United States
                    Environmental Protection
                    Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
                    Research and Development
EPA/600/S2-87/020 May 1987
SEPA          Project  Summary
                    Dewatering  of  Dilute Aqueous
                    Hazardous  Wastes  Using
                    Reversible  Gel  Absorption
                    Walter J. Maier and Edward L Cussler
                     The feasibility of using crosslinked
                    gels in a reversible process for extract-
                    ing pure water from aqueous waste so-
                    lutions has been investigated. It has po-
                    tential for conentrating waste streams
                    that contain hazardous chemicals. Near
                    critical gels have been developed that
                    swell and collapse as a function of proc-
                    ess conditions. At low temperature or
                    high pH, these gels swell by absorbing
                    water. During swelling, they exclude
                    larger solutes, including many haz-
                    ardous materials. Solutes and colloidal
                    pollutants thus may be concentrated to
                    facilitate treatment and disposal.  The
                    gels are made from water soluble
                    monomers; they can be easily regener-
                    ated by warming or by adding acid to
                    release the purified water. The  col-
                    lapsed gels are separated  by filtration
                    and reused.
                     In this study gels have been con-
                    tacted with solution containing a wide
                    range of different molecular weight sol-
                    utes and colloids to determine water
                    extraction capacity and efficiency of
                    solute rejection.
                     This Project Summary was devel-
                    oped by EPA's Hazardous Waste Engi-
                    neering Research Laboratory* Cincin-
                    nati, 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 infor-
                    mation at back).

                    Introduction
                     Recently published studies have re-
                    ported the use of water soluble gels to
                    achieve essentially the same type of
                    separation that can be obtained with
                    membrane  processes. The physical-
chemical concepts that underlie gel ex-
traction are related to the separation
mechanisms encountered in molecular
size ultrafiltration membrane separa-
tions, e.g., size exclusion, and ion selec-
tive membrane separations, e.g., ion ex-
clusion. However, operational problems
that stem from the need for high pres-
sure and concentration polarization are
overcome in gel separation because of
the large surface area of the gels. Fur-
thermore, the previous research on gels
indicated that a  reversible gel absorp-
tion process could be developed.  This
would allow reusing gels in a  continu-
ous process that meets the operating
criteria listed above. Research on the
development and use of gels was there-
fore undertaken.

  This reserach had two objectives; the
first was to develop near critical water
soluble polymer gels which would swell
and collapse in aqueous solutions  with
minor changes of temperature or pH;
the second was  to evaluate the effec-
tiveness of these gels for concentrating
dilute solutions of hazardous wastes.
  The gels were made of substituted
crosslinked  acrylamides.  One of the
more successful temperature-sensitive
gels is a polyisopropylacrylamide that
was made by polymerization of N-
isopropylacrylamide. pH-sensitive  gels
were made by polymerization and  sub-
sequent hydrolysis of acrylamide or by
co-polymerization of acrylamide  and
sodium acrylate. With these two gels in
hand,  laboratory tests were set up to
define mechanisms by which the  gels
worked, how much water they can im-
bibe and release, and what types of sol-

-------
ute molecules and colloids would be re-
jected from entering the gel phase.
  The theoretical  background that ex-
plains the behavior of gels relates to be-
havior of solutions near a phase transi-
tion. Phase transitions are illustrated in
Figure 1. In the upper left-hand corner of
this figure, the phase diagram for a reg-
ular solution is shown as a plot of tem-
perature vs. solute mole fraction x2. The
phase diagram is shaped like a
parabola. Above this parabola, the sys-
tem is a  single phase; under the
parabola, it is two phases whose com-
positions are exemplified by the ends of
the horizontal dashed line. When a sys-
tem of known concentration is cooled, it
can initially be one phase and later  be-
come two phases, as suggested by  the
vertical dashed  line.
  The phase distribution of a regular so-
lution as  shown  by the parabolic  do-
mains is a special case of solution  be-
havior. The more general case is
illustrated in the upper right-hand cor-
ner of Figure  1; it is the phase diagram
for water-triethylamine. As  before,  the
phase diagram consists of a plot of tem-
perature vs. composition, which is  ex-
pressed as a volume fraction fa. Like a
regular solution, this system forms one
phase at very high temperatures. Unlike
regular solutions it also  forms one
phase at low temperatures; in between,
it forms two phases. When a solution of
water-triethylamine containing one
phase is cooled, it again follows a locus
like the vertical dashed  line. At high
temperature  it initially contains one
phase. As it is cooled, it  forms two
phases. As it  is cooled further,  it forms
one phase again. The behavior of a reg-
ular solution and of water-triethylamine
is often discussed in terms of critical
points, or, as they are more commonly
called in  liquid  solution, consolute
points. These  consolute  points are
shown as the  heavy  black dots in Figure
1. A regular  solution forms only one
consulate  point, called an upper conso-
lute point.
  These same concepts  have been  ex-
tended to describe the behavior of
water soluble  polymers. For un-
crosslinked polyacrylamides, the phase
diagram is shown in the lower left-hand
corner of Figure 1. This phase diagram
agains plots temperature vs. composi-
tion  expressed  as volume fraction  2.
The phase  diagram for the un-
crosslinked polyacrylamide, shown by
the lower of the two curves, looks like a
regular solution, which has been bent
strongly to the left. This bending is a
    Regular Solution
     Polyacrylamide
                                                           02
              Reversible Gel
                                                              Crosslinked
                                                                 Polymer
              02
Figure 1.    Phase diagrams.
           T = temperature; 02 = volume fraction
                     02
consequence of the polymeric  chains,
which imply a solute of dramatically dif-
ferent molecular weight than the sol-
vent.
  However, if the polyacrylamide is
crosslinked like our gels then the phase
diagram is like the upper curve. At low
temperature, the upper curve is similar
to the lower curve for an uncrosslinked
polymer. However when 4>2 is small,
that is, when the polymer is dilute, the
uncrosslinked polymer dissolves com-
pletely in  the solvent. The crosslinked
polymer swells to a maximum value. As
a result a  crosslinked polyacrylamide
never fully dissolves, and its phase dia-
gram shoots upwards.
  The polymers used in this study can
be represented by the superposition of
the phase diagram for the crosslinked
polyacrylamide and the phase diagram
for triethylamine-water. At low temper-
atures, the phase diagram is like th
shown in the lower left-hand corner, b
at high temperatures the phase diagra
follows  the schematic shown in tl
upper right-hand corner. Overall beha
ior is characterized by a phase diagra
like that shown in the lower right-hai
corner of Figure 1.
  Gels that have a phase diagram li
that in the lower right-hand corner
Figure 1, would be expected to  she
temperature related volume changes
described in the experimental results
this report. In particular, when the gel
warmed, 2 will get slightly smaller ai
the gel volume will get slightly biggi
However, at a critical higher tempei
ture, 2 will get much bigger and t
gel volume will get much smaller.
many cases, this volume change wi
temperature can be abrupt.  For exai
pie, the  crosslinked polyisoprop

-------
 lacrylamide polymer described in this
 report showed a 5-fold volume change
 over a temperature  increase  of only
 0.1 °C.
   These models were used as a theoret-
 ical framework for analyzing and corre-
 lating data  on volume changes of the
 gels. The models also gave insight on
 development of new gels and the phys-
 ical configuration of the gels.

 Conclusions
   The overall  objectives of the project
 were to demonstrate the feasibility  of
 using temperature- or pH-sensitive gels
 to separate water from solutions to con-
 centrate them. The gels are effective for
 concentrating  solutions of macromolec-
 ular solutes, and hence are a commer-
 cially attractive alternative to ultrafiltra-
 tion. The gels  are less  effective for
concentrating small solutes. The overall
separation process of a temperature-
sensitive gel for  extracting water,
thereby concentrating the residual solu-
tion, is shown in Figure 2. In the first
step,  solution  to be separated is con-
tacted  with a bed of  temperature-
sensitive gel  beads. The gel beads
swell, absorbing water, while excluding
macromolecular solutes.  The swollen
gel and the concentrated raffinate are
then separated by  filtration.  The con-
centrated  raffinate becomes the
product. When swollen  gel  itself  is
warmed, it collapses,  releasing the ab-
sorbed  solution. The collapsed gel  is
again removed from this solution by fil-
tration, producing  an  extract that con-
tains little if any macromolecular solute.
The warmed collapsed gel is reused by
cooling which starts the cycle over
again.
  Three types  of gels have been tested
to determine their ability to concentrate
different solutes.  Typical results are
given in Table  1 for three gels. The first
gel.  A, is a non-ionic, temperature-
sensitive gel. Gel B is an ionic, tempera-
ture-sensitive gel. The third gel, C, is an
ionic, pH-sensitive gel. The  first two
columns of the table describe some of
the solutes that have been successfully
concentrated using at least one of the
three gels. The separation efficiencies
reported in  the  last three columns of
Table 1 are  calculated by dividing the
concentration change actually observed
by that expected from the gel volume
change. For example, a 100% efficiency
means the solute is completely rejected
by the gel while 0% efficiency would re-
sult if the solute moved freely into the
gel with the water.  100% efficiencies
were never observed because traces of
solution are trapped between gel parti-
cles when the swollen gel is removed by
filtration.
  Table 2, arranged parallel to Table 1,
shows several lower molecular weight
solutes that are separated less effi-
ciently. Both tables are consistent with
the idea of the gel as a polymer mesh;
larger  solutes are screened out while
smaller solutes are not rejected as effec-
tively. More complete listings of chemi-
cals are presented in the full report.
  If we now look at a pentachlorophenol
(PCP) on row four of Table 2, we notice
a discrepancy  in  the efficiencies re-
ported for the three gels. At high pH,
PCP is anionic and is effectively concen-
trated by the ionic gels, but not by the
non-ionic gel. At low pH, PCP is non-
ionic and is not separated by either gel.
Rejection of anionic species was also
observed with other solutes. It repre-
sents a second mechanism of exclusion
that stems from the ionic character of
pH sensitive gels. The rejection of ionic
solutes has been described quantitively
in terms of a modified Donnan  equi-
librium model analogous to membrane
separations.


Recommendations
  The use  of gels for extracting  rela-
tively pure water has numerous possi-
ble practical  applications for treating
wastewaters. Potential applications in-
clude concentrating aqueous solutions
to facilitate reuse-recycle of products,
reducing volumes to facilitate storage
and  ultimate disposal, and extracting
water to eliminate the need for dis-
charging wastewater. However, further
      Add
     Solution
     le) Cool the Gel

  The gel is now ready
      for reuse.
                                 (a) Gel Swells

                                   Solvent is
                                  preferentially
                                   absorbed
                       (b) Filter
                       Separate gel
              (d) Filter

      Separate gel particles
        from the extract
                     (c) Warm the Gel

              ''The gel collapses releasing
                  absorbed solution.
                                        Figure 2.    A separation process based on temperature sensitive gels.

                                                                               3

-------
research and development are needed
to fully exploit this separation technol-
ogy.
  The dynamics of swelling and con-
traction should be studied. The work de-
scribed in the full report was designed
to analyze and describe water extrac-
tion capacities and rejection efficiencies
at equilibrium. The time rate of change
of gel  expansion and contraction has
not been studied. It has been observed
that gel volume changes  are very fast
for small  aggregates of gel. However,
rates of water movement and exchange
of ions by larger aggregates of gel
should  be studied  in order to define
mass transport limitations. A better
understanding of the effects of gel ag-
gregate size is needed in order to opti-
mize process design conditions for con-
tacting  during the expansion  and
contraction cycle.
   Process studies to evaluate the effects
of staging should be carried out. Single
stage extractions that involve large dif-
ferences in concentration between gel
phase and raffinate are thermodynami-
cally unfavorable. Staged  or sequential
extraction would be more efficient and
allow for more complete rejection  of
solutes from the extracted water phase.
   Modified gels should be developed
that combine the swelling and contrac-
tion characteristics of temperature-
sensitive gels with the solute rejection
characteristics  of the ionic gels. Such
gels would be capable of rejecting ionic
species as well as separating by molec-
ular size. There are several advantages
to using temperature cycles instead of
pH changes to extract water; energy
costs are low, there is no need for chem-
ical addition, there is no increase in salt
(ions) concentration. Another approach
that should be pursued is grafting  re-
verse  osmosis membranes onto the
outside  of the gels  themselves. Such
grafting should provide the additional
selectivity necessary for separating low
 molecular weight  species like pen-
tachlorophenol. Use of gels for other
 chemical  systems  including alcohol,
 water and linear vs. branched alkanes
 should also be investigated.
Table 1.    Solutes Successfully Separated by Gel Extraction
Solute
Albumin
Hemoglobin
Gelatin
Polyethylene
Oxide
Blue Dextran
Polystyrene
Latex
Virus*'
iVHsidfUtai
Weight
45,000
64,500
65,000
600,000

2,000,000
—

—
A
97
—
98
97

97
95

80
B
84
—
97
84

99
96

—
C
9.
9
-
-

—
8.

8;
 *A = n-lsopropylacrylamide Gel
  B = Diethylacrylamide/Acrylic Acid Copolymer Gel
  C = Hydrolyzed Polyacrylamide Gel
**A = Avian Flue Virus:  80%
  C = E. Coli: 85%
Table 2.    Solutes That are Less Successfully Separated by Gel Extraction
Solute
Vitamin B-12
Polyethylene Glycol
Sucrose
Pentachlorophenol
Urea
Molecular
Weight
1355
342
267
60

A
30
400 10
18
2

B
15
5
86**
3

C
6
93'
0
 "A = n-lsopropylacrylamide Gel
  B = Diethylacrylamide/Acrylic Acid Copolymer Gel
  C = Hydrolyzed Polyacrylamide Gel
 **Pentachlorophenol gave high separation efficiencies at pH above 7.5 but separation efficiei
  cies were poor at lower pH levels because negative charge is neutralized by protonation.
   Walter  J. Maier  and Edward L.  Cussler are with  University of Minnesota,
    Minneapolis, MN 55455.
   Mark J. Stutsman is the EPA Project Officer (see below).
   The complete report, entitled "Dewatering of Dilute Aqueous Hazardous Wastes
    Using Reversible Gel Absorption," (Order No.  PB 87-168 761/AS; Cost:
    $13.95, subject to change) will be available only from:
          National Technical Information Service
          5285 Port Royal Road
          Springfield, VA22161
           Telephone: 703-487-4650
   The EPA Project Officer can be contacted at:
          Hazardous Waste Engineering Research Laboratory
          U.S. Environmental Protection Agency
          Cincinnati. OH 45268

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
    BULK RATE
POSTAGE & FEE!
       EPA
  PERMIT No  G
Official Business
Penalty for Private Use $300

EPA/600/S2-87/020
            0000329   PS

            U S  EIIVIR  PROTECTION AGENCY
            REGION 5 LIBRARY
            230  S DEARBORN STREET
            CHICAGO              IL   60604

-------