Note: Descriptions are shown in the official language in which they were submitted.
CA 02235337 1998-04-17
W O 97/14652 PCTAUS96/16753
PARTIALLY CRYSTALLINE LAYERED SODIUM TITANATE
CA 0223~337 1998-04-17
W O 97/14652 PCT~US96/16753
BACKGROUND OF THE INV~NTION
(1) Field of the Invention
This invention concerns a novel partially crystalline sodium nonatitanate composite having
S a layered structure. The novel sodium titanate exhibits ion exchange properties, and it is
particularly designed to be an excellent ion-exchanger for strontium.
(2) Description of the Art
Ever since nuclear weapons were first produced at the end of World War II, large
amounts of nuclear waste have been generated and stored at various facilities. The nuclear
10 waste, which consists largely of the byproducts of uranium and plutonium production and
purification, was disposed of in ways which were deemed suitable at the time, but which in
retrospect are now inadequate. Much of the nuclear waste is now stored in tanks as a highly
alkaline mixture of salts and liquids which, if not recovered and properly remediated, will
potentially create severe environmental problems.
Most of the stored aqueous nuclear waste is alkaline (pH 14), and contains high
concentrations of sodium nitrate. The tanks contain various complexing agents, fission products,
transuranic elements and other materials. Much of the stored nuclear waste is in the form of
sludge created when alkali was added to the waste to prevent tank corrosion. Some of the
radioactive material has been incorporated into salt cakes which is the evaporative product of
20 the alkaline aqueous material. It is desired to remove the radioactive elements from the waste
in order to allow for subsequent safe disposal of the non-radioactive materials. The removal of
two of the metallic radionuclides, cesium and strontium, is particularly hn~o,~nt because their
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
half-lives are long enough to repl~s~ilt a h~7~rd for an extended period of time.
Sodium nonatitimate is known as a strontium ion exchanger. J.Lehto; J. Radioanal.
Nucl. Chem. Letters, 118: 1-13 (1987) describes the ion exch~n~e behavior of Na4Ti9O20 ~ xH20
toward strontium. The sodium titanate was ~,~paled hydrothermally at 300~ C followed by
5 boiling with NaOH.
Other references that disclose the preparation and use of sodium titanates as strontium
ion exchangers include R.G. Dosch, "Final Report on the Application of Titanates, Niobates,
and Tantalates to Neutralized Defense Waste Contamination Materials Properties, Physical
Forms and Regeneration Techniques; National Technical Information Service" (1981); R.M.
Merrill; Journal of Radioanalytical Chemistry; 43: 93-100 (1978); J.Lehto et al.; J. Chem. Soc.
Dalton Trans., 101-103 (1989); and S.P. Mishra et al.: J. Rodioonalytical an~ Nuclear
Chemistry, Articles; 162:2, 299-30S (1992).
Various types of sodium titanates are described in the prior art as well as various uses
for sodium titanates as an ion exchanger for ions besides strontium. M. Watanabe; Journal of
Solid State Chemistry, 36: 91-96 (1981) describes the preparation of sodium titanate compounds
from hydrothermal reactions involving TiO2 with NaOH. The titanates described were TiO2,
Na2O ~ nTi2Na"TiO2. J. Akimoto et al., Journal of Solid S~ate Chemis~ry, 90: 147-154 (1991)
describes the synthesis of monosodium titanates NaTigO,3, which are distinct from the sodium
non~tit~n~t~s disclosed here. H. Leinonen et al., Reactive Polymers, 23: 221-228 (1994)
20 describes the use of sodium titanates as ion exchangers for nickel and zinc.
Lehto et al.; Radiochem. Radioanal. Letters, 50:6, 375-384 (1982) describes the effects
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
of gamma radiation on sodium titanate and other solid ion exchangers. The study eoncluded that
gamma radiation had very little impact on the strontium ion exehange eapacity of sodium
titanate.
PCT Applieation WO 94/19Z77 discloses silieo-titanates and methods for making and
5 using them. The silieo-titanates diselosed are useful for removing eesium from radioactive
wastes. U.S. Patent 4,156,646 discloses removal of plutonium and amerieium from aqueous
alkaline waste solutions using sodium titanate ion exehangers. The sodium titanate used is a
monosodium titanate.
U.S. Patent No. 5,352,644 describes a titania bound zeolite made by eombining the
10 zeolite, a low acidity titania binder material, and an aqueous slurry of titanium oxide hydrate.
CA 0223~337 1998-04-17
WO 97/14652 PCTAJS96/16753
SU~ ARY OF THE INVENTIO N
It is an object of this invention to provide a novel partially crystalline layered sodium
titanate that exhibits superb ion exchange properties towards strontium.
It is another object of this invention to provide a novel partially crystalline layered
S sodium titanate that is capable of ion exchanging strontium under highly alkaline solutions.
It is still another object of this invention to provide a novel partially crystalline layered
sodium titanate that has a d-spacing of from about 8 to about 9.9 angstroms.
It is yet another object of this invention to provide a novel method for preparing a sodium
titanate.
In one embodiment, this invention is a partially crystalline layered sodium titanate having
a d-spacing of from about 8 to about 9.5 angstroms.
In another embodiment, this invention is a partially crystalline layered sodium titanate
having a d-spacing of from 9.0 to 9.9 angstroms, a Langmuir surface area of from about 60 to
about 110 m2/g, and 001 reflection peak width greater than I degrees and less than about 4.5
degrees.
In still another embodiment, this invention is a partially crystalline layered sodium
titanate composite having the x-ray diffraction pattern of Figure 1 prepared by hydrothermally
treating a sodium titanate gel at a temperature of from about 120~C to about 200"C in the
presence of aqueous NaOH for a period of time ranging from about 1 to about 20 hours.
In yet another embodiment, this invention is a partially crystalline layered sodium titanate
having a d-spacing of from 9.0 to 9.9 angstroms, a 001 reflection "eak width greater than 2
CA 02235337 1998-04-17
W 097/14652 PCTAUS96/16753degrees and less than about 4.5 degrees prepared by the method comprising hydrothermally
treating a sodium titanate gel prepared by a reflux method and then hydrothermally treating the
reflux product at a temperature of from about 150~C to about 170 ~C in the presence of added
aqueous NaOH for a period of time ranging from about 1 hour to about 20 hours.
CA 02235337 1998-04-17
W O 97/14652 PCTrUS96/16753
DF-~CRIPrrlO N OF llHE DRA W nNGS
There is shown in the drawings a presently p,efelled embodiments of the layered partially
crystalline sodiurn titanate of this invention wherein:
Figure 1 is an X-ray diffraction pattem of a layered partially crystalline sodium titanate
S composite of this i~vention having a d-spacing of 9.4 angstroms;
Figure 2 are X-ray diffraction patterns of partially crystalline layered sodium titanate
sarnples 1-1 to 1-6 as prepared in Example 1;
Figure 3 are X-ray diffraction patterns of partially crystalline layered samples 1-7 to 1-1
as prepared in Example 1;
Figure 4 is an X-ray diffraction pattern of partially crystalline layered sodium titanate
sample 2-1 prepared in Example 2.
Figure 5 is a plot of the log (K~) versus pH for the uptake of strontium by sodium
titanate;
Figure 6 is the X-ray diffraction pattern of dehydrated layered partially crystalline sodium
15 titanate of this invention having a d-spacing of 9.00 angstroms as prepared in Example 9;
Figure 7 is the X-ray diffraction pattern of hydrated layered partially crystalline sodium
titanate that has a d-spacing of 10.2 angstroms as prepared in Example 9;
Figure 8 is a plot of the strontium uptake of a partially crystalline layered sodium titanate
prepared in Example 5 as a function of the full-width half maximum (FWHM) peak height of
20 the 001 reflection peak wherein FWHM and sodium titanate crystallinity are inversely related;
Figure 9 is a plot of the internal temperature in a reactor vs. time during sodium titanate
CA 02235337 1998-04-17
WO 97/146~;2 PCT/US96/16753
hydrothermal treatment;
Figures IOA, lOB, lOC, and lOD are scanning electron micrographs (SEM) of each
hydrothermally treated pilot plant batch of partially crystalline layered sodium titanate prepared
in Example S magnified 3000 times;
S Figure 11 is a plot of the concentration of strontium in the eMuent of a partially
crystalline layered sodium titanate packed column over time before and after regeneration of the
sodium titanate ion exchange material; and
Figure 12 is a plot of the kinetics of strontium uptake by bound and unbound partially
crystalline layered sodium titanate of this invention.
CA 0223~337 1998-04-17
W O 97/14652 PCT~US96/16753
DE;SCRIPI'ION OF THE CURRENT ~MBODIMENT
The present invention relates to a partially crystalline layered sodium titanate having a
d-spacing of from about 8 to about 9.9 angstroms, and a 001 reflection peak half width of from
1~ to about 4.5~. The new partially crystalline sodium titanate has strong ion exchange
S prope-Lies towards strontium as a result of physical plo~l~ies that distinguish the composition
from all other sodium titanates.
The partially crystalline layered sodium titanate of this invention has the formula Na3 4
4 4Ti8 4 9 2O,~ 5 20 6 ~ xH2O. It is preferred that the partially crystalline layered sodium titanate
has the formula Na4Ti9O20 ~ xH,O.
The non-crystalline layered sodium titanate of this invention also comprises amorphous
sodium tit~n~tto As will be rli5cllcced below, the combination of the layered partially crystalline
sodium titanate and amorphous sodium titanate is the result of hydrothermally treating sodium
titanate gel at specified conditions. The resulting partially crystalline layered sodium titanate has
a unique X-ray diffraction pattern as well as other unique physical and performance
15 characteristics.
The partially crystalline layered sodium titanate of this invention has been formulated to
maximize its strontium ion exchange capacity and selectivity. The strontium ion-exchange
plO~l ly has been deliberately manufactured into the sodium titanate of this invention by
controlling the sodium titanate hydrothermal treatment step to produce a sodium titanate product
20 that it is partially cryst~lline. The hydrothermal treatment variables that affect the crystallinity
of the sodium titanate product and, thereby, its strontium ion-exchange capacity and selectivity
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/16753
are, hydrothermal treatment temperature, the time of treatment, and the concentration of sodium
hydroxide used in the hydrothermal treating solution. An additional advantage of the
hydrothermal treatment is that it renders the sodium titanate essentially insoluble in strongly
~lk~line solutions.
S Various methods are used to prepare a sodium titanate gel for hydrotherma] treatment.
Two methods: the reflux method; and the sol-gel method are described in more detail in the
examples. The method used to prepare sodium titanate gel does not significantly affect the
strontium ion-exchange prope, Lies of the final sodium titanate product.
Strontium ion-exchange affinity is measured by the strontium distribution coefficient, Kd.
The distribution coeMcient, Kd is calculated using the following equation:
(c-C,~ v
where Ci and C, are the initial and final solution concentrations of strontium or any other
solution ion beirlg tested, V is the volume of the starting test solution, and W is the weight of
the sample tested. Kd is typically reported in units of mL/g. The strontium Kd is determined by
contacting a known sample of sodium titanate with a solution of known strontium concentration
for a controlled period of time, preferably 24 hours. Strontium Kd is solution specific and for
most Kd's reported herein, a solution consisting of SM NaNO3/O. 1 M NaOH/55ppm Sr was used.
If a reported Kd is derived from different strontium containing solution, then the solution
composition is reported. Strontium Kd is also sample size specific and unless otherwise reported,
--~o-
CA 0223~337 1998-04-17
W O 97/146S2 PCT~US96/16753
each test used 20 mL of solution and 20 mg of solid sample (V/M = 100 mL/g).
We have discovered that the strontium Kd of sodium titanate is a function of sodium
titanate crystallinity. If the sodium titanate product is too crystalline, mass transfer into the
sodium titanate will be slowed thereby lowering the strontium Kd. Likewise, if the sodium
5 titanate is not crystalline enough, the sodium titanate will not have the requisite d-spacing to be
a good strontium eYch~nger and the strontium Kd will be low. Thus the sodium titanate of this
invention is characterized as "partially crystalline" -- the crystallinity of the sodium titanate of
this invention is tailored during the hydrothermal treatment step to control d-spacing and other
physical properties, thereby maximizing the strontium Kd of the partially crystalline sodium
10 titanate composition.
At least three physical ~).ope.lies are characteristic of a partially crystalline layered
sodium titanate composition of this invention with the required crystallinity to be a good
strontium ion exchanger. The properties are, strontium Kd, X-ray diffraction d-spacing, and the
001 reflection peak half width. The latter two ~lupe.Lies are derived from the X-ray diffraction
15 pattern of the partially crystalline layered sodium titanate composition of this invention.
The relationship between d-spacing and the X-ray angle of reflection is set forth in
Bragg's Law:
2dsin~ = n~
Where d is the d-spacing in angstroms, n is an integer, ~ is the X-ray wavelength in angstroms
20 and ~ is the X-ray angle of reflection in degrees. The Bragg's Law Equation is a useful tool for
eLing X-ray diffraction patterns since the X-ray diffraction pattern is a trace of 2~.
-I 1-
CA 02235337 1998-04-17
WO 97/14652 PCT~US96/16753
The spacing between the sodium titanate layers is ascertained from the x-ray difi'ractior
pattern using Bragg's Law and is known as d-spacing. Sodium titanate is composed of layers
of titanium and oxygen atoms separated by voids containing sodium ions and water. The d-
spacing is the ~lict~nc~ from one titanium or oxygen atom to the identical atom one crystal layer
S away. D-spacing is the thickness of one titanate layer and one void space. For optimum
strontium ion exchange capacity, the sodium titanate of this invention must have a d-spacing of
from 8 to 9.9 angstroms, and preferably a d-spacing of from about 9.0 to about 9.9 angstroms.
A sodium titanate having the requisite d-spacing admits hydrated strontium, excludes hydrated
sodium, and exhibits strontium Kds in excess of 20,000 mL/g.
Our discovery of a sodium titanate having an optimum crystallinity for strontium ion-
exchange derived from our efforts to understand how hydrothermal re-treatment of sodium
titanate gels affects the crystallinity of the partially crystalline sodium titanate product. This led
to our discovery that sodium nonatitanate has an optimum crystallinity for use as a strontium ion
exchanger. The optimum crystallinity can be quantified by measuring the width of the 001
15 reflection peak at half of it's height (FWHM, units=degrees). The FWHM is inversely
proportional to the crystallinity of the sodium titanate material meaning the greater the FWHM,
the less crystalline the material is. High strontium Kd's are obtained when the FWHM is greater
than 1 degrees and less than about 4.5 degrees: 1.0 degrees < FWHM ~ 4.5 degrees, and
most preferably when the FWHM is greater than 2 degrees and less than about 4.5 degrees: 2.0
20 degrees < FWHM c 4.5 degrees.
The layered structure of partially crystalline sodium titanate of this invention is the source
-12-
CA 0223~337 1998-04-17
Wo 97/14652 PCT/US96/16753
of its ability to selectively exchange strontium. The partially crystalline sodium titanate
co.,.position contains spaces between the layers that are large enough to accept hydrated
strontium ions. The layer spacing is small enough, however, to exclude hydrated sodium and
this prevents hydrated sodium from competing for ion-exchange sites with hydrated strontium.
- 5 Sodium titanate surface area is related to the rate at which an ion exchanges. The larger
the surface area, the faster the strontium exchange rate. Typically, the greater the crystallinity
of sodium nonatitanate the 1Ower its surface area is. Thus, there is also a fine balance between
crystallinity and exchange rate. A partially crystalline sodium titanate of this invention will have
a surface area of from 25-200 m2/g. It is preferred, however, that the partially crystalline
sodium titanate of this invention has a surface area of from 60 to 1 10 m2/g.
Partially crystalline layered sodium titanates of this invention can be prepared by at least
two techniques; by the sol-gel technique, and by the reflux technique. Both techniques produce
a sodium titanate gel that must undergo hydrothermal treatment which crystallizes at least a
portion of the so~dium titanate gel to give a partially crystalline layered sodium titanate of this
invention. It is the parameters of the hydrothermal treatment parameters, including the NaOH
concentration, that are important for producing a partially crystalline sodium titanate composition
that has the desired Kd, d-spacing, and FWHM properties.
The sol-gel method for preparing a sodium titanate gel comprises combining titanium
isopropoxide and methanol at a weight ratio of from 1: 1 to about 1: 100 to form a first reaction
mixture. A second reaction mixture is prepared by combining NaOH and methanol in a weight
ratio that allows ~he sodium hydroxide to dissolve completely in the methanol. The first reaction
-13-
CA 0223~337 1998-04-17
WO 97/1465Z PCT/US96/16753
mixture is added to the second mixture to form an admixture and a third reaction mixture
comprising water and methanol slowly added to the admixture to initiate gel formation. The
final mixture is allowed to gel for a period of time ranging from about 15 minutes to about 2
hours or more. Preferably, the reactor is sealed and agitated at a high rate for a period of time
S of 30 minutes or longer to keep the sodium titanate gel fluid. The solvent and byproducts are
then evaporated from the sodium titanate gel in a vacuum oven operated at from 40~C to about
80~C for a period of time ranging from about 2 hours to about 24 hours or more until most of
the methanol solvent and reaction byproduct, isopropanol, is volatilized from the sodium titanate
gel.
An alternative and preferred method for producing sodium titanate gel is the reflux
method. The reflux method does not use methanol and only produces the corresponding
isopropanol byproduct thereby reducing the amount of volatile fumes and wastes produced by
the sodium titanate gel formation process as well as making solvent recovery easier and more
energy efficient. The reflux process comprises first preparing a solution of sodium hydroxide
by dissolving sodium hydroxide pellets in deionized water to give a first solution consisting of
from about 10 molar to about 19.2 molar NaOH. Neat -- 99% -- titanium isopropoxide (TiP)
is then added slowly to the first solution until the mole ratio of Ti to Na in the mixture ranges
from 1:1 to 1:10, and preferably from 1:5 to 1:9. The sodium hydroxide/titanium isopropoxide
mixture is then refluxed for a period of time ranging from about 60 minutes to about 4 hours
or more at a temperature of from 100-150~C to form an amorphous sodium titanate gel.
The sodium titanate gel produced by either method must be hydrothermally treated in
CA 0223~337 1998-04-17
W O 97/14652 PCTnJS96/16753
order to produce the partially crystalline layered sodium titanate of this invention. The
hydrothermal treatment is accomplished in a reactor at a pressure ranging from about 45 to about
1000 psig and at a temperature ranging from about 100 to about 250~ C. The sodium titanate
~el is hydrothermally treated for a period of time ranging from about 1 hour to about 1 day or
5 more.
As the hydrothermal treatment occurs, the autogenous pressure in the reactor vessel
increases with increasing L~ ,c,dt~lre. Therefore, the preferred average reactor pressure during
hydrothermal treatment ranges from about 50 psig to about 350 psig depending on reaction
temperature. The hydrothermal treatment may occur under alkaline conditions created by adding
10 water or a sodium hydroxide solution to the sodium titanate gel before or after it is added to the
hydrothermal treatment reactor or autoclave. The hydrothermal treatment step may be repeated
at least once in order to increase the crystallinity of the partially crystalline sodium titanate.
To obtain a partially crystalline layered sodium titanate product that falls within the scope
of this invention, the hydrothermal treatment step is preferably conducted for I to about 20 hours
at 160-200~C. By reducing the temperature of the hydrothermal treatment from 200~C to
160~C, the autogenous pressure is reduced from 247 to 90 psia, thereby making the process
safer and requiring less expensive equipment. The time reduction from 20 hours to 5 greatly
increases the overall productivity of the process.
In some instances, a partially crystalline sodium titanate will not be crystalline enough
20 to have the physical propel Lies of a partially crystalline layered sodium titanate of this invention.
In this situation, the partially crystalline sodium titanate can be hydrothermally treated a second
-15-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
and possibly subsequent times at the conditions recited above in order to increase the crystallinity
of the partially crystalline sodium titanate to the desired level.
Table 1, below, shows that the strontium Kd of a partially crystalline layered sodium
titanate can be improved in some cases by a second hydrothermal treatment. By closely
S controlling the hydrothermal treatment step parameter, we can obtain the optimal crystallinity
for strontium uptake using one, two, or more hydrothermal treatment steps.
TABLE 1
Table 1. The Effect of Two Hydrothermal Treatments on the Strontium Kd of
Sodium Titanate*
-t Hydrothermal Treatment 2nd Hydrothermal Treatment Kd at 24 hours
Temperature NaOH M mL/g
~C
200~C/Water 48,700
200~C/Water 240 0 75,400
200~C/Water 240 1 127,600
1~
* All hydrothermally treated for 20 hours
The partially crystalline layered sodium titanate of this invention has the formula Na3 4,
44Ti8492O,8.5206. The partially crystalline sodium titanate is removed from the reactor and
washed with deionized water and filtered. The washed product is dried at a temperature from
about 50~C to about 100~C for a period of time ranging from about 3 hours to about 2 days or
more.
CA 0223~337 1998-04-17
W O 97/14652 PCTAUS96/16753
The amount of sodium hydroxide added to the sodium titanate gel can be a critical
hydrotherrnal tre~fm~nt parameters and must be closely controlled in order to produce a sodium
titanate having an optimum crystallinity for strontium ion exchange. In order to manufacture
a partially crystalline sodium titanate having a d-spacing of from about 8 to about 9.9 angstroms
and a 001 reflection peak half-width greater than 1~ and less than about 4.5~, the amount of
sodium hydroxide added to the sodium titanate gel during hydrothermal treatment should range
from about 0.0M to about 6M, and preferably from 0.50M to about l.SM. The sodium titanate
gel inherently is naturally contaminated with NaOH. So sodium hydroxide will become
dissolved in any water added to the sodium titanate gel during gel hydrothermal treatment. It
is preferred, however, that an aqueous solution of NaOH in the molarity ranges given above be
added to the sodium titanate gel prior to hydrolysis.
The optimum hydrothermal treatment temperature is from about 100~C to about 250~C
and preferably from 160-200~C. Finally, it is preferred that the strontium K~, of the resulting
partially crystalline sodium titanate is at least 20,000 mL/g based on 20mL of solution consisting
of 55 ppm Sr/5M NaNO3/O.lM NaOH added to 200 mg of solid sample.
The partially crystalline sodium titanate of this invention is very useful when used as an
ion exchanger. In order to use the powdered partially crystalline sodium titanate as an ion
exchanger, it must be bound into larger particles to reduce the pressure drop in the ion exchange
column and to ease handling. Any binder known in the art for binding catalysts and ion-
exchangers may be used. However, the bound sodium titanate of this invention is very useful
in removing strontium and other radioactive waste from highly caustic aqueous solutions. So~
-l 7-
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/167S3
it is preferred that the binder be selected from materials that withstand radiation and alkaline
conditions, and the material should not inhibit or block strontium or other ions from entering
the partially crystalline layered sodium titanate. A composite material made up from 40 to 95
wt% of partially crystalline sodium titanate with 5-60 wt% of a binder is preferred.
S Both organic and inorganic binders can be mixed with partially crystalline sodium titanate
to make a bound ion exchange composition for strontium. For applications with nuclear waste,
inorganic binders offer the advantage of increased radiation resi~t~nce. For other applications,
organic binders may be easier to extrude into pellets than inorganics.
Examples of inorganic binders include silica or silica gel, silicon carbide, clays, and
silicates, including synthetically prepared and naturally occurring ones, which may or may not
be acid treated, for example, attapulgus clay, china clay, diatomaceous earth, fuller's earth,
kaolin, kieselguhr, etc.; ceramics, porcelain, crushed firebrick, bauxite; refractory inorganic
oxides such as alumina, titanium dioxide, zirconium dioxide, chromium oxide, beryllium oxide,
vanadium oxide, cerium oxide, hafnium oxide, zinc oxide, magnesia, boria, thoria, silica-
alumina, silica-rnagnesia, chromia-alumina, alumina-boria, silica-zirconia, etc.; crystalline
zeolitic aluminosilicates such as naturally occurring or synthetically prepared mordenite and/or
faujasite, for exarnple, either in the hydrogen form or in a form which has been exchanged with
metal cations; spinels such as MgAI2O4, AnAI2O4, CaAI2O4, and other like compounds; and
combinations of naterials from one or more of these groups.
Other examples of inorganic binders include various metal salts in powder, sol, or gel
form, as well as graphite and hydraulic cement may be used to bind any type of sodium titanate.
CA 0223~337 1998-04-17
W O97/14652 PCT~US96/16753
In addition, Ciment Fondo XR calcium aluminate, and Portland type 3 cement are good sodium
titanate binders w;th excellent strength and resistance to high alkalinity. When cements are used
as binders, the bound sodium titanate can be in the form of pellets, can be fashioned with dies,
or extruded.
~ S Organic binders may also be used to bind partially crystalline sodium titanate. Examples
of organic binders include polymers, starches, cellulose, cellulose acetate and other organic
catalyst and ion-exchanger binders known in the art.
Pore formers, surface area enhancers and other materials may be added to the partially
crystalline sodium titanate before, during, or after binding to improve the porosity and surface
area of the bound crystalline sodium titanate. A preferred pore former is one which can be
removed from the ion-exchanger chemically, or thermally before the bound material is used as
an ion-exchanger.
A preferred binder is a hydrolyzable titanium compound. The hydrolyzable titanium
compound is use~ul for binding any form of crystalline titanates including crystalline sodium
titanates, crystalline hydrogen titanates, and the preferred partially crystalline sodium titanate.
A hydrolyzable titanium compound of this invention will have the formula Ti XX,X"Xll, wherein
X is any constituent, and Xl,Xn,and Xlll are each chosen from the group consisting of Cl, Br,
I, or OR where R is any acyl or alkyl group containing 10 carbons and wherein R may make
one or two points of contact with Ti and oxygen.
Preferred hydrolyzable titanium compounds include titanium alkoxides and especially
titanium isopropoxide. The hydrolyzable titanium compound is preferably used to bind a form
~ -19-
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/16753
of crystalline sodium titanate that is a good strontium ion exchanger because the resulting titania
binder has little detrimental impact on the strontium Kd of the bound product. A preferred
crystalline sodium titanate is a partially crystalline sodium titanate having the formula: Na34.
44Ti8~9.2OI8s.206~ a d-spacing of from 8.0 to 9.9, and a (001) reflection peak half-width greater
S than 1~ and less than about 4.5~.
The hydrolyzable titanium compound, preferably titanium isopropoxide, and a crystalline
sodium titanate are combined and water from the air and from the crystalline sodium titanate
slowly hydrolyze the hydrolyzable titanium compound to form a titania bound crysta]line sodium
titanate.
The titania bound crystalline sodium titanate may be bound in the presence of an alcohol
such as methanol in which case the solid should be dried before use or before further processing.
The titania bound crystalline sodium titanate may be dried at ambient conditions or it may be
dried in an oven. In a preferred method, the titania bound crystalline sodium titanate is dried
in an oven at a temperature of from 75~C to about 100~C for a period of time ranging from
15 about 1 hour to about 12 hours or more.
The dried titania bound crystalline sodium titanate can be used as is, it can be ground and
sieved into smaller particles for use as an ion exchanger, or it can be processed further to
improve its mechanical ~u~ Lies. It is preferred that the dried titania bound crystalline sodium
titanate is further processed first by compaction, and then by calcination. The dried titania
20 bound crystalline sodium titanate can be compacted as produced, or it can be ground into small
particles, or into a powder and then compacted. It is preferred that the dried titania bound
-20-
CA 0223~337 1998-04-17
W 097/14652 PCTnJS96/167S3
crystalline sodium titanate is ground into smaller particles that can be easily compacted.
The titania bound crystalline sodium titanate may be compacted in any powder
compaction equipment known such as molding presses, tableting presses, and extruders.
Molding presses comprise a mechanically or hydraulically operated press and a two part mold
5 attached to the platens of the press, consisting of top (male) and bottom (female) portions. The
action of pressure and heat cau~ a particulate charge to flow and take the shape of the cavity
of the mold.
Tableting presses produce simpler shapes at higher production rates than do molding
presses. A single-punch press is one that will take one station of tools consisting of an upper
10 punch, a lower punch, and a die. A rotary press employs a rotating round die table with
multiple stations of punches and dies. Older rotary machines are single-sided; that is, there is
one fill station and one compression station to produce one tablet per station at every revolution
of the rotary head. Modern high-speed rotary presses are double-sided; that is, there are two
feed and compression stations to produce two tablets per station at every revolution of the rotary
15 head.
The titania bound crystalline sodium titanate may be dry granulated. In dry granulation,
the blended dry ingredients are first densified in a heavy-duty rotary tableting press which
produces pellets. The pellets are subsequently crushed into particles of the size required for ion
exch~nge. Densification can also be accomplished using a rotary compactor-granulator system.
20 A third technique, direct compaction, uses sophisticated devices to feed the blended dry
ingredients directly to a high-speed rotary press.
_
CA 0223~337 1998-04-17
W O 97/14652 PCT~US96/16753
Roll presses can also be used to tablet the titania bound crystalline sodium titanate by
directing a powder feed into a gap between two rolls rotating at equal speeds. The size and
shape of the compacted pellets are determined by the geometry of the roll surfaces. Pockets
or indentations in the roll surfaces form briquettes from a few grams up to 5 Ib. or more in
5weight. Smooth or corrugated rolls produce a solid sheet which can be granulated in the desired
particle size on conventional grinding equipment.
Lubricants added to the powder feed can aid in the transmi~sion of compaction forces and
reduce sticking to the die surfaces. Lubricants that are removed from the bound material at
calcining temperatures may be incorporated into the titania bound crystalline sodium titanate
10prior to compaction. Such lubricants may be selected from the group comprising boric acid.
graphite, oils, soaps, starch, stearic acid, and waxes. A preferred lubricant is stearic acid and
it is preferably present in the powder compactor feed in an amount ranging from about 0.1 to
4.0 weight percent.
The compaction step should produce a compact particle or pellet having a piece density
15ranging from about 1.5 to about 2.5 g/ml. It is most preferred that the compressed titania bound
crystalline sodium titanate particle has a piece density ranging from 1.8 to 2.2 g/ml.
The compacted titania bound crystalline sodium titanate particles are calcine~d at a
te~ dture ranging from about 150~C to about 500~C for a period of time ranging from 30
minutes to 10 hours or more. It is preferred that the particles are calcined in air at a
20temperature of from about 200~C to about 400~C for a period of time ranging from about 30
minutes to about S hours. It is most preferred that the particles are calcined in moist air at the
-22-
CA 0223~337 1998 - 04 - 17
wo 97/14652 PCT/US96/16753
conditions identified above.
Bound and unbound sodium titanate ion exchangers are useful in removing strontium and
other radioactive and non-r~ tive metals form aqueous solutions by packing bound or
unbound sodium titanate into a column and removing target metals from the aqueous streams
5 which are fed in to the column. Using this configuration, crystalline sodium titanate can remove
metals form large volumes of aqueous solutions. Sodium titanate ion-exchangers can also be
regenerated with an acid, and reused without loss of performance. Furthermore, bound
crystalline sodium titanate is able to remove strontium from aqueous streams having a pH of at
least 9.95 up to 13 or more, and a Na ion molarity of from about 1.0 to about 5.0 or more
10 without significant loss of exchanger capacity or physical integrity.
When used in processing nuclear waste, the ion-exchange columns should be made out
of glass, and may be lined with a polymer for caustic protection. Once the exchanger is spent,
the sodium titanate, along with the glass column can be vitrified to act as a impervious barrier
for the radioactive strontium.
The unbound and bound sodium titanate of this invention is useful as an ion exchanger
for metals besides strontium. Specifically, the sodium titanate is useful, either alone or bound,
in removing actinides, and especially uranium, from aqueous solutions. Other actinides which
can be similarly removed include thorium, plutonium and americium.
Sodium titanate may be used as an ion-exchanger as produced or it may be converted into
20 H-titanate and used as an ion-exchanger for metals such as ytterbium, zirconium, molybdenum,
silver, thallium, lead, chromium, vanadium, iron, cesium, tin, arsenic, and other metal ions with
-23-
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/16753
a high affinity towards hydrogen titanate.
The sodium titanate of this invention is also useful either a]one or with a binder, to
exchange a wide variety of metals at ~ line conditions. The metals are ranked according to
uptake in Table 2 below. For example, lithium, m~gnesium, nickel, cobalt, and barium are
5 exchanged by a partially crystalline sodium titanate in amounts equal to or greater than
strontium. Among the priority pollutants, partially crystalline sodium titanate has a high affinity
for zinc, copper~ cadmium, mercury, thallium, and lead. Among the precious metals tested,
silver has a very high uptake by sodium titanate. As seen in Table 2, these results indicate that
sodium titanate can be used to remove metals from industrial eMuent and other aqueous metal
10 containing waste, as well as recover target materials.
The ion exchange results reported in Table 2 are based on batch tests using 200 mg
sodium titanate, (sample 5-1 from Example 5, infra) an aqueous 20 mL solution containing 20
ppm of the metal ions being screened, at al'kaline conditions (pH 10-11). Sodium was present
in all samples, and the results reflect high selectivities for the indicated metals over sodium.
15 Sodium titanate is useful only as a cation exchanger, and therefore it did a poor job exchanging
some metals, including As, Sb, Mo, and Pb at high pH.
-24-
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
TABLE 2
Metals Affinity for Sodium Titanate. C~m~itionc initial metal --20 ppm, final pH 9.95-11.14.
Target K~ Industries where rnetal is present in ef~luent
Metal mL/g
Lithium > 19900 Nuclear, Military
Mq-~n~cium 14900 Mining Paint
Nickel (II)> 7400 El.~t~")lati~g, Explosives, Foundry, Wood, Tanning. Mining,
Paint, Petroleum, Fnqm~ling, Paper, Textiles
Cobalt (II)6900 Mining, Paint, Fn~m.oiing
Barium 6570 Paint
St~ulltiuLu6570 Mining, Nuclear
Zinc 5570 Ele~L-~,platiug, E~cplosives, Foundry, Wood, T~nning, Mining,
Paint, Fnqm~ling, Paper, Textiles
Silver 5150 Electroplating, Explosives, Foundry, Wood. Minin_, Paint,
Petrolwm, Textiles
Copper (Il)2900 Electroplating, Explosives, Foundry, Wood, Tanniny, Mininy,
Paint, Petroleum, Fnqm.~ling, Paper, Textiles
Cadmium 2120 El~troplatiny, Foundry, Wood, Mining, Paint, Petrolwm.
F~qm- ling, Textiles
Calcium 1900 Mining, Pamt
Mercury (II) ~ 1800 Foundry, Wood, Minin~. Paint, Petroleum. Fnqm.~lin~. Paper.
Textiles
Thallium (1) > 1800 wood, Mining, Paint, Petlu' Textih~s
Lead (Il) 1700 Electroplating, Explosivés, Foundry, Wood, Tanning, Paint,
Petroleum, Fnqm~ling, Paper, Textiles
Ytterbium (111) 1400 Nuclear
Mangane~ (Il) 1030 Mining, Paint, Fnqm~-ling
Rubidium 658 Nuclear
Cesium ** 492 Nuclear
Tin (Il) 217 Electroplating, Paint, Enameling
Chrûmium (111) 92 Electroplating, Foundry, Wood, Tanning, Mining, Paint,
Petroleum, FrqmPling, Paper, Textiles
Antimony (V) 87 Explosives, Foundry, Wood, Mining, Paint, Petroleum,
Fnqm-~ling, Textiles
Alurnmum 69 Mining, Paint, Fnqm~ling
Vanadium (V) 33 Mining, Paint
Iron (111) 33 Electroplating, Mining, Paint, Fn~nn.~ling
Zirconium 17
Arsenic (V) o Foundry, Wood, Mining, Paint, Petroleum Fnqm.~ling~ Textiles
Mining, Paint, Fnqm.~ling
Molybdenum (Vl) 0
45 ** Initial [(:S] = 43 ppm
-25-
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/167~3
EXAMPLE 1
This Example describes a reflux method for manufacturing sodium titanate gel of this
invention followed by hydrothermal treatment to produce a partially crystalline sodium titanate.
5.85 g of NaOH pellets were dissolved in 14 mL of distilled deionized water (DDI). 32
S mL of titanium isopro~ide (TiP) was then added slowly to the NaOH solution. This mixture
was then refluxed for 3 hrs at which time it was then transferred into a Teflon lined bomb using
distilled deionized water (except for Sample 1-2 which used 20 mL of 0.82 M NaOH solution).
The bomb was sealed and placed in a 190~C oven and allowed to react for 20 hrs. The material
was collected by filtration and washed one time with DDI and three times with methanol. The
10 initial Ti concentration in the bomb was 1.58 M and the initial concentration of NaOH in the
bomb was 2.46 M. The Ti:Na ratio was 1:1.56.
Table 3 lists the properties of partially crystalline sodium titanates prepared by the
method of this Example. The solution used to derive the strontium K~ data included SM NaNOl~
0.1M NaOH and 90-95 ppm Sr2+. The solution to sample weight ratio was 200:1 unless
15 otherwise reported.
-26-
CA 02235337 1998-04-17
W O 97/14652 PCT/US96/16753
TABLE 3
Sodium Titanates Pl~pd.c~d by Method of Example I
Sample Ti NaOH Ti: Na Reflux Hydrothermal Sr Kd
M M Mole timeTemp. (time) (mL/g)
ratio (hrs)
1-1 1.1 1.5 1:1.37 3.25200 (20 h) 3,180
1-2 1.6 2.5 1:1.56 3190 (20 h) 11,800*
1-3 1.0 2.0 1:2 3200 (20 h) 5,340
1-4 1.0 4.0 1:4 3200 (20 h) 19,900
1-5 0.63 3.1 1:5 2170 (22.5 h) 295,000
1-6 0.65 4.5 1:6.9 3.2193 (19.7 h) ~95,000
1-7 0.58 5.1 1:8.9 3.2193 (19.7 h) ~95,000
1-8 0.90 9.0 1:10 3.2190 (21 h) ---
1 9 0.44 8.35 1:19.4 16.5200 (1 d) ---
1-10 0.75 5.1 1:6.8 1 hr200 (4 d) 42,800
1-11 0.43 0.87 1:2 3.5145 (4 d) 692**
1-12 0.69 1.4 1:2.1 3 200 (7 d) 710**
1-13 gel gel 1:1.37 NA 200 (1 d) 3,360*
* The solution to sample v/w ratio was 250: 1.
** The solution to sample v/w ratio was 400: 1.
The XRD patterns of samples 1-1 to 1-13 are shown in Figures 2 and 3. Samples 1-6, 1-7, and
1-10 each have d-spacing below 9.9 and greater that 8.0 and 001 reflection FWHM's of between
1~ and 4.5~.
-27-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
EXAMPLE 2
This Example describes a method for preparing a sodium titanate designated as sample
2-1 by subjecting sample 1-2 of Example 1 to a second hydrothermal treatment. 3.50 g of
sample 1-2 was placed into a Teflon lined bomb. 38 mL of 4.2 M NaOH solution was added
S to the solid. This mixture was then treated hydrothermally for 2 days at 170~C. The sample
was collected by filtration and washed one time with DDI, four times with methanol, and two
times with ethanol. The X-ray diffraction pattern of sample 2-1 is found in Figure 4.
As seen in Table 5, the Kd before the second hydrothermal treatment (sample 1-5) was
3,060 mL/g and after the additional hydrothermal treatment (sample 2-1), 1270 mL/g.
CA 02235337 1998-04-17
W O97/14652 PCTAJS96/16753
EXAMPLE 3
This example details the preparation of a partially crystalline layered sodium titanate of
this invention by the sol-gel method.
Solution B was made in a plastic beaker by dissolving 40.793 g of 98.4% NaOH pellets
5 in 400 mL of methanol. The NaOH is not readily soluble in methanol. Therefore, the mixture
was stirred for over 30 minutes with a magnetic stirrer.
Solution A was made by mixing 294.2 g of titanium isopropoxide (TiP) and 250 mL of
methanol in the reaction vessel. The heat of mixing is exothermic (the temperature reached 72 ~
C). The reaction vessel was a 2-liter glass beaker wrapped with insulation tape. Next, solution
10 B was slowly added to solution A in the reaction vessel. The contents were stirred with a
magnetic stirrer, and heated to 53~C on a hot plate.
Solution C was made in a 100 mL beaker by mixing 0 mL water and 50 mL methanol.
Using a buret, solution C was added drop-wise to the heated mixture of A and B (53~C). The
resulting sodium titanate gel was mixed for another 15 minutes, then transferred to a ceramic
15 evaporating dish. The solvents were allowed to evaporate overnight in a hood. Next, about half
the gel was loaded into a liter round bottom flask, which was then connected to a rotary
evaporator. The gel was dried under vacuum for 1.5 hours at 60-75~ C. The rotary evaporation
procedure was repeated for the second batch. This procedure yielded 135 g of dried gel.
To make the final sodium titanate product, 20 g of the precursor gel was mixed with
20 either 40 mL of deionized water, or 40 mL of l M NaOH in a Teflon lined Parr autoclave. The
contents were then hydrothermally treated at 160-200~C for 5-20 hours.
-29-
CA 0223~337 1998-04-17
W O 97/14652 PCTAJS96/16753
EXAMPLE 4
This example details the preparation of a partially crystalline sodium titanate of this
invention using the reflux method followed by hydrothermal treatment.
A 30% NaOH solution was prepared by dissolving 419.88 g of 98.4% NaOH pellets in956 g of deionized water. This solution was transferred into a 3000 mL, 3-necked round bottom
reaction flask. 295.3 g of titanium isopropoxide (TiP) was added drop-wise from a 500 mL
addition funnel into the NaOH solution in the reaction flask. During the addition, the mixture
was stirred at 500 rpm. A large amount of white solids precipitated out of solution.
The reaction vessel was placed on a heating mantle, and equipped with a condenser,
thermocouple and temperature controller, and mixer. The mixture was stirred at 250 rpm and
refluxed at 110~ C for 5.25 hours. The sodium titanate gel product was allowed to cool and
settle out overnight. The water was then decanted, and the solids dried at 60~C for 2 days. The
yield was 152 g of sodium titanate gel.
To make the final product, 20 g of the precursor was mixed with either 40 mL of
deionized water, or 40 mL of 1 M NaOH in the Teflon lined Parr autoclave. The contents were
then hydrothermally treated at 160-200~C for 5-20 hours.
-30-
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/167~;3
EXAMPLE S
10 kg of sodium titanate gel were made using the sol gel method according to this
example. A 50 gallon reactor was charged with 30 kg titanium isopropoxide and 20.6 kg
methanol. To this mixture, a solution of 4.2 kg NaOH and 32.8 kg methanol was slowly added
S from one of several 20 gallon mixing tanks. A 2.1 kg water/4. 1 kg methanol solution was then
slowly added to initiate gel formation. Methanol evaporation from the reactor was elimin~tf~d
by applying pressure, and high agitation rate, (681 ft/min), to keep the gel fluid. This meant that
after a gelation lperiod of 30 minutes, the gel flowed freely from the reactor. The gel was
emptied onto trays which were placed in a 60~C vacuum oven at ''7 in. Hg for 1~' hours.
10 Methanol solvent (about 56 kg), and byproduct isopropanol (about 21 kg) were evaporated from
the product in the oven.
The dried gel was divided into four batches and treated hydrothermally in a S gallon
st~inless steel-lined autoclave. The autoclave liner was capped with a sheet of Teflon. Table
4, below, summarizes the conditions of each hydrothermal treatment. For each of batches 5-1
lS and 5-2, 4.5 kg of dry gel was hydrothermally treated in 9 kg of water for 20 hours at 160-200~
C. For batches 5-3 and 5-4, 2.6 kg of dry gel was treated in 5.2 kg of water for 5 hours at 160-
200~C. All non-stirred batches formed a solid chunk by the end of the autoclave step. This
solid was easily dispersable in water. Batch 5-4 was agitated throughout the treatment. The
material from batch 5-4 was a finely suspended slurry, and therefore was very easy to remove
20 from the autoclave.
The top layer of water on each batch in the autoclave was suctioned off and the
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
remaining solid slurry was transferred to a filter bag within a centrifuge. 7 kg of deionized
water was used to wash the soiids clinging to the sides of the autoclave. The solids were
centrifuged for 10 minutes to remove the water. An additional 4 kg of water was added to the
solids and then centrifuged to wash the product. The product was dried overnight in a 65~C
S oven with a flowing nitrogen purge. The combined yield of all four batches was 9.2 kg of
partially crystalline sodium titanate.
TABLE 4
-
Time Internal Internal Mixer Dried Yield of
(hr) Temperature Pressure Gel Final
~ C ** (psig) Charged Product
(kg) (kg)
Batch Heat At high Ave Range Ave
Up & Temp
Cool *
Down
5-1 26 20 200 200 140- 275 off 4.50 2.37
310
5-2 20 20 200 198 190- 285 off 4.50 3.98
305
5-3 15 5 200 199 155- 300 off 2.60 1.61
330
5-4 13 5 210 205 150- 350 on 2.60 1.23
372
* At time = at ~,.~dture between 160-200~ C
** Internal te.,l~.~ture high and average during when the autoclave was at temperature.
-32-
CA 0223~337 1998-04-17
W O 97/14652 PCT~US96/16753
EXAMPLE 6
X-ray powder diffraction (XRD) patterns of the synth~ci7~d materials from Examples I
and 2 were prepared and the results are shown in Figures 1-3. The XRD patterns indicate, for
the reflux samples, that as the mole ratio of Na/Ti increases, the crystallinity of the titanate
S material also increases until a ratio of 5 is reached. At this ratio, what appears to be a less
crystalline compound, forms.
The most crystalline samples 1-8 and l-9 were made with the highest ratios of NaOH to
Ti and in the strongest base solutions. The X-ray pattern for sample 2-l shows that this twice
hydrothermally treated sample is more crystalline than its precursor l-2. Samples 2-l, l-8 and
10 1-9 exhibited Kd values that were considerably lower than those of the less crystalline samples.
For example, sample l-8 had a strontium Kd of 7000 mL/g in 0.lM NaNO3. Under the same
conditions, sample l-l had strontium Kd value in excess of 100,000 ml/g.
Examination of the X-ray patterns leads to a conclusion that different types of materials
have been made. A more crystalline variety which appears to resemble sample 2- l are samples
lS 1-1, 1-2, 1-3, and 1-5 and a more gel-like group as exemplified by l-6, 1-7, and l-l0, and a
sample of intermediate crystallinity between the two, sample 1-4. It is sample types l-l0 and
1-7 that provide ~he highest Kd values.
Sample 1-13 (gel-based) and 1-2 (reflux-based) have similar Na:Ti mole ratios and were
reacted for approximately the same time period and temperature, but turn out to be different
20 structure nonatitanates. The X-ray diffraction pattern for the gel sample was similar to the high
Na:Ti ratio reaction material 1-10, but has much lower ion eY~h~nge selectivity toward Sr2+.
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
Upon comparing the X-ray diffractions patterns of sample 1-12 and 1-3 it-appears that
longer hydrothermal treatment does not change the crystallinity of the composition but it does
have a definite effect upon its ion exchange p,~,pe.Lies.
CA 02235337 l998-04-l7
W O 97/14652 PCTnJS96/167~3
EXAMPLE 7
This Example describes the strontium distribution coefflcient (Kd) results for sodium
tit~n~tPc prepared in Examples 1 and 2. The distribution coefficient as defined in this example
is the ratio of the concentration Sr2+ in the exchanger to the concentration of Sr7+ in solution at
5 equilibrium. Re~ P the analyses were carried out an a weight basis, the units are g/g instead
of the usual mL/g. The distribution coefficients (K~) were determined for Sr2+ as a function of
pH for three of the samples, 1-1, 1-10, and 2-1. The resulting data are shown in Figure 5. As
the pH of the solution decreases, the selectivity for Sr decreases. The Kd at pH values above
11 are reported as being greater than 300,000 because at this pH, the Sr concentration was
10 below the detection limit of the AA unit and a value of 0.1 ppm was used to calculate the Kd.
It is illlpOI Lant to mention that there is typically a large difference between the initial pH of the
Sr solutions and equilibrium pH after exchange. This is due to the hydrolysis of the sodium
nonatitanate whose reaction is shown below. This property is found in all the layered titanates.
Na4TigO2o + xHzO > H,~Na4 ~TigO20 + xNaOH
Distribution coefficients for S~+ were also determined for these same samples in the
presence of 5 M NaNO3 and 1 M NaOH and 90-95 ppm Sr2+ with a solution mass:sample mass
ratio of 400:1. The results are found in Table 5, below. Sample 1-10 exhibits the best
o~ ance under the conditions studied. When toY~mining samples treated for longer
hydrothermal periods with that of their X-ray diffraction equivalents (sample 1-12 with sample
20 1-3), the Sr Kd's decrease by a factor of 2. This suggests that very short hydrothermal treatment
periods will produce highly selective materials.
CA 02235337 1998-04-17
W O97/146S2 PCT~US96/16753
When reduction of loading was done, a corresponding increase in K~ was observed. For
' example, the strontium distribution coefficient of sample 1-10 increased to 58,700. The increase
in K~ on reduction of the solution volume from 400 mL to 200 mL indicates that the larger
volume of solution contains enough strontium ion to exceed the capacity of the exchanger in 6
S M Na+. We may assume that a Kd of 58,700 g/g represents an uptake of 99% of the added
Sr2+. The exchange capacity of Na4Ti9O20 is 4.75 meq/g and in the 200g solution the uptake was
0.2 meq/g thus 4.2% of the capacity was utilized in the experiment where solution to solid
weight ratio was 400 and the Kd was 13,400 g/g. Calculation shows that 97% of the Sr2+ was
taken up. Thus, the loading was 2 x 0.2 x 0.94 = 0.388 meq or 8% of the exchange capacity.
This figure represents a very high value for such strong solutions with such high sodium
concentrations .
TABLE 5
Sample Initial pH Final pH Sr K" (glg)
1-10 13.63 13.80 9.780
l-l l 13.63 13.82 692
1-12 13.63 13.74 710
1-1 13.63 13.74 l,270
2-1 13.63 13.74 1,270
1-5 13.6 13.8 3,060
1-4 13.6 13.7 1,730
VIM = 400
-36-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/1 67S3
EXAMPLE 8
A portion of sample 1-6 was placed on a piece of Whatman filter paper and inserted into
a longneck round bottom flask which contained distilled deionized water. The flask was gently
S heated to produce water vapors which would then impregnate the partially crystalline layered
sodium titanate sample. A second portion of sample 1-6 was placed in a 75~C oven until
needed. For the Kd measurements, TGA, and XRD pattern of this dried material, the second
portion of sample 1-6 was removed from the oven and immediately transferred to the appropriate
testing container to help prevent absorption of moisture. The Kd measurements were done using
a solution containing 5 M NaNO3, 1 M NaOH, and 91.67 ppm Sr~+. The results are shown in
Table 6 below. From the XRD patterns (Figures 6 and 7), it can be seen that the dehydrated
1-6 sample possessed a d-spacing in the range of 9.0-9.9 A while the hydrated sample possessed
d-spacing in the range of 9.4-10.2 A. The results, found in Table 6, show an initial average
sodium titanate d-spacing less than about 9.9 angstroms plays a considerable role in promoting
15 the uptake of Sr2+ and is related to the state of hydration of the initial exchanger.
-37-
CA 02235337 1998-04-17
WO 97/14652
PCT~US96/16753
Table 6 -- Sr Kd's for a hydrated and slightly dehydrated partially crystalline sodium htanate.
Sample Sr Kd (~
1-6 (1) dehydrated ~254,000
1-6 (2) dehydrated 229,000
1-6 ( 1 ) hydrated 18, 600*
1-6 (2) hydrated 21,200*
~0 * The weight of the sample was adjusted for the excess water which is contained by multiplying
by 0.8163 which was determined by TGA data.
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
EXAMPLE 9
This Example ex~mines the effect of the hydrothermal treatment step on the strontium
Kd of a partially crystalline layered sodium titanate. Four parameters were studied: method of
gel ple~aldtion (~ol-gel or reflux), concentration of NaOH in the autoclave, the time, and the
~ S te,-")eldture of the hydrothermal treatment step.
The source material for the experiments was sodium titanate gel made by the sol-gel
method and by the reflux method and made according to Examples 3 and 4. The results in
Table 7 indicate that the strontium Kd's for partially crystalline sodium titanates prepared by
hydrothermally treating amorphous sodium titanates are two times greater for material treated
at 160~C for S hours than at 200~C for 20 hours. The most statistically significant hydrothermal
treatment variable affecting the 24 hour Kd data is the relationship between the concentration of
NaOH and the temperature during the hydrothermal treatment step. Control of both of these
parameters is necessary to fine tune the crystallinity of the final product.
-39-
CA 02235337 1998-04-17
WO 97/14652 PCTAJS96/16753
TABLE 7
.
The Effect of the Hydrothermal Treatment Conditions on the Strontium Kd of Sodium
Titanate
Sample Gel Method Temperature NaOH M Time hr Kd at 24 hours
Number ~C (mL/g)
9-1 sol-gel 160 0 5 22,400
9-2 reflux 160 1 5 35,500
9-3 reflux 160 1 5 35,500
9-4 sol-gel 160 1 20 21,700
9-5 reflux 200 0 5 8,1 40
9-6 sol-gel 200 1 5 24,100
9-7 sol-gel 200 0 20 10 . 600
9-8 reflux 200 1 20 21,700
9-9 reflux 200 0 20 6,660
-40-
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
EXAMPLE 10
In this Example, the strontium Kd of partially crystalline sodium titanate prepared in
Example S was evaluated. Table 8, below, summari~s the hydrothermal treatment conditions
and p,o~Lies of the four batches of sodium titanate prepared in ~xample 5. Batches 1 and 2,
S which were hydrothermally treated for the longest period of time, (160-200~ C for 20 hours),
were the most crys~lline, as in~ic~t~d by the low full width at half maximum of the largest peak
in their X-ray pattem. These most crystalline materials also have the lowest Kd's because high
crystallinity hinders the interlayer diffusion, thereby lowering strontium uptake.
TABLE 8
Summary of the Pilot Plant Hydrothermal Treatments
Batch Time at Mixer Yield of Crystallinity Sr Kd3
Temp. I Final Prod- FWHM2
uct kg deg mL/g
5-1 20 off '.37 1.13 12,100 +
2,100
S-2 20 off 3.98 1.92 10,200 + 420
5-3 S off 1.61 2.67 19,800 + 410
lS 5-4 5 on 1.23 2.23 21~100 +
2.310
' At temp. = at ~"l~lature between 160-200~C.
2 FWHM = 001 reflection peak full width at half maximum.
3 SM NaNO3/0. lM NaOH/68.6 ppm Sr.
-41-
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/16753
The sodium titanate from all four batches have a high strontium Kd. But the strontium
Kd for the batches hydrothermally treated for S hours are about two times higher than for the
batches treated for 20 hours (20,500 vs. 11,500 mL/g). This difference may be correlated with
the degree of crystallinity of the samples. Figure 8 is a plot of the strontium Kd of the batches
5 as a function of FWHM (full width at half maximum peak height of the 001 reflection peak).
The larger the FWHM, the less crystalline the material. The materials with the higher
crystallinity (batches 1 and 2) have the lower K~s.
The X-ray pattern for batch 2, shows a major peak at 8.76 angstroms. This peak is still
present after stirring the sample with 0.1 M NaOH. By this observation, the new peak is not
10 a H4Ti9O20 phase (which should result if the sample were washed with too much water in the
pilot plant). Batch 2 may contain another sodium titanate phase, possibly Na2Ti~O, which has
its major peak at 8.270 angstroms. All other batches contain only nonatitanate. Compared to
the other batches, Figure 9 shows that batch 2 has the longest period at the internal temperature
of 200~C (18 hours). This extended heating period may have allowed the new phase to form.
Figures 10A, B, C, and D show scanning electron micrographs (SEM) of each of the
pilot plant batches at 3000 times magnification. The partially crystalline sodium titanate of
batches 1 through 3, Figures 10A, l0B, and l0C, shows the material to have formed nuggets,
while batch 4, Figure 10D consists of elongated bundles of fibers. While mixing does not affect
the resulting strontium Kd, it does influence the gross morphology of the exchanger. Mixing
20 the gel during hydrothermal treatment causes shear stresses in the newly forming sodium
titanate, which in turn results in a higher number of loose fibers and elongated bundles.
-4~-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
EXAMPLE 11
Fondo XR cement (8.6g) and sodium titanate (20g, - 200 Mesh) prepared by the method
of Example 5 were dry blended vigorously for 5 minutes to form a bound sodium titanate.
Subsequently, deionized H2O (14.7g) was slowly added while mixing continuously. The mixture
5 was allowed to blend for an additional 5 minutes. The paste like mixture was thinly spread over
a Teflon sheet. The sheet was placed in covered pans and exposed to water saturated in an oven
at 30~C and cured for 48 hr. The cured sheets were broken down to 20 x 40 mesh size with
mortar and pestle.
-43-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/167!;3
EXAMPLE 12
This example details a method for preparing an amorphous sodium titanate bound
crystalline sodiurn titanate.
In a beaker located on a magnetically stirred hot plate, 293 g of titanium isopropoxide
5 (TiP) was mixed together with 250 mL of methanol. The solution was heated to 50~C. Next,
a solution of 40.7 g NaOH in 400 mL methanol was added to the beaker. To this mixture, 10
g of through 200 mesh sodium titanate prepared by the method of Example S was added. After
adding a solution of 50 mL methanol and 20 mL water, the solution plus solids mixture was
stirred for l hour. The resulting thick paste was spread out on a Teflon sheet and air-cured for
6 days. The cured plaque was ground with a mortar and pestle, then sieved to obtain 20-40
mesh particles. The resulting bound ion exchange compound is a semi-crystalline sodium
titanate bound with amorphous sodium titanate.
-44-
CA 02235337 1998-04-17
W O 97/1465Z PCT~US96/16753
EXAMPLE 13
This example details the p~cfel~d method for preparing an titania bound partially
crystalline sodium titanate.
Six grams of sodium titanate, having a particle size that passes through 200 mesh,
S prepared by the method of Example 5, was mixed with 10 mL of titanium isopropoxide (TiP).
The mixture was agitated until it formed a thick paste. The paste was subsequently spread on
a Teflon sheet and air-cured for 45 minutes. After air-curing, the plaque was cured in the oven
at 85~C overnight. The cured plaque was ground with a mortar and pestle, then sieved to obtain
20-40 mesh particles. The resulting compound is a partially crystalline sodium titanate bound
10 with an amorphous titania.
-45-
CA 02235337 1998-04-17
WO 97/14652 PCT~US96/16753
EX~MPLE 14
Column experiments were carried out on sample 1-2 (sodium nonatitanate) using a model
solution in order to .os~mine the performance of the exchanger under dynamic conditions.
Composition of the model solutions was as follows: S M NaNO3, I M NaOH, 0.15 M KCl, Cs-
S 13.9 ppm, Sr-8.75 ppm and Ca~.08 ppm. The volume of the absorbent was 0.85 cm3, the
absorbent's layer height was lSSmm, the absorbent granules size was <0.1 mm, and the flow
rate was 3 bed volumes (b.v.) per hour. The regeneration of the exchanger was achieved by
passing 15 mL of a 0.5 M HCl solution through the column at a rate of 0.5 b.v./hr and then
subsequent treatment with ~0 mL of a solution containing 5 M NaOH, and 0.15 M KCI (0.5
b.v./hr) to give a regenerated partially crystalline sodium titanate.
The breakthrough curve for Sr sorption is shown in Figure 11. The curve shows that the
partially crystalline sodium titanate effectively purifies approximately 700 column volumes,
(c.v.), of the model solution in the first sorption cycle, and after regeneration, the material
purifies nearly twice the number of c.v.'s (1200) in the second cycle.
-46-
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
EXA M PLE 15
Pellets of sodium titanate prepared by the method of Example 5 combined with Portland
cement and calcium alumin~tP (Ciment Fondu) were prepared according to the method of
Example 11, based on the recipe set forth in Table 9, below.
Sr-Kd measurements for the Fondu and Portland cement at various curing times are found
in Table 9 below. The Kd values were taken using standard test conditions. The Sr-Kd values
at all curing times are low when compared to Sr-Kd values of 1633 mL/g for the -200 mesh
virgin partially crystalline sodium titanate. The Ciment Fondu binder exhibits a slightly higher
Sr-Kd than the Portland cement. There was no obvious effect of the curing time on the final
properties of the inorganically bound partially crystalline sodium titanates.
TABLE 9
Effect of Curing Time and Cement Composition of Sr-Kd of Bound Sodium Titanate.
Sample IDBinder Type Curing Time Final Sr Concentration Sr-Kd
50 wt% (days) (ppm) (ml/g)
15-1 Ciment Fondu 4 40.2 136
15-2 * 1 1 29.9 252
15-3 * 21 31.0 236
15-4Portland Type III 4 54.1 50
15-5 * 1 1 55.5 44
15-6 * 21 56.4 40
To determine if the sodium titanate was blinded by the inorganic binder, samples were
analyzed via nitrogen porosimetry. The results are shown in Table 10 below and indicate that
-47-
CA 0223~337 1998-04-17
WO 97/1~652 PCT/US96/16753
the inorganic binders have enc~rsul~ted some of the sodium titanate powder as shown by the
decrease in both the BET and Langmuir surface areas in the bound samples as compared to the
sodium titanate powder.
To determine the extent of mass transfer difficulties, Sr-Kd measurements were taken over
a period of 120 hours for bound and unbound samples. Figure 12 and Table 10 show that the
bound sodium titanate exhibits the fastest uptake of Sr. Both organic and inorganic binders
reduced the rate of Sr uptake. The organic binder made with the pore former has a relatively
quick Sr uptake and eventually reaches the value of the sodium titanate powder. However, the
inorganic binders have very slow Sr uptake and do not asymptotically reach the same
concentration as with the sodium titanate powder. As mentioned above, the inorganic binder
appeared to blind the sodium titanate powder. Since the final concentration of Sr does not
approach the final concentration observed with the sodium titanate powder, a portion of the
sodium titanate appears be totally encapsulated. Cement bound absorbents using only pure
cement have both caustic and mechanical stability, but some mass transfer ability is lost as a
result of the binder.
-48-
CA 02235337 1998-04-17
WO 97/14652 PCTAJS96/16753
TABLE 10
Effect of Inorganic Binder on Pellet on Surface Area
Sample T ~nE~muir Surface BET Surface Area Average Pore Radius
Area (m2lg) (A)
(m2lg)
SSodium Titanate 106.7 ~9.9 39.7
Bound with 39.3 22.8 . 43.4
Ciment Fondu
10Bound with 52.1 30.0 64.1
Portland Type III
-49-
CA 0223~337 1998-04-17
W O97/14652 PCT~US96/16753
EXA~PLE 16
This Example describes methods of introducing porosity into the inorganic binderlion
exchanger system to improve mass transfer. By increasing the porosity of the ion exchanger
pellet, mass transfer of cations into the exchanger increases and therefore increases the amount
5 of radioactive oation c~ uled. Two methods investigated for increasing porosity were gas
foaming of the binder-exchanger and addition of inert fillers to the binder-exchanger system.
The inert filler was leached out to form porous pellets. All samples using pore formers were
prepared by using the general procedure listed in Example 15.
In the first method of producing porous pellets, aluminum was added to the mixture to
10 induce gas formation in the binder-exchange system. The aluminum reacts with water under
caustic conditions to produce hydrogen. The hydrogen produces bubbles or voids in the pellet.
Results are shown in Table 11 for varying loadings of aluminum and particle size. All of the
samples passed the standard caustic stability test. Samples 16-1 and 16-2 were subjected to a
small normal force by placing a small weight on the sheet. These sheets did not expand to the
15 same extent as the other samples and have slightly lower product yields. The Kd values for all
the porous pellets are significantly lower than that seen with the virgin partially crystalline
sodium titanate powder which has a Kd value typically greater than 10,000 mL/g.
-50-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
TABLE 11
Effect of Al as Porosity Modifier on Sr-Kd of Fondu Bound Sodium Titanate.
Sample ID Binder ExchangerAluminum Al Size Final Sr Sr-Kd
~ Wt% Wt% Wt% (Micron) Concentration (ml/g)
(ppm)
16-1 49 49 2 20 29.6 132
16-2 49 49 2 45 19.8 246
16-3 49 49 2 20 48.0 43
16-4 49 49 2 45 46.2 48
16-5 48 48 4 20 30.0 129
1 0 1 6-6 48 48 4 45 29. 3 1 34
One aluminum filled porous sample was sent for surface area determination via nitrogen
porosimetry and exhibited a Langmuir surface area of 111.1 m2/g, a BET surface area of 63.
m2/g, and an average pore radius of 32.2 angstroms. The aluminum filler increased both the
15 BET and Langmuir surface areas in the bound samples to surface areas similar to that seen in
the unbound sodium titanate. Even with the increased surface area, the Sr-Kd for this sample
was small.
In the second method of producing porous pellets, inorganic or organic fillers were
blended and cured with the binder-exchanger system described in Example 11. The inorganic
20 fillers were calcium carbonate and calcium sulfate while the organic filler was a low molecular
weight hydrocarbon wax. These fillers have low solubility in water but can be leached out with
the ~plu~-iate leaching solutions.
CA 0223~337 1998-04-17
WO 97/14652 PCT/US96/16753
TABLE 12
Effect of Inorganic Fillers on Sr Kd of Fondu Bound Sodium Titanate.
Sample ID Sample Particle Size Sr Kd
(mesh) (ml/g)
16-7Sodium Titanate bound with 50% 20 x 40 150
Ciment Fondu (20 vol % extracted
wax)
16-8 n -200 1 12
16-9Sodium Titanate bound with 50% 20 x 40 150
Ciment Fondu (20 vol % extracted
wax)
16-10 " -500 132
Typical results for the two inorganic fillers are shown in Table 12. These samples were
each leached with acid and the sodium titanate was regenerated using sodium hydroxide. All
of the Kd values of the inorganic salt filler pellets were lower than the Kd of the virgin sodium
titanate powder.
Organic pore formers were used to increase porosity without introducing additional
15 calcium to the system. The organic pore formers were added into the system at different volume
percent loading as listed in Table 13. There was no increase in Kd as the porosity due to leached
wax increased. The strontium Kd did increase with increased exchanger loading for samples 16-
13 and 16-14. This was also observed with the inorganic fillers. There appears to be an upper
limit to the amount of exchanger that can be added without loss of mechanical strength. The
20 Kd values were again lower than the virgin sodium titanate powder.
-52-
CA 02235337 1998-04-17
W O 97/14652 PCTrUS96/16753
TABLE 13
Effect of Organic Fillers on Sr-Kd of Fondu Bound Sodium Titanate.
Sample IDBinder Wt% ExchangerFiller Type Filler Wt% Sr-Kd
Wt% (vol %) (ml/g)
16-11 34.6 34.6Petrolite 30.9 383
185 (60%)
16-12 30.8 46.5Petrolite 22.7 524
185 (40%)
16-13 36.9 36.9Petrolite 26.3 299
185 (50%)
16-14 43.0 28.8Petrolite 27.7 552
185 (50%)
16-15 37.0 37.0Petrolite 26.0 372
185 (50~)
16-16 37.0 37.0Petrolite 26.0 429
185 (50%)
CA 0223~337 1998-04-17
wo 97/14652 PCTtUS96/16753
EXAlVlPLE 17
Cement or calcium exchanged bound sodium titanate can be fully regenerated with an
acid wash followed by a caustic wash. However, when the acid wash was tested on cement or
calcium exchanged, cement bound sodium titanate, the pellet broke apart, indicating acid attack
5 on the binder.
To select the best acid regenerate which does not harm the binder, we measured the
amount of Ca+2 in solution after shaking either SO0 mg of Ca-titanate or cured cement (without
an exchanger) in test solutions for 72 hours. The calcium titanate was made by shaking sodium
titanate in a 2 M solution of CaC12 for several days. The test solutions where made from
10 different concentrations of acetic acid, nitric acid, nitric acid plus 1 M NaCI~ and NaCI. For
each type of solution, lower pH's removed more Ca+2 from the titanate. Below pH 5.5, nitric
acid mixed with 1 M NaCI is the best regenerant, removing the most calcium from calcium
titanate. Acidic conditions are required to displace Ca+2 with H+l. The presence of NaCI
further enhances the amount of Ca+2 released by increasing the ionic strength of the solution,
15 thereby increasing the driving force for calcium to leave the exchanger.
The same solutions tested with the Ca-titanate were tested with cured cement containing
no exchanger. Regardless of the solution, the cement broke apart when the pH < 2. For pH
~3, the cement released very little Ca+2 for all solutions except acetic acid. Based on these
observations, a solution of 0.01 M nitric acid plus 1 M NaCI (initial pH=3.1) is the best
20 regenerant.
Also evaluated were procedures to regenerate bound Na-titanate (assumed to be both the
CA 02235337 1998-04-17
W O 9711465Z PCTAJS96/16753
Ca and Na form) into a form with a larger amount of Na using a two step procedure. For the
first step, a solution of 0.01 M nitric acid plus 1 M NaCI was passed through a column packed
with sodium titania bound with 50 wt% Cement Fondu-XR (sample 16-1). The acid solution
was passed through the column until the eMuent pH reached 4 (about 350 column volumes).
S Next, half of the acid-washed sample was treated with 82 mL 0.1 M NaOH, and the other half
with 68 mL of 2 M NaOH. All samples maintained mechanical strength after the treatments.
The untrea~d sample, the acid washed sample, and the 0.1 M NaOH treated sample all
have nearly the same strontium uptake (Kd between 161-264 mL/g based on sodium titanate
weight within the pellet). However, the strontium uptake of the sample regenerated with the 2
M NaOH was 3.9 times higher than the original sample (623 vs 161 mL/g).
The low Kd of the regenerated bound sample with respect to the virgin sodium titanate
powder (623 vs. 10,000 mL/g) probably indicates that mass transfer within the pellet must be
improved by increasing the porosity.
-55-
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
EXAMPLE 18
Partially crystalline layerecl sodium titanate unbound and bound with 50 wt% Ciment
Fondu was tested for radiation stability. The samples were irradiated for 250, 500, 750 and
1000 Mrad of exposure. All of the sodium titanate powders exhibited a Kd of approximately
5 40,000 ml/g. The Sr-Kd values of the irradiated bound samples are all equivalent with a value
of 150 ml/g, which is consistent with the values seen in Table 9. The unirradiated sample had
a slightly higher Kd and may be within experimental error of the analytic technique. All of the
bound samples qualitatively exhibited excellent mechanical strength after irradiation and
treatment with caustic.
-56-
CA 02235337 1998-04-17
W O 97/14652 PCTnJS96/16753
EXAMPLE 19
In this Example, sodium titanate gel was tested as a binder for partially crystalline
sodium titanate and found to have good mechanical strength, is resistant to caustic, and enhances
the strontium uptake of sodium ti~nz~te Since the sodium titanate gel is inorganic, this binder
will have a high resist~nce to damage by radiation.
Sodium titanate as p,~,~ar~ in Example S was bound with sodium titanate gel by two
different methods. In the first method according to Example 12, the binder was formed by
making a sodium titanate sol-gel (methanol, NaOH, and titanium isopropoxide) and adding the
sodium titanate exchanger before curing. In the second and preferred method according to
Example 13, titanium is~p.upo~ide (TiP) was mixed with the sodium titanate ion exchanger and
allowed to solidify in moist air before curing.
Table 14 summarizes the effectiveness of titanate as a binder for sodium titanate with
respect to ion exchange capacity and mechanical strength. Bound samples passed the mechanical
test if they withstood being shaken for 24 hours in solutions containing 64 ppm Sr, 5 M NaNO~,
and 0.1 M NaOH (marked as "P" in Table 14); Those that failed are designated by "F".
The presence of the titanate binder enh~nces strontium uptake by sodium titanate.
Compared to the unbound sodium titanate (19-1), samples 19-2 and 19-3, bound with the gel
method, took up 1.76 times more strontium; and sample 19-6, bound with the TiP method, took
up 2.9 times more strontium. Amorphous sodium titanate (gel) by itself gives a high Kd for Sr
(17,000 mL/g). The high strontium Kds may be the result of a combination of the binder
enh~ncing mass transfer and the powder having a high capacity for strontium.
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
TABLE 14
Strontium Uptake by Sodium Titania bound with Amorphous Sodium Titanate.
Bound sarnples 20-40 mesh -- Feed 64 ppm Sr/5. 1 M NaNO3/O. 1 M NaOH
Sample Binder Form Curing Method Mechanical Kd
Strength* mL/g
19-1 none powder ---- --- 10,930
(batch 2)
19-2 gel plaque 6 days in air at room P 19,220
temperature
19-3 gel extrudate 6 days in air at room P 19,220
temperature
19-4 gel extrudate 4 days in water at room F
tem~"erature; 85 ~ C in air
overnight
19-5 TiP plaque 2 days in air at room F ---
temperature
19-6 TiP plaque 85~C in air overnight P 31,800
-58-
CA 02235337 1998-04-17
W O97/14652 PCTAJS96/16753
EXAMPLE 20
- This example details the measurement of strontium equilibrium batch capacities for titania
bound Na-titanate prepared according to Example 13, as well as each of the unbound pilot plant
batches prepared in Example 5. As shown in Table 15, the 24-hour Kd of titania bound sodium
titanate is 3.5 times higher than powder (37,000 vs. 10,800 mL/g for 64 ppm Sr, 5.1 M NaNO3,
and 0.1 M NaOH). However, the capacity of the TiP-bound sodium titanate is nearly the same
as for the unbound material (batch 2). Re~llse the presence of the titania binder does not affect
capacity, the binder itself may take up strontium. Kd measurements of the titania binder
indicates that this is the case, with a Kd of 17,000 mL/g (64 ppm Sr, 5.1 M NaNO3, 0.1 M
NaOH). The powder combined with the binder has a higher strontium uptake than the powder
of the binder by itself. The titania binding process produces a composite material which has
enhanced mass transfer.
TABLE 15
Capacities of Strontium Exchangers. Fe~d: 55.4 ppm Sr, 5.1 M Na, 0.1 M OH.
Exchanger Capacity Kd
(meq/g) (mL/g)
Na-Titanate pilot plant batch 1 0.897 12,100
Na-Titanate pilot plant batch 2 1.06 10,800
Na-Titanate pilot plant batch 3 1.112 14,800
Na-Titanate pilot plant batch 4 1.19 21,000
Na-Titania bound with amorphous Na-titanate 0.991 37,000
-59-
CA 0223~337 l998-04-l7
WO 97/14652 PCT/US96/16753
EXAMPLE 21
This Example details a rnethod for manufacturing sodium titanate ion exchange pellets
using an organic binder with introduced porosity to improve mass transfer inside the pellet.
Partially crystalline sodium titanate powder produced in Example S had a Sr-+ distribution
S coefficient (K~) of 16900 mLlg. The sodium titanate powder was subsequently bound with 20
wt % cellulose acetate. While Kd testing on the preliminary pellets were positive (Kd Ca 2500),
later pellets consistently tested low with Kd values less than 300. The extreme decrease in Kd
was due to decreased mass transfer through the cellulose acetate. Incorporating porosity in the
pellet using a binder with pore former, consistently recovered the Kd to values greater than 1600
mL/g.
The process to produce pellets with pore former consisted of dissolving cellulose acetate
(22.5 wt %) in an acetone (47.5 wt %)/formamide (30.0 wt %) mixture. Pores are formed by
a phase ir.version of the acetone/formamide/water system when the pellet is later extruded into
water. The solution was blended with sodium titanate at a weight ratio of 1:4 cellulose acetate
to ion exchanger. Once blended, the mixture was stirred to volatilize enough acetone to leave
a thick paste. The paste was extruded into a water wash bath containing 0.1% Triton XL-80N~
50% glycol and 25 % glycerol at 10~C. The purpose of the water bath is to leach the
formamide/acetone solvent and to minimize the pore collapse due to drying of the membrane
The low ~,npe-dtllre facilitates the precipitation of the polymer and freezing of the pore
structure. The surfactant reduces pore collapse upon drying by reducing capillary pressures.
The reason for the loss of mass transfer upon binding is apparent from Table 16. The
-60-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/1 6753
surface area of vir~,in sodium titanate is greater than 139 m2/g, but the extruded powder blended
with cellulose acetate and acetone showed very little surface area ( < 1 m2/g). The cellulose
acetate had blinded the surface of the sodium titanate powder. By using the pore former solution
and a cooled surfactant water bath, the surface area increased to greater than 91 m2/g.
5 Additionally, the wetability of the pellets using the new binder process also increased. The
pellets with pore former wetted more easily due to intact pores and presence of residual
surfactant in the pores.
CA 02235337 1998-04-17
WO 97/14652 PCTAJS96/16753
'~ c . -e ~~ ~ ~ e o
E ~ V V S~ x
o o
~, o ~ ~ -- o o
~; ~ o o o o
c~
c~ F oo V V ~' ~ oo
V~ --
o o o o
c , a --a
C O ~ ~ ~ ~ o, ~ C O~ ~ O'
3 a 3 oc~ 3 E V 3 ~
~~ oo ~~L,-o ~~~
~ o
- 62 -
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
EXAMPLE 22
This example demonstrates that heat treatment increases the mechanical strength of titania
bound sodium titanate. When p~tially crystalline sodium titanate is mixed with a hydrolyzable
titanium compound in air and stirred for about one hour, a paste like material is obtained due
S to the hydrolysis of TiP to form titania/titanium hydroxide. This material was then cured at
room teln~eldtu~e in air and then dried at 90~C overnight. After drying the material was ground
to below 40 mesh. The titania bound sodium titanate powder was col,.y~:ssed in a pelletizer to
a piece density of 1.8 - 2.2 g/ml. The pellets were then calcined at 200~C, 300~C and 400~C,
respectively. After 4 hours of thermal treatment, the pellets were crush tested using the ASTM
10 method D4174. The physical ~opelLies of the crushed product were combined with product
AW-300 zeolite, manufactured by Des Plaines, Illinois, UOP. The instrument used was an
Instron model 4502 with a 100 Newton load cell. The results are found in the following Table
17.
TABLE 17
Tip-bound Sodium Titanate Pellets AW-300 Pellets
Radial stress at maximum 519 PSI 343 PSI
Axial stress at maximum 2354 PSI --------
20A11 the calcined titania bound sodium titanate pellets have very good mechanical strength
with strong resistance against degradation in alkali solution. The strontium distribution constants
-63 -
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/167~3
(Kds) for these pellets were measured both as intact pellets and after crushing to 40-100 mesh.
These results are found in Table 18.
TABLE 18
s
Tip Bound Maeerial Strontium Kd
Calcined at 200~C, 40-100 mesh 21531 mL/g
Calcined at 200~C, intact pellets 24063 mL/g
Calcined at 300~C, 40-100 mesh 19079 mLJg
Calcined at 300~C, intact pellets 3661 mL/g
Calcined at 400 ~ C, 40-100 mesh 171 14 mL/g
Calcined at 500~C, intact pellets 20000 mL/g
Uncalcined, 40-100 mesh 51181 mL/g
Uncalcined, intact pellets 59014 mL/g
-64-
CA 02235337 1998-04-17
W O 97/14652 PCTAJS96/16753
EXA M PLE 23
Partially crystalline sodium titanate made according to the method of Example 5 was
added to a slurry of hydrous titania. Methyl cellulose was added as organic binder to form the
material before calcination. The mixture was extruded and calcined at 560~C for 4 hours. The
S calcined extrudates exhibited good mechanical strength, but powdered in alkali solution.
-65-
CA 02235337 1998-04-17
WO 97/14652 PCT/US96/16753
EXAMPLE 24
Partially crystalline sodium titanate made using the method described in Example S was
mixed with titania hydrate powder. The mixture was pelleted to a piece density of 1.8 - 2.2
gtml. The pellets were then calcined at 400~C and 560~C1 respectively. After 4 hours of
S therrnal treatment, calcined extrudates have reasonably good mechanical strength, but powdered
in alkali solution.
-66-
CA 02235337 1998-04-17
WO 97/14652 PCTAJS96/16753
EXAMPLE 25
The titania bound sodium titanate powder made using method describing in Example 13
was mixed with titania hydrate powder. The mixture was pelleted to a piece density of 1.8 - 2.2
g/ml. the pellets were then calcined at 400~C and 560~C, respectively. After 4 hours of
5 thermal treatment, all the pellets have very good mechanical strength with strong resistance
against degradation in alkali solution.
CA 0223~337 1998-04-17
Wo 97/14652 PCT/US96/16753
EXAMPLE 26
This example demonstrates a process of making titania bound crystalline sodium titanate
pellets using a motorized pellet former, which is facilitated by the presence of a solid lubricant,
namely stearic acid. The titania bound crystalline sodium titanate powder was mixed with 0.5%
- 4% of stearic acid and pelleti7~d with tablet press, Coulton 215, to a piece density of 1.8 - 2.2
g/ml. The pellets were then calcined at 400~C either in the air or in air saturated with moisture
at room t~-"pe,dture. After 4 hours of thermal treatment, all the pellets have very good
mechanical strength with strong resistance against degradation in alkali solution.
The strontium Kd was measured for three samples: intact pellets; pellets ground to 16-45
mesh; and to below 45 mesh. The intact pellets had a Kd of 11,860 mL/g. The 16-45 mesh
material has a Kd of 11,531, and the 45~ mesh material had a Kd of 12,813.
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
EXAMPLE 27
In this example, titania b~ound crystalline sodium titanate prepared by the method of
Example 13 is used to remove uranium from an aqueous solution.
200 mg of 20~0 mesh titania bound sodium titanate pellets were placed in a beaker. A
25ml solution comprising 50ppm uranium was contacted with the solid sodium titanate for 24
hours. The solution concentration of uranium after 24 hours was 3ppm. This tr~ncl~t~s to a
uranium Kd of over 1600 mL/g.
-69-
CA 02235337 1998-04-17
WO 97/14652 PCTrUS96/16753
EXAMPLE 28
This example describes a method for converting partially crystalline sodium titanate into
crystalline hydrogen titanate followed by testing the hydrogen titanate as an ion-exchanger.
10 grams of partially crystalline sodium titanate ple~ ed as in Example S was added to
S a liter solution of 0.1 N HNO3. The mixture was shaken for 18 hours during which time the
solution pH increased from about 1.38 to about 1.46. The solids were filtered from the solution,
recovered, and dried. The resulting crystalline hydrogen titanate was tested for its ability to
exchange various metal ions and the ion-exchange test results are reported in Table 19
-70-
CA 02235337 1998-04-17
W O 97/14652 PCT~US96/16753
TABLE 19
Hydrogen Titanate Ion Exchange Capacities
Target Metal Initial Concentration Kd (mL/g)
(ppm)
Yb 78 38900
Zr 51 12650
Mo 11 10900
Ag 19 9400
Ti 17 5566.667
Pb 5.6 5500
Cr 4.3 5280
V 20 4900
Fe 15 490~
Cs 19.7 4280
Sn 26 2790
Sb 20 2500
As 19 1800
Y 18 1190
Cd 20 614.2857
K 29 400
Hg 18 386.4865
Cu 24 352.8302
Ba 3.6 350
CA 02235337 1998-04-17
WO 97/14652 PCTAJS96/16753
TABLE 19 Cont.
Hydrogen Titanate Ion FYcll~nFe Capacities
Rb 19 273
Sr 20 156.4103
Zn 18 80
Pd 18 64
Co 25 17.05882
Ni 23 43.75
Al 26 30
Pt 18 29
Mg 19 26.3337
Solution pH = 2-3
