I niled Slides Hii\ ironnu'iK;il Pmieclion .\»enc\	Office »f Research iiikI l)c\clopmenl

National Kxposure Research Laboratory
Research Abstract

Government Performance Results Act (GPRA) Goal #2
Annual Performance Measure #565

Significant Research Findings:

Data on Solubilization and Mobility of Arsenicals from Iron Oxide
Residuals Produced by Drinking Water Treatment

Scientific	The implementation of the Arsenic Rule will require some drinking water utilities to

Problem and	implement treatment options to comply with the new maximum contaminant level (MCL).

Policy Issues	Two of the most widely used treatment strategies are iron co-precipitation and the

adsorption of arsenic on iron-based media. These treatments generate iron
oxide/hydroxide wastes containing arsenic. The waste material is classified as hazardous
or nonhazardous via the Toxicity Characterization Leachate Procedure (TCLP). The
wastes are commonly backwashed off treatment filters to settling ponds, streams and/or
disposed to a nonhazardous waste landfill (if it passes the TCLP). The mobility of the
arsenic in these cases is controlled by the oxidation state of the arsenic [i.e., As(III) or
As(V)] and the manner in which the arsenic is incorporated within the solids. Arsenic can
either be associated with the crystalline structure and be difficult to liberate, or adsorbed
electrostatically to the surface, where it can be liberated easily. Therefore, it is essential to
understand the conditions that can induce arsenic mobility in these solids. This knowledge
will aid in developing treatment and waste management strategies which maximize the
removal efficiency of arsenic while minimizing the mobilization of deposited arsenic
within the distribution system or its release back into the environment. To date, the ability
to estimate the oxidation state distribution of arsenic on a solid has been limited to X-ray
absorption near edge spectroscopy (XANES). This analysis, which is available in only a
few locations within the U.S., is costly and cannot distinguish between the arsenic bound
within the crystalline structure and the arsenic adsorbed to the surface via weak
electrostatic interactions. Thus, the goal of this research is to devise a cost-effective
analytical tool which can estimate the As(III) /As(V) concentrations on treatment solids
and, in turn, improve the ability to predict which water chemistries may generate increased
mobility.

The National Exposure Research Laboratory (NERL) has developed a sequential extraction
and analysis procedure to provide information on both the oxidation state of the arsenic
and the nature of the interaction between the removal media and the arsenic., The
sequential extraction procedure provides information on how the arsenic is incorporated
within the solid, while ion chromatography inductively coupled plasma mass spectrometry
(IC-ICP-MS) provides the As(III)/As(V) distribution associated with that part of the solid.

The two sequential extraction fluids chosen to estimate the ionically-bound As(III) and
As(V) were magnesium chloride and sodium phosphate, respectively, due to their
geochemical/environmental relevance. The magnesium chloride extraction fluid almost
exclusively extracts the ionically-bound As(III), which has a weak ionic bond with iron
particles, while the sodium phosphate predominantly liberates the more strongly ionically-
bound As(V). Because the selectivity of the extraction fluids is not 100%, the As(III) and
As(V) are speciated in each extraction fluid using IC-ICP-MS. The unextracted
As(III)/As(V) incorporated in the solid is estimated by utilizing an oxalate extraction fluid

Research
Approach


-------
followed by IC-ICP-MS analysis. Finally, the total arsenic that is available within the
solid is estimated via a nitric acid digestion of a second sample. This nitric acid digestion
procedure is similar to that used in solid waste methods for the dissolution of solids prior
to metal analysis. Using a mass balance approach, the sum of the ionically-bound arsenic
(in the magnesium chloride and the sodium phosphate extracts) and arsenic incorporated
within the solid (from the oxalate extract) is compared to the total arsenic (from the nitric
acid digestion) to estimate the percentage of arsenic speciated in the solid. Ultimately, this
technique improves upon the traditional XANES analysis because it provides an analysis
of the oxidation state of the arsenic while it differentiates between the more mobile
ionically adsorbed As(III)/As(V) and the As(III)/As(V) incorporated in the solid.

Results and	Significant findings of this research were as follows:

Impact	| j An automated sequential continuous flow extractor system was developed and utilized

to produce arsenic speciation results for iron-based drinking water treatment medias from
three different sources. The sum of the speciated arsenic results agreed favorably with the
results obtained from XANES. This is the first non-XANES technique capable of
providing arsenic speciation results on drinking water treatment media.

2.)	The sequential extraction results indicate that between 0.1 and 2.3% of the total arsenic
is liberated from treatment solids by magnesium chloride (the weakest extraction fluid).
This arsenic would be considered very mobile and under most environmental conditions,
this arsenic would be liberated. In all cases, the majority of the arsenic liberated by the
magnesium chloride was As(III). The sodium phosphate solution used to extract the
ionically bound arsenic from the media liberated between 15 and 43% of the total arsenic.
In all cases, the majority of the arsenic extracted by the sodium phosphate solution was
As(V). These findings are important for two reasons. First, while iron-based media
removed the arsenic from the drinking water, the amount of arsenic on the spent media that
is fairly mobile at near neutral pHs varies considerably from one media to the next: and the
amount of mobile arsenic would be significant for typical spent treatment media containing
4,000 parts per million arsenic. Second, the first extraction fluid removes almost
exclusively As(III), while the second extraction fluid removes almost exclusively As(V).
This finding may facilitate the use of a simpler direct analysis of the extraction fluids as a
means of estimating the ionically adsorbed As(III)/As(V) on a solid.

3.)	The arsenic liberated by the third sequential extraction fluid (oxalate solution)
dissolved the arsenic incorporated in the solid. This arsenic is far less mobile and would
require a relatively aggressive environmental condition to liberate it. The oxalate
extraction fluid liberated between 10 and 65% of the total arsenic from the drinking water
treatment media. This is significant because it indicates that some of the arsenic associated
with these solids could be considered relatively immobile. From a waste stabilization
perspective, this immobility translates into low risk with respect to future exposures.

Research
Collaboration
and Research
Products

4.) Although the initial demonstration of this technique focused on drinking water
treatment media, the technique could be used to estimate the mobility of arsenic from other
solids (e.g., iron oxide suspended solids and scales in distribution systems) by using water
chemistry changes likely to be utilized as treatment strategies for Arsenic Rule compliance.
This research has been a collaborative effort between NERL and the National Risk
Management Research Laboratory, which provided the drinking water treatment media.

U.S. EPA Publications

Creed, P.A., Schwegel, C.A., Creed, J.T. "Investigation of Arsenic Speciation on Drinking Water
Treatment Media utilizing Automated Sequential Continuous Flow Extraction with IC-ICP-MS
Detection." J. Environ. Monit2005, 7, 1079 - 1084.


-------
U.S. EPA Presentations

Creed, J.T., Creed, P.A., Schwegel, C.A. "Arsenic Mobility from Iron Oxide Solids Produced During
Water Treatment." Oral presentation at the 4th International Remediation of Chlorinated and
Recalcitrant Compound, Monterey, California, May 2004.

Parks, A.N., Gallagher, P.A., Schwegel, C.A., Ackerman, A.H., Creed, J.T. "An Investigation of
Arsenic Mobility from Iron Oxide Solids produced during Drinking Water Treatment." Poster
presentation at the Pittcon Conference, Orlando, FL, March 2003.

Future Research The initial demonstration of capability utilized iron-based treatment media previously
analyzed by XANES. An initial comparison between the continuous flow extraction
system and XANES was analytically very important because XANES is considered the
gold standard for arsenic speciation on solids. Future research will investigate the mobility
of arsenic on the smaller particle sized solids, resulting from iron co-precipitation, which
escape filtration and settle out in the distribution system. These solids can be mobilized by
hydraulic disturbances such as hydrant flushes or by water chemistry changes [e.g.,
hardness (ionic strength), pH, phosphate concentrations] that are brought on by new
treatment strategies or through the use of alternative source water due to the high cost of
treatment. The continuous flow extractor will be used to identify specific drinking water
chemistries that may mobilize arsenicals from these solids.

Questions and inquiries can be directed to:

JohnT. Creed, Ph.D.

U.S. EPA, Office of Research and Development
National Exposure Research Laboratory
26 W. Martin Luther King Dr.

Cincinnati, OH 45268-1564
Phone: 513/569-7833
E-mail: creed.jack@epa.gov

Funding for this project was through the U.S. EPA's Office of Research and Development,
National Exposure Research Laboratory, and the work was conducted by the
Microbiological and Chemical Exposure Assessment Research Division.

Contacts for

Additional

Information


-------