vc/EPA
                                   United States
                                   Environmental Protection
                                   Agency
                                   Municipal Environmental Researc
                                   Laboratory
                                   Cincinnati OH 45268
                                   Research and Development
                                   EPA-600/S2-81-082  July 1981
Project  Summary
                                  The  Equilibrium  Fluoride
                                  Capacity  of Activated Alumina
                                  Gurmderjit Singh and Dennis A. Clifford
                                    Because design limitations for a
                                  defluoridation process were unavail-
                                  able, laboratory studies were performed
                                  to establish the maximum (equilibrium)
                                  fluoride capacities for activated alumi-
                                  na as a function of pH and competing
                                  ions.
                                    Fluoride adsorption tests  were run
                                  using mini-columns containing 1 gram
                                  of acid-treated Alcoa F-1*  activated
                                  alumina equilibrated for 6 to 10 days
                                  with a continuous flow of fluoride
                                  solution at a constant pH. This proce-
                                  dure is very different from the usual 1 -
                                  hour batch equilibration  test and
                                  should give a more accurate estimate
                                  of the maximum fluoride-adsorption
                                  capacity of the alumina.
                                    Maximum fluoride adsorption ca-
                                  pacities in distilled water solutions of
                                  sodium fluoride were found to be
                                  8000 g FVm3 at 3.0 mg  FYL and
                                  9100 g FVm3 at 6.0 mg  FVL. The
                                  optimum pH range for fluoride adsorp-
                                  tion was 5 to 6. Based on data now in
                                  the literature and preliminary results
                                  from pilot studies,  less than 50 per-
                                  cent of these capacities are estimated
                                  to be attainable in actual municipal
                                  defluoridation practice because of
                                  poor kinetics and  competing ion ef-
                                  fects including silicate, which was not
                                  tested here.
                                    The alumina adsorbent preferred the
                                  common groundwater anions in the
                                  following order: OH~^>F~>SOr>Cr
                                  > HCO3. Fluoride was highly preferred

                                  *Mention of trade names or commercial products
                                   does not constitute endorsement or recommenda-
                                   tion for use by the U S Environmental Protection
                                   Agency
                                   over sulfate;  however, at a typical
                                   concentration of 250 mg SOijVL, the
                                   fluoride adsorption capacity was re-
                                   duced  28 percent at pH 6. Fluoride
                                   capacities were not reduced signifi-
                                   cantly by competition from chloride or
                                   bicarbonate.
                                    Although kinetics was not a  major
                                   objective of these studies, it was ob-
                                   served that as pH increased from 6 to
                                   8, the rate of fluoride adsorption also
                                   increased. Adsorption rate should be
                                   further studied.
                                    This laboratory research on fluoride
                                   removal was done in  support of field
                                   studies on the removal of fluoride with
                                   the use of activated alumina columns
                                   in a mobile pilot plant designed and
                                   constructed at the University of Hous-
                                   ton. The Mobile Drinking Water Treat-
                                   ment Research Facility was completed
                                   on April 9, 1980, and transported to
                                   its first field location—Taylor, Texas, a
                                   small community (population 13,000)
                                   with a  high-fluoride  (3.0 mg/L)
                                   groundwater supply. The mobile facil-
                                   ity containing activated alumina and
                                   ion exchange columns and reverse
                                   osmosis and electrodialysis units may
                                   be transported to any U.S. community
                                   with an inorganic contaminant problem.

                                    This  Project Summary was devel-
                                   oped by EPA's Municipal Environmen-
                                   tal Research Laboratory, Cincinnati,
                                   OH, to announce key findings of the
                                   research project that is fully docu-
                                   mented in a separate report of  the
                                   same title (see Project Report ordering
                                   information at back). A second project
                                   report by O. Clifford and M. Bilimoria

-------
 covering the design, construction, and
 operation of the Mobile Drinking
 Water Treatment Research Facility
 and a third report by D. Clifford, C. C.
 Lin, and M. Bilimoria covering the
 pilot-scale fluoride removal studies at
 Taylor, Texas, will be published at a
 later date.

 Introduction
   Packed beds of activated alumina
 granules have been used for about 30
 years to remove excess fluorides from
 potable water, especially at levels > 5
 mg/L current estimates are that there
 are several thousand public water sup-
 plies in the United States with excessive
 fluorides, especially in the 1.4 to 5.0
 mg/L range. Many of these communities
 may eventually install activated alumina
 treatment systems to remove the excess
 fluoride pursuant to the requirements of
 the 1974 Safe Drinking Water Act.
   When this research began, reliable
 process  design criteria for defluorida-
 tion of groundwater  using activated
 alumina were unavailable in the litera-
 ture. This was especially true for fluoride
Plastic Tubing-^
Tee Connection_
For Influent
Sample
Mini-Column-
Glass Wool—
 Pinch Clamp for—
 Flow Adjustment

2 L Beaker
For Effluent
                        20 L Fluoride
                        Solution
                        Reservoir
   Column
   Stand
 1 gm
- Pretreated
Alumina
Figure 1.    Typical  alumina  mini-
             column   apparatus   for
             adsorption runs. Repre-
             sents one often columns
             used simultaneously.
                                                10,000
                                                  3000
                                                  8000
                                           II
                                           o

                                           i<
                                                  7000
                                                  6000
                                                  5000
                                                                 2.0        4.0         6.0        8.0

                                                                       Liquid Phase Fluoride Cone. mg/L
                                                                                                              10.0
                                         Figure 2.
                                                     Fluoride adsorption isotherm. pH 6.0; temperature 22°C; distilled water
                                                     solutions of NaF; pH adjusted with
                                        concentrations less than 5 mg/L. There
                                        was no systematic treatment available
                                        of the effects of pH, competing ions, and
                                        fluoride concentration on alumina's
                                        capacity for fluoride at any concentra-
                                        tion. Because fluoride "adsorption" on
                                        acid-treated activated alumina is thought
                                        to be an ion exchange process, all these
                                        variables would be expected to influence
                                        capacity. Furthermore, the available
                                        literature data were  conflicting and
                                        obtained either from short contact time
                                        (1 -hour) batch equilibration tests or
                                        from actual municipal defluoridation
                                        plant operating  results.  Both  of these
                                        sources have serious disadvantages.
                                        Fluoride adsorption kinetics are known
                                        to be very slow, and the  effects of
                                        variable feed water  quality cannot
readily be studied in full-scale plants.
Also,  some of our preliminary experi-
ments indicated that it would take
several days of continuous exposure to
continuously flowing fluoride solution
to achieve 90 percent of equilibrium
adsorption capacity. Thus itwasdecided
to perform  long-duration  column tests
using feed solutions of varying fluoride
concentration, pH, and anion composi-
tion to reliably establish the effects of
these variables  on the equilibrium  ad-
sorption capacity of activated alumina.
The ultimate  objective was to develop
design limitations for municipal defluo-
ridation processes. The laboratory data
would also be used to plan pilot-scale
defluoridation experiments  in  various
U.S. communities.

-------
 Methods and Materials

  Alcoa,  28 x 48  mesh,  F-1  grade
 activated  alumina was conditioned by
 decanting to remove fines and washing
 exhaustively in a column with hUSOA-
 acidified,  pH 5, fluoride-free tap water
 before drying at 110°Cfor 48 hours. An
 alumina exhaustion run consisted of
 passing 10,000 to 30,000 bed volumes
 of fluoride solution through a  mini-
 column containing 1  gram of conditioned
 sulfate-form alumina over a period of 6
 to 10 days. A  sketch of the apparatus
 showing  one  of the 11-mm ID mini-
 columns is shown in Figure 1. Fifty runs
 were performed in  all. The  early runs
 were duplicated to  verify that, indeed,
 the procedure was reproducible. The
 fluoride capacity of  the exhausted alu-
 mina was determined by eluting the
 fluoride from the alumina over 4 hours
 using a 50-fold stoichiometric excess of
 caustic (100 ml of 1 percent NaOH solu-
 tion). This procedure recovered at least
 95  percent of the adsorbed fluoride.
 Regenerant solutions were analyzed for
 fluoride with the use of a fluoride-
 selective  ion electrode and a special
 buffer  to prevent interference from the
 dissolved  aluminum. Sulfate was deter-
 mined by  a BaCI2 turbidimetric method,
 bicarbonate by the inorganic channel of
 a TOC analyzer, and chloride by poten-
 tiometric titration  using a Ag/AsS
 electrode.
  Fluoride concentrations in distilled
water in the range of 0 to 10 mg/L and
pH in the range of 5 to 8 were evaluated.
 Minimal concentrations (< 0.006 meq/L)
 of H2S04,  NaaCOs, and NaOH were used
to adjust pH.  Individual competition
from sulfate, chloride, or bicarbonate
was evaluated in concentrations in the
 range  of 0.5 to 1 5  milliequivalents/L.
The effects of high ionic strength (high
TDS) were evaluated using solutions
containing the same number of equiva-
 lents of sulfate, chloride, and bicarbon-
ate at  total solution concentrations in
the range of 3.8 to 56 millimoles/L.
Reagent grade sodium salts were used
throughout the experiments.
  Fluoride analyses of column effluents
were made  periodically throughout
each run  to determine when a given
column was exhausted. Although pre-
cise flow  control was not  maintained,
the total volume fed to a given column at
a particular time was precisely  mea-
sured.  These data were used to deter-
mine average flow rates, which  varied
from 1 /2 to 4 ml/min.
 Results

   The effects of fluoride concentration
 on fluoride adsorption capacity at pH 6
 are shown in Figure 2; the effects of pH
 at constant fluoride concentration are
 illustrated in  Figure 3. Table 1  sum-
 marizes  both these effects on fluoride
 adsorption capacity.
   Figure 4 depicts the effects of varying
 concentrations of sulfate or chloride on
 the equilibrium fluoride capacity of the
 alumina  at pH 6 and constant fluoride
 concentrations (5.7 mg/L).  Finally, the
 time to  90 percent of equilibrium at
 various pH values and fluoride concen-
 trations  is depicted in Figure 5; for all
 the data  points, throughput (T = fluoride
 fed  to column/fluoride capacity of
 column)  was approximately 2.5.
                              Table 2 is a very useful compilation of
                            relevant conversion factors for design
                            purposes and literature comparisons. It
                            is included because the defluoridation
                            literature contains so many different
                            units expressing fluoride adsorption
                            capacity.
                              The full report was submitted in ful-
                            fillment of Grant No. R806073010 by
                            the University of Houston under the
                            sponsorship of the U.S. Environmental
                            Protection Agency.
         10,000
          8000
    u
    <0

      5   5000
          4000
          2000
                                        Fluoride
                                        Concentration
                                                           j_
Figure 3.
               4.0        5.0         6.0        7.0         8.0        9.0

                                           pH
Effects of pH on fluoride adsorption capabity. Distilled water solutions
of NaF; pH adjusted with Na2C03 and HzSO*

-------
Table 1.    Effects of pH and Aqueous-Phase Fluoride Concentration on Fluoride
           Adsorption Capacity*
                             Fluoride Adsorption Capacity in gm F /m3
                                           of Alumina
 Fluoride Concentration
         mg/L
               5.0
  Adsorbate pH
6.0          7.3
                                                                                250
         8.0
2.0
4.0
6.0
8.0
10.0
7390
8680
9360
9200
9395
7445
8605
9075
8965
9340
3030
4580
5510
6195
6645
2160
3280
3835
4250
4650
^Capacities are based on an alumina packed bed density of 55 Ib/ft3.

-------
G. Singh and D. A. Clifford are with the University of Houston, Houston,  TX
  77004.
Thomas J. Sorg is the EPA Project Officer (see below).
The  complete report,  entitled  "Equilibrium Fluoride Capacity of Activated
  Alumina," (Order No. PB 81-204 075; Cost: $8.00, subject to change) will be
  available only from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield. VA 22161
        Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
        Municipal Environmental Research Laboratory
        U.S.  Environmental Protection Agency
        Cincinnati, OH 45268
                                                                                     « U*. GOVERNMENT HUMTING OFflCC 1«1 -757-01Z/7216

-------
,                                                                                                           Postage and
United States                          Center for Environmental Research                                      Fees Paid
Environmental Protection                Information                                                           Environmental
Agency                                Cincinnati OH 45268                                                  Protection
                                                                                                           Agency
                                                                                                           EPA 335

Official Business
Penalty for Private Use $300

RETURN POSTAGE GUARANTEED                                         PS,



                                                                       CHlCA-t,  II

-------