Limitations and potential of commercially available rhodamine WT as a groundwater tracer

We conducted chemical characterization, batch, column, and modeling studies to elucidate the sorption and transport of rhodamine WT (RWT) in the subsu rface. The sand-pack material from the Lizzie field site near Greenville, N orth Carolina, served as our porous media. Our study confirms earlier resul ts that RWT consists of two isomers with different sorption properties. It also shows that the two isomers have distinct emission spectra and are equa lly distributed in the RWT solution. The presence of the two isomers with d ifferent sorption properties and distinct emission spectra introduces an er ror in measuring the RWT concentration with fluorometers during porous medi a tracer studies. The two isomers become chromatographically separated duri ng transport and thus arrive in a different concentration ratio than that o f the RWT solutions used for fluorometer calibration and test injection. We found that this groundwater tracer chromatographic error could be as high as 7.8%. We fit six different reactive-solute transport models of varying c omplexity to our four column experiments, A two-solute, two-site sorption t ransport model that accounts for nonequilibrium sorption accurately describ es the breakthrough curves of the shorter-timescale column experiments. How ever, possibly due to the groundwater tracer chromatographic error we disco vered, this model, or a similar one that accounts for a Freundlich isotherm for one of the solutes, fails to describe the RWT transport in the longer- timescale column experiments. The presence of the two RWT isomers may compl icate the interpretation of field tracer tests because a shoulder, or any t wo peaks in a breakthrough curve, could result from either aquifer heteroge neity or the different arrival times of the two isomers. In cases where iso mer 2 sorbs to such an extent that its breakthrough is not recorded during a test, only isomer 1 is measured, and therefore only 50% of the injected m ass is recorded. Isomer 1 of RWT can be accurately modeled with a one-solut e, two-site, nonequilibrium sorption model. This conclusion and the results from our batch studies suggest that RWT isomer 1 is an effective groundwat er tracer but that the presence of isomer 2 hampers its effectiveness.