking water because of its high solubility and mobility, known effects on thyroid hormone production, and treatment cost. The need for perchlorate treatment is especially acute in southern California’s Inland Empire region. The Inland Empire’s perchlorate plume is at least six miles long and impacts four cities’ water supplies, resulting in the impairment of 61,790 acre-feet (approximately 76.2 million cubic meters) per year of potable water.

As of 2010, ion exchange (IX) is the only perchlorate treatment technology fully approved by the California Department of Public Health (DPH) for drinking water; biological reduction using a fluidized bed bioreactor has only been conditionally approved. IX using a perchlorate-selective resin followed by resin disposal or destruction is the dominant market technology for perchlorate treatment. Integrated Ion Exchange (IIX) involves the regeneration of perchlorate-selective resin using tetrachloroferrate (FeCl₄^-) anion and then returning the resin to service. Perchlorate in the spent FeCl₄^- regeneration solution is subsequently destroyed or disposed of as a liquid waste stream. The application of IIX for the treatment of perchlorate will provide a more effective form of treatment and may reduce the amount of purge water required during resin installation.

The objective of this project was to demonstrate a reliable, more cost-effective method of treating low concentration perchlorate in drinking water supplies using IIX at an operating municipal water treatment plant. Specific objectives included demonstrating that regenerated resin could achieve California maximum contaminant levels (MCL) for perchlorate (≤6 μg/L) and that the regenerated resin effluent could maintain concentrations at or below MCL, secondary maximum contaminant levels (SMCL) for nitrate and Title 22 metals, and at or below notification level (10 ng/L) for nitrosamines. The performance of IIX was evaluated through several cycles to demonstrate the effectiveness and stability of the regenerated resin; the demonstration goal was to achieve 80-120% of virgin resin performance for treatment volume at breakthrough and perchlorate mass removal with regenerated resin.

This IIX demonstration project included four perchlorate loading cycles to saturation using a 150 gallons per minute (gpm) wellhead treatment unit and a single batch of perchlorate-selective resin. After the perchlorate loading cycle, the resin was regenerated at an off-site facility using FeCl₄^- solution. The modestly perchlorate-impacted fraction of the regenerant from each event was re-used in subsequent regeneration events without further treatment. A portion of the highly impacted fraction was used to evaluate chemical perchlorate reduction with ferrous chloride in a high pressure, high temperature pilot-scale reactor.

IIX produced water comparable to that produced with the baseline technology—single-use perchlorate-selective resin treatment. No degradation in perchlorate-selective resin performance or impacts from metals carryover was found with IIX through three regeneration cycles, despite re-using the FeCl₄^- regenerant. Two effluent water samples contained measureable volatile organic compounds (VOCs) at concentrations below the U.S. EPA MCL. The VOC source may have included influent contamination, regeneration reagent contamination, or residuals from regeneration facility construction. IIX was not found to increase nitrosamines or other semi-volatile organic compounds (SVOCs) in treated water.

Perchlorate loaded resin was regenerated in a multi-step process utilizing FeCl₄^- as the principal regenerant. The bulk of the perchlorate and nitrate were eluted within the first two bed volumes (BV) of FeCl₄^- regenerant. Pre-treatment procedures were able to remove naturally occurring uranium from the resin prior to perchlorate regeneration. Post-regeneration procedures were able to return the resin to service with no significant water quality impact on initial wellhead effluent following a regeneration cycle.

Parametric destruction tests of spent FeCl₄^- regenerant indicated pseudo-first order reduction of perchlorate using ferrous iron. Higher temperatures and residence times were associated with higher destruction efficiencies. Simulated perchlorate destruction tests routinely achieved greater than 95% destruction efficiency. In addition, high nitrate concentrations caused gas generation that led to process difficulties and VOCs were produced in destruction reactions.

There are several challenges that need to be addressed before IIX technology can be successfully implemented at sites. Presently, IIX technology is licensed to Calgon from Oak Ridge National Laboratory. Commercialization is expected to proceed using off-site regeneration at regional facilities contracted with lead times similar to new resin purchases. Operational differences between single-use resin and IIX will be isolated to the regeneration vendor and thus will not affect the utility. However, reductive perchlorate destruction requires further development prior to commercialization; the vendor is likely to initially commercialize IIX with incineration of the spent regenerant until the perchlorate destruction process can be optimized. Cost savings with this technology have been predicted, but they will ultimately depend on the market price of the regeneration service and resin purchases, as well as the sales/use tax treatment of these items.