vvEPA
                            United States
                            Environmental Protection
                            Agency
                         Environmental
                         Research Laboratory
                         Athens, GA30613
Research and Development    EPA/600/M-89/031    June 1990

ENVIRONMENTAL
RESEARCH    BRIEF
          Lanthanide Ion Probe Spectroscopy for Metal Ion Speciation
                                L.V. Azarragai and L.A. Carreiraa
Abstract
A  unique process  model  that intimately  involves  the
experimental and mathematical aspects of  metal-organic
interactions was developed. The technique, LIPSMIS (Lan-
thanide Ion Probe Spectroscopy for Metal Ion Speciation), in
its present form, provides for experimental  verification of
predictions of the quantitative values derived from modeling
organic interactions. Although it is cast in its present form to
treat binding of metal ions by humic substances, LIPSMIS is
sufficiently general that it can be modified to model metal
ion binding with inorganic anions or with sites on organic
substances other than humics.

Introduction
Research in metal-organic interactions at the  Environmental
Research Laboratory, Athens, Georgia  has the  prime
objective of providing the necessary scientific knowledge
with which to build decision-making tools for EPA's
regulatory offices. For example,  implementation of both the
Hazardous and Solid Waste and Superfund  Acts requires
tools for  evaluation of alternative waste management  and
treatment technologies, based on potential  human  health
and environmental impacts. These tools include exposure
assessment models for estimating the fate and transport of
toxic metals to either an environmental or human receptor.
Currently, the prime model for this purpose is MINTEQA2, a
thermodynamic equilibrium  model for prediction of metal
Speciation, and thus of metal mobility.

Dissolved organic material (DOM), e.g., humic substances,
are an  important component of most surface waters and soil
systems,  and even occur to a  significant extent in some
 1 Environmental Research Laboratory, Athens, GA

 2 Department of Chemistry, University of Georgia, Athens GA
                      aquifer solids. Toxic chemicals, such as metals, may  bind
                      with DOM, leading  to mobilization  if  the  DOM  is  in a
                      dissolved <3r colloidal state. MINTEQA2 does not contain a
                      term representing the interactions of metals  with naturally
                      occurring organic materials.

                      The long  range goal of our research is  to develop a
                      quantitative term (a mathematical  process model) to
                      describe metal-natural organic binding. However, before this
                      is  possible, it is necessary to  develop techniques to
                      experimentally measure the strength and extent of metal-
                      organic  interactions  under the  influence of environmental
                      variables  such as  pH,  ionic  strength,  and metal  ion
                      competition for the various binding sites of the DOM without
                      severely  perturbing the state  of the system as these
                      measurements are made.

                      Experimental
                      The LIPSMIS technique  is based on  the  unique
                      fluorescence properties of the trivalent europium ion, Eu(lll).
                      The basis of this technique is the existence  of a
                      hypersensitive transition associated with the fluorescence
                      emission  spectrum of the Eu(lll) ion. The intensity of the
                      ~616nm  emission band of  the Eu(lll) ion  is extremely
                      sensitive to ligation, whereas  the intensity of the ~592 nm
                      emission  band is not. This results in  an increase in  the
                      intensity of the 616  band relative  to  the 592 band as the
                      Eu(lll) metal ion is bound to  a  ligand site, for example, a
                      carboxylic site on a humic  material. This  intensity ratio
                      (Ratio =  L2/|616) can then be quantitatively related to the
                      concentrations of the  free  and  bound  metal species.
                      Therefore, the change in this ratio as a function  of  total
                      metal added is essentially a  spectroscopic titration of the
                      metal binding sites of the humic (other) material present.

                      Two experimental problems had to be addressed, however,
                      before this type of measurement could be deemed viable.

-------
First, the cross section for spectroscopic absorption for the
Eu(lll) ion is vanishingly  small (~10au  as compared  to
10000 for most organics). This was solved by exciting the
Eu(lll) emission with a high peak power, low average power
pulsed dye laser operating at the Eu(lll) absorption line at
~394nm. Second, at this wavelength  the humic  material
fluoresced with very high efficiency. The humic  emission
could be up to 106 times as strong as the Eu(lll) signal at
low Eu(lll) concentrations. To overcome this large mismatch,
the spectral  properties of Eu(lll) were used to separate the
signals.  The fluorescence lifetimes of most organics are in
the nanosecond  range, whereas the lifetime  of the nearly
forbidden Eu(lll) transitions are in the millisecond range. By
using a combination of phototube  pulsing and  signal  time
gating,  the  Eu(lll)  fluorescence  signal  could be easily
extracted  from the  extremely intense  humic  emission.
Details of the experimental setup  are discussed by Dobbs
and coworkers1.

A Model for Metal-Organic Binding

The present metal-organic interaction model is based on the
continuous distribution model used for proton binding1 with
humic  substances. This distribution is characterized by
three descriptor variables: CL, the total  concentration  of
metal or proton binding sites, n, the mean  Log K, value and
o, the breadth of the distribution of K, values. K, is the cation
- organic binding constant, a measure of binding strength.
The effect of competitive binding was incorporated into the
model by using a form of the competitive Langmuir equation
such  that all  cations  (hydrogen  ions  and metal  cations)
compete for the same distribution  of ligand binding sites2.
The effect of ionic  strength of the aqueous medium on
binding was found to be easily modeled by expressing the
activities of  all species in equilibrium. The simple Davies
model was found  to be satisfactory for calculating activity
coefficients.
Although the binding  of Eu(lll) to  humic materials is not of
particular environmental  interest,  it  can  be  used  as  a
sensitive  probe  for  measuring  effects  of  pH  and  ionic
strength on metal binding. When  a second  metal, e.g.,  a
toxic environmental pollutant, is added, the competition for
sites between  Eu(lll) and this metal can be monitored as  a
displacement  of Eu(lll)  and concomitant  increase  in  the
measured fluorescence intensity ratio

Effect of pH

Figure 1  shows the effect  of  pH  on metal binding with
Suwannee River  fulvic  substances.  The  symbols  are
experimentally  observed values and  the solid lines  are
calculated from the metal organic interaction model
The effects of  pH are most pronounced in  pH ranges below
5. As  the pH is decreased, the probe metal is released from
the  humic substance, and the fluorescence ratio increases.
The complexation capacity of the humic material is a strong
function of the  pH of the aqueous medium

Effect of Ionic Strength

Figures 2 and  3 show the effect of ionic strength on metal
binding with  Suwannee  River  dissolved organic  matter
(SRDOM) at pH 3.5 and pH 6 2
In both cases an  increase  in  ionic strength  results in  a
release of metal ion and an increase in  the fluorescence
ratio.  The effect is most pronounced at lower  pH.

Effect of metal competition

Figure 4 shows the effect of Pb(ll) ions competing with the
europium probe ion for humic binding sites.
The symbols represent observed values and the solid lines
are calculated from the model.  The  competitive  binding
model developed  here  allows  the LIPSMIS technique to
serve as a general  tool for measuring the binding strength of
any metal that competes with the probe metal for  binding
sites on the humic  material.

Conclusions

Comparison of the experimental data with model predictions
shows that humic-bound  metal ions can be released rapidly
as free metal ions as a result of changes in acidity and ionic
strength in  the aqueous medium. LIPSMIS  has  yielded
                        Eu (III) Titrations
     1.2
                                              -2.5
 Figure 1.  Ratio is the fluorescence rate = |592/|616 • Cm is the
          total metal added, x = pH of 2.5. * = pH of 3.0.  +
          = pH of 3.5. Humic concentration of 20 ppm.
                                                              20 E-1
                                                                              NaCIOandO.lM  pH=3.5
                                                                           -6
                                -4

                             Log Cm
                                                                                                          -2
 Figure 2.  Humic concentration 20 ppm. O = no NaCI x
          0.1 M NaCI pH = 3.5.

-------
   25 E-1 T
   ra
   o:
        0 ;
                   NaCI 0 and 0.1M  pH = 6.2
                 -6
                                -4
                             Log Cm
                                                -2
Figure 3.  Humic concentration 20 ppm. 0 = no NaCI x =
          0.1M NaCI pH = 6.2.

reasonable values tor the humic complexation capacity, CL
(1.5  x 10-3 meq/g of SRDOM). Addition of environmentally
important metal ions  such as Pb or Al results in competition
for binding  sites with the probe metal and an observable
release of Eu. The ability to model and measure competitive
metal binding allows  this technique to determine the binding
strengths  of metals  that are difficult if not  impossible to
measure by other techniques, e.g. Al(lll). To  date  we have
measured the mean Log K, values of three metals: pEu = 6.4,
u,Pb = 4.8, and HAI = 55.
For  more  immediate application,  portions of  the  model,
which have been verified to the extent shown in this paper,
will be incorporated  into MINTEQA2  to allow an estimation
of metal-natural  organic  binding,  thus  providing  a more
accurate prediction of metal speciation and fate.
                                                            20
              Lead 0 and 2E-4M  pH=3.5
  o
  a
  
-------