Removal of Uranium From Aqueous Wastes Using Elect PDF
Removal of Uranium From Aqueous Wastes Using Elect PDF
Removal of Uranium From Aqueous Wastes Using Elect PDF
net/publication/267701987
Article
CITATIONS READS
2 77
5 authors, including:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Elliot Kennel on 08 February 2016.
Shannon L. Stover
Master of Science
In
Chemical Engineering
2000
Shannon L. Stover
The presence of aqueous uranium wastes is a problem in the United States and
their treatment/disposal is desirable. Treatment methods have been developed but result
in concentration opposed to conversion to a disposable form. This technique involves
recovery of uranium as a solid, providing an advantage over other methods. The
technique utilizes carbon nanofibers as electrodes which successfully electrosorb uranium
ions. Fibers with varying surface characteristics were evaluated in the removal process
and all were determined to be equally and extremely effective. Various experimental
parameters were evaluated including applied potential, pH, and flow rate. The critical
applied potential at which significant removal is achieved is between –0.3 and –0.4 V.
Decreasing pH hinders the electrosorption process while increasing it enhances the
process. As expected, an increase in flow rates results in decreased removal. It was
determined that cyclic loading/unloading increased fiber performance and a capacity of at
least 5.45 guranium/gcarbon can be achieved. These results illustrate that this technique can
be effectively implemented to solve current waste management problems.
DEDICATION
you are truly a gift from God. Neither will you ever understand how much I
appreciate you staying awake and keeping my feet warm while writing.
iii
ACKNOWLEDGEMENTS
iv
CONTENTS
Abstract ............................................................................................................................ii
Dedication ........................................................................................................................iv
Acknowledgements...........................................................................................................iii
Contents............................................................................................................................v
List of Tables....................................................................................................................vii
List of Figures...................................................................................................................viii
Chapter 1: Introduction and Research Scope....................................................................1
Chapter 2: Literature Review ...........................................................................................4
2.1 Summary of Treatment Methods Currently in Use.....................................4
2.1.1 Coagulation-Filtration ..................................................................4
2.1.2 Lime Softening..............................................................................5
2.1.3 Reverse Osmosis Hyperfiltration .................................................6
2.1.4 Electrodialysis Process ................................................................6
2.1.5 Activated Alumina........................................................................7
2.1.6 Carbon Adsorption .......................................................................8
2.1.7 Ion Exchange.................................................................................9
2.2 Possibility of Using Electrochemical Removal Techniques .....................11
2.2.1 Electrosorption.............................................................................11
2.3 Related Electrosorption Techniques.........................................................14
2.3.1 Traditional Adsorption Processes ................................................14
2.3.2 Removal by Microbial Reduction.................................................17
Chapter 3: Theoretical Background..................................................................................19
3.1 Uranium Chemistry ...................................................................................19
3.1.1 Properties of Uranium Compounds...............................................20
3.1.2 Uranium Ions in Solution ..............................................................21
3.1.2.1 Trivalent Uranium.......................................................21
3.1.2.2 Tetravalent Uranium ...................................................21
3.1.2.3 Pentavalent Uranium...................................................23
3.1.2.4 Hexavalent Uranium ...................................................23
v
3.1.3 Uranium Species as a Function of pH...........................................27
3.1.4 Relationship Between pH and Electrochemical Potential.............29
3.2 Electrochemistry.......................................................................................32
3.3 Possible Electrosorption Mechanisms......................................................32
Chapter 4: Experimental Apparatus and Methods.............................................................34
4.1 Carbon Materials......................................................................................34
4.2 Reagents ...................................................................................................35
4.3 Electrolytic Cell and Fiber Loading .........................................................35
4.4 Experiments and Sampling Techniques.....................................................38
4.5 Analytical Methods...................................................................................39
4.6 Fiber Unloading........................................................................................41
4.7 FT-IR and X-Ray Diffraction Analyses....................................................42
Chapter 5: Results and Discussion ...................................................................................43
5.1 Performance Experiments.........................................................................43
5.2 Effect of Cell Potential on Fiber Performance..........................................53
5.3 Effect of Flow Rate on Fiber Performance...............................................55
5.4 Effect of pH on Fiber Performance...........................................................57
5.5 Effect of Cyclic Loading/Stripping of Fibers............................................59
5.6 Fiber Capacity..........................................................................................59
5.7 Possible Effect of Residual Iron...............................................................62
5.8 Attempt to Identify Surface Functional Groups Using FT-IR Analysis .....63
5.9 Identification of Uranium Solid By X-Ray Diffraction .............................64
5.10 Examination of Fibers Using SEM............................................................64
Chapter 6: Conclusions and Recommendations ................................................................67
6.1 Performance Experiments.........................................................................67
6.2 Variation of Parameters............................................................................67
6.3 Removal Efficiency and Fiber Capacity...................................................68
6.4 FT-IR, X-Ray Diffraction Analysis, and SEM..........................................69
6.5 Recommendations for Future Work ..........................................................70
References........................................................................................................................71
Appendix ..........................................................................................................................75
vi
LIST OF TABLES
vii
LIST OF FIGURES
viii
Effluent Concentrations.........................................................................................57
5-19. Effect of pH Variation on Fiber Performance.......................................................58
5-20. Effect of Cyclic Loading/Stripping on Fiber Performance....................................60
5-21. Evaluation of Fiber Capacity................................................................................61
5-22. Percent Residual Ash in Fiber Samples................................................................62
5-23. Results of X-Ray Diffraction Analysis .................................................................65
5-24. SEM of PR-1-AG Showing Smooth Surface.........................................................66
5-25. SEM of PR-19-HT Showing Surface Roughness Following Post-Treatment........65
ix
CHAPTER 1
INTRODUCTION AND RESEARCH SCOPE
2
loading each electrode with uranium to analyze its performance as a function of time
required to reduce the inlet concentration of uranium to acceptable levels. Each study
was performed at least twice to ensure accuracy and reproducibility. Following this
study, a select fiber was further tested to determine its absorptive capacity. An additional
performance experiment was also performed and involved the cyclic loading and
unloading of the fibers to identify any decrease in efficiency upon regeneration.
The second set of experiments was used to obtain further information on the
mechanism that makes this technique a success. These tests included several different
types of analyses. Experimental parameters including the pH and flow rate of the feed
and the cell potential were varied to determine their effect. Identification of the solid
precipitate mentioned previously was attempted through x-ray diffraction analysis of both
the precipitate itself and of a nanofiber sample that was loaded with the uranium
precipitate. An attempt was made to analyze the carbon nanofibers using Fourier
transform infrared (FT-IR) spectroscopy to identify any surface functional groups present
that may have been contributing to their effectiveness; however, these tests were
unsuccessful, the reason for which is discussed in more detail later in the report. Finally,
photographs were taken using a scanning electron microscope (SEM) in an attempt to
identify differences in the surface characteristics of the fibers.
The results obtained from these experiments were effective in demonstrating that
not only can the electrodes be reproduced and optimized, but that the potential for
commercial development of this technique in the future could prove to be very valuable
in the effort to restore and control the declining integrity of those sections of the world
that are battling with radioactive contamination in the form of uranium pollution.
3
CHAPTER 2
LITERATURE REVIEW
2.1.1 Coagulation-Filtration
The coagulation-filtration process involves the addition of a chemical substance,
typically iron or alum, which attracts uranium ions and/or complexes, resulting in larger
particles that are then easily filterable. A significant amount of laboratory testing has
been completed in this area and impressive removal efficiencies have been realized [6].
These efficiencies are dependent upon the coagulant dosage and the final pH of the
solution. Testing has been conducted on samples of pond water containing 83-ppb
uranium in the form of UO2(CO3)22- and UO2(CO3)34-, common species that are typical of
4
most carbonate-dominated surface and well waters. Upon the addition of 25 mg/L of
aluminum sulfate or ferric sulfate to the waste, removal efficiencies of more than 85%
can be achieved. Previous tests were successful at pHs of both 6 and 10; however, the
removal decreases significantly at pHs both below 6 and above 10. Success at a pH of 6
is due to the attraction of the uncharged UO2CO3 complex (the dominant uranyl complex
at this pH) to the relatively uncharged aluminum or iron oxyhydroxide surface.
At pH 10 it is believed that CaCO3 precipitation caused a decrease in the
carbonate concentration that results in the formation of (UO2)3(OH)5+ complex, which in
turn is attracted to the negatively charged coagulant surface. Below a pH of 6 the
coagulant does not form an appreciable amount of floc while at pHs above 10 the
precipitant becomes increasing soluble. Both of these phenomenon result in a decrease in
removal efficiency. These results have also been duplicated, reinforcing the finding that
uranium can be efficiently removed by coagulation-filtration techniques [7]. Although
this technique is effective, continuous costs are incurred due to the required chemical
addition. The technique described also requires a settling time prior to filtration that
results in the need for large storage tanks and a batch-type process. The process could
possibly be run in a continuous manner but would require large continuously stirred tank
reactors, increasing the capital investment.
5
carbonate precipitates removed only 10 to 30% of the uranium while at pHs above 10.6,
the addition of magnesium raised removal efficiencies to between 93 and 99%. It was
concluded that the presence of Mg(OH)2 precipitates at pHs greater than 10.6
significantly increases uranium removal via lime softening. Despite the success of this
technique it too, like coagulation-filtration, incurs a continuing cost for the addition of
lime and possibly other chemicals such as MgCO3 added to enhance performance. This
technique would also have to be performed in a batch-type process, proving it to be
unfeasible for large volume, low concentration wastes. Also like coagulation-filtration, if
this process were to be run in a continuous manner, large reactors would be required.
6
involves an electrochemical separation process in which ions are driven through anion-
and cation- selective membranes from a less concentrated to a more concentrated solution
as a result of the flow of direct electric current. The electrodialysis reversal (EDR)
process now in common use is a refinement of the ED process, in which depleting and
concentrating channels are switched and the polarity of the electrodes is reversed
typically four times per hour to ionically backflush the membranes in order to minimize
scaling and fouling [9].
A recent study in Chimney Hill, Texas, demonstrated that ED can be used to
effectively treat groundwater contaminated with uranium [9]. It was determined that by
using ED, uranium removal efficiencies of more than 95% can be achieved when treating
water containing 120 ppb uranium during a continuous 28-day test. However, it was later
determined that the uranium removed was not removed by ion transport as is typical with
ED processes, but by accumulation of uranium in the anion membranes of the ED stack.
Therefore, this process is not desirable when typical anion exchange resins are capable of
achieving the same effects while utilizing a less complicated process.
7
concentrating the waste which must then be treated in some other way to obtain a solid
form, increasing the overall cost of the process as a whole.
8
2.1.7 Ion Exchange
Ion exchange has proven to be the most effective and economical treatment
technique, therefore making it the most popular. Both cation and anion exchange resins
have been used for uranium removal; however, the anion exchange resin has proven to
achieve removal efficiencies superior to those of cation resins. Several cationic resins
have been tested including H+, Na+, and Ca2+ forms [13]. The hydrogen form cationic
resin resulted in 93 to 97% uranium removal with the effluent pH equal to 3.5. Both the
calcium and sodium forms were unsuccessful at a pH of 8.2; however, at a pH of 7, the
sodium-form resin achieved 85% removal while the calcium form was still unsuccessful.
The calcium-form resin did not begin to remove uranium until a pH of 5.6 was utilized.
The removal efficiencies continued to increase with decreasing pH. By the time a pH of
4 was reached, removal efficiencies of 60 to 65% were realized. The sodium-form resin
maintained a removal efficiency of 70% at all lower pH values.
The abovementioned results demonstrate the strong dependence of uranium
removal on both the pH of the feed and the form of the resin, indicating the importance of
the uranium speciation in solution. The varying efficiencies of the resin forms are related
to the selection sequence of the cations. Although the cation resins are considered to be
somewhat effective, the requirements for a low feed pH and the low removal efficiencies
indicate that the process is not feasible for treatment of drinking water.
Anionic resins result in much higher removal efficiencies when used to treat
contaminated water. It has been suggested that anionic resin treatment would be the only
cost effective treatment method for small community water supplies [12]. Both bench-
and pilot-scale studies have shown that the strong-base anionic resins exhibit very high
capacity (up to 60,000 BVs) for the removal of the uranyl carbonate complexes including
UO2(CO3)22- and UO2(CO3)34-. One test, using a feed of water containing 22 to 104 ppb
uranium, resulted in treating 8,000 to 60,000 bed volumes before exceeding 1.0 ppb in
the effluent [14]. Another study using anionic resins also proved successful and achieved
70% removal after a 4,000 BV throughput [13]. Additional studies using real waste
samples containing 300 ppb uranium resulted in successfully treating over 9,000 bed
volumes before uranium was detected in the effluent [11]. A final study involved
analyzing two forms of resin including Dowex 21K and Ionac 641, which were operated
9
to 17,400 BV and 31,300 BV, and still removed 95% and 90%, respectively, of the total
uranium in the feed [10]. The results of these tests indicate that anionic resins display a
high selectivity for the charged uranyl carbonate complexes commonly found in
groundwater.
Due to the excellent performance of anionic exchange resins and the ability to
easily regenerate the materials, this technique has become the EPA’s treatment of choice
in the remediation of uranium laden aqueous wastes. The decommissioning and
treatment of a weapons grade production facility located in Fernald, Ohio illustrates one
example of the effectiveness of anionic exchange resins. The US Department of Energy
and Flour Daniel are each contributing to the clean up of the site, which involves
contaminated soils and structures, resulting in contaminated surface and subsurface
waters. The subsurface contamination involves the contamination of a small portion of
the Great Miami Aquifer, one of the largest sources of drinking water in the country. The
DOE has implemented a pump-and-treat system to treat the contaminated section of the
aquifer and inject the treated water back into the aquifer. The process involves the
treatment of 2,900 gpm of water contaminated with between 60 to 1,000 ppb uranium in
the form of complex uranium carbonate ions. The waste is treated using an anionic
exchange resin that successfully reduces the uranium concentration to below acceptable
limits (i.e., 20ppb).
The columns are regenerated approximately once a month, producing
concentrated (300ppm U) brine solution that must then be treated. The pH of the brine is
increased through the addition of lime and the uranium is precipitated and recovered
through filtration. The supernatant liquid, which contains small concentrations of
uranium, is recycled to the front of the process and bled slowly into the fresh waste
entering the ion exchange columns.
Although this technique is very effective and combines the best features of two
methods, ion exchange and precipitation/filtration, a treatment process that achieves the
same results in one step would be desirable. A successful process, in which uranium is
removed from liquids and recovered in a solid form, without the addition of chemicals to
adjust pH, would prove to be ideal.
10
2.2 Possibility of Utilizing Electrochemical Removal Techniques
Treatment of aqueous wastes containing metals by electrochemical reduction is an
attractive possibility due to the economic advantage of recovering valuable metals [1].
However, while this technique is effective for several metals including copper, lead,
cadmium, and nickel, it is not useful for uranium. The reduction potential required to
reduce the uranyl ion (UO22+) to elemental uranium is between -1.79V (dc) and -2.11V
(relative to a Ag/AgCl electrode), which is significantly more negative than the reduction
potential of water, which is -0.83V (dc) [15]. It is for this reason that the direct reduction
via electrodeposition of uranium cannot be accomplished and alternative electrochemical
methods such as electrosorption are being investigated.
2.2.1 Electrosorption
One cousin to electrodeposition, known as electroadsorption or simply
electrosorption, involves the adsorption of UO22+ onto a negatively charged electrode
surface. The technique does not involve the reduction of uranium to its elemental state,
eliminating the aforementioned difficulties associated with electrodeposition. The
negative potential placed on the electrode is significantly higher than the natural negative
surface charge formed as a result of the dissociation of oxygen functional groups present
on the surface. This technique enhances the adsorption capacity of the electrode and
makes it possible to take advantage of the increased surface areas and electrical
conductivity of some forms of carbon and graphite. It is possible that the success of this
technique could result in a minimally polluting, energy-efficient, and potentially effective
alternative to current treatment methods [16].
Electrosorption is a technique in which low concentration aqueous wastes can be
treated and the wastes accumulated in a solid form as opposed to a concentrated aqueous
one. Upon researching this technique, it became apparent that there are no parallel
investigations in this area based on publications in past or recent literature. It appears
that the WVU/ASI group may have conducted the only research in this area to date,
which has paved the way for the work presented herein [4].
Previous studies at WVU involved the use of carbon nanofibers produced using
an iron catalyst [4]. Two fibers were tested, one which was tested “as-grown” with no
post fabrication treatment and another which was oxidized following fabrication to
11
increase the amount of oxygen-containing functional groups of the surface. Through
experiments utilizing a small electrolytic cell, a flow rate of 0.7 mL/min, and a feed
concentration of 100ppm uranium in the form of UO22+ (in 0.1M KNO3 to increase the
ionic strength of the solution), it was determined that the post-oxidized fibers were
significantly more effective than the unoxidized fibers. Additional studies were
conducted to determine the effects of altering pH, flow rate, and applied potential.
Results were affected by all three factors. The process is ineffective at pHs below 3.5
while it is enhanced at all pHs above 3.5. As expected, the removal efficiency decreased
with increasing flow rate. Finally, a unique dependence on the applied potential was
established. The fibers were ineffective at potentials less negative than –0.45 V;
however, upon achieving the potential of –0.45 V, the increase in fiber performance was
significant and continued to increase with increasing potential. Upon reversing the
potential, the uranium was obtained in a solid form that has yet to be identified.
Although no information on electrosorption of uranium is available, aside from
the previous research at WVU, research on the electrosorption of other ions has been
published. One example involves the electrosorption of hexavalent chromium ions
(present as HCrO4-, CrO42-, and Cr2O72-) from groundwater through the use of carbon
aerogel electrodes [16]. The electrodes have a high surface area, typically between 400
and 1100 m2/g, and an exceptionally low electrical resistivity (less than or equal to 40
mÙ· cm). Upon applying a potential to the column, it was discovered that the ions were
removed from the electrolyte and electrosorbed onto the electrode surface. Influent
concentrations were reduced from 35 ppm to 2 ppm, well below the acceptable level of
11 ppm, as per surface water regulations. It is believed that the Cr(VI) separation was not
the result of simple double-layer charging but chemisorption, a process reversed by
cathodic polarization.
The electrosorption technique has also proven to be successful in the removal of
lithium ions from aqueous solutions using chemically and physically activated carbons
[17]. The performance of the carbon electrodes was a strict function of their
physicochemical properties including pore volume, pore size distribution, and surface
area. Testing included runs during which electrochemical polarization was present and
runs where it was not. Results indicated that the cathodic treatment of the activated
carbons significantly enhanced the adsorption of the lithium ions.
12
Another test has been conducted that involves both the use of electrosorption and
electrodialysis using an ion exchange resin [18]. The technique was tested for its
effectiveness in removing Ni2+ and Cu2+ from synthetic wastes that were designed to
imitate electroplating effluents. The tests involved placement of 5.0 grams of an ion
exchange resin into the desalination compartment of an electrodialysis cell containing a
10cm2 membrane. The feed contained 10 mg-eq/L NaCl and a metal ion (Cu2+ or Ni2+) to
other ion (Na+) ratio of 3:1. The pH and linear flow rate of the feed were 5 and 0.1 cm/s,
respectively. Tests were conducted both with and without the presence of an electric
current (10 mA/cm2). Ion removal in the presence of the electrical current was found to
be six times that of the experiment with the absence of current. The increase in removal
efficiency was attributed to electrochemical reactions taking place in the resin/solution
boundary layer that resulted in the formation of insoluble metal hydroxides. It should
also be noted that the technique was more effective for the treatment of Cu2+ ions than for
Ni2+ ions. Although this technique appears somewhat promising, it has not yet been
proven effective for the treatment of uranium contaminants.
Tests have also been conducted on the removal of organic compounds from
aqueous solutions using electrosorption onto GAC [19],[20],[21]. These studies include
the adsorption and electrosorption of chloroform onto three types of GAC. The results
are dependent upon operating parameters, including applied potential, temperature, pH,
contaminant concentration, and chemical properties of the GAC. It was determined that
an increase in the negative potential, increases in initial chloroform concentration, and
decreases in temperature resulted in an enhanced affinity for chloroform while the
imposition of a positive potential and neutral pH values inhibited sorption. Although
previous studies have shown GAC to be inferior in uranium removal, it is undetermined
whether the presence of an applied potential would increase performance capabilities.
The abovementioned studies, in conjunction with previous testing conducted at
WVU, provide significant reason to believe that electrosorption-type technologies can be
developed and implemented for uranium removal. The technique would provide uranium
treatment that is economically competitive with treatment using anionic exchange resins;
however, proposed electrosorption techniques result in the acquisition of uranium from
solution in a solid form, eliminating the need for post treatment techniques following
13
regeneration. The implementation of this process would provide an economical
advantage to other removal methods and also fewer process complications.
14
a very high affinity for uranium and is stable in aqueous environments.
Several naturally occurring substances have also been tested for their
effectiveness in treating uranium. Clays have been used for several years in ion exchange
applications and have recently been tested for their effectiveness in uranium removal
[24]. It was determined that the performance of these clays was not only a function of pH
of the feed solution, but also of the concentration of humic substances and the clay-to-
solution ratio. The presence of humic substances enhances uranium adsorption at low pH
due to the formation of a clay-humate complex.
When considering natural materials, the use of iron nodules to treat uranium is
worth mentioning [25]. The iron nodules, which have proven effective in scavenging
uranium from groundwater, have a high capacity for uranium uptake. Experiments show
that a uranium enrichment of 8% (by weight) can be achieved, much higher than that of
other forms of iron including fissure fillings and clay coatings. The mechanism is
believed to involve an initial adsorption step followed by precipitation.
In several cases of environmental pollution, high ionic strengths and high pHs are
present in conjunction with uranium pollution such as in the case of storage tank liquids
and leachates from vitrified, saltstone, and grouted waste forms. Studies have been
completed on the effect of these conditions on the mechanism of treatment methods
involving adsorption onto natural sediments containing carbonate mineral phases [26]. It
was determined that neither the concentration of uranium nor the ionic strength affect the
adsorption mechanism; however, removal efficiencies greatly increased when the pH was
increased from 8.3 to 9.3. When the pH surpassed 10.3, precipitation of uranyl solids
occurred causing removal efficiencies to increase dramatically but due to precipitation
rather than adsorption; and as stated previously, precipitation is a costly process for large
quantities of low concentration wastes due to the required chemical addition and storage
requirements.
The above-mentioned studies involved the removal of uranium from aqueous
wastes where other contaminants were not present; however, several studies have also
been completed that involve the removal of uranium in the presence of other
contaminants. One study involved the removal of uranyl ions from spent nuclear fuel
[27]. An innovative anion exchange resin (AR-01) was utilized. The material is
composed of a resin embedded in silica beads with benzimidazole functional groups
15
present on the surface. The majority of fission products including Cs(I), Sr(II), Mo(VI),
Rh(III) and trivalent rare earth metals were immune to adsorption while Ce(IV) was
strongly adsorbed. It should be noted, however, that Ce(IV) was eventually reduced to
Ce(III) by the resin, a non-adsorptive form. Zirconium (IV) also presented weak
adsorption and partially mixed with U(VI). Both Ru(III) (in the form of anionic
nitrosylnitrato-complexes) and Pd(II) exhibited strong adsorption.
Several types of bacteria have also been tested for their effectiveness in the
uranium adsorption process. One type of bacteria, scenedesmus obliquus 34, was found
to exhibit an energy independent sorption of uranium, even in the presence of other
metals [28]. These metals included Cu2+, Ni2+, Zn2+, Cd2+, and Mn2+, each of which
competed only slightly with the uptake of uranyl ions. A maximum capacity of 75
mguranium/gsorbent (dry weight) was achieved and the results could be modeled using the
Freundlich adsorption isotherm. Increases in electrolyte concentrations increased the
affinity for uranium sorption while pretreatment of the cells with NaOH, NaCl, ethanol,
or heat resulted in only a slight decrease. Treatment of the bacteria with a buffer of pH 4
results in desorption. The uranium uptake using this substance is believed to be
attributed to active groups or capillary action in the cell wall capturing the uranium.
An immobilized biomass known as Rhizopus arrhizus has also been tested for its
ability to adsorb uranium [29]. Tests were conducted using both synthetic and industrial
wastes. It was discovered that when using influents with concentrations at or below 500
ppm, effluent concentrations of as low as 1 ppb were obtained. Breakthrough of the
column began at 57 bed volumes and continued through 76 bed volumes. It is also
possible to regenerate the biomass and obtain highly concentrated solutions of uranium
through elution. Use of synthetic uranyl nitrate solutions resulted in no deterioration of
the biomass performance over twelve successive sorption-elution cycles; however,
treatment of mine leachate solutions resulted in an 18% reduction in performance after
only four cycles. It should be noted that the presence of additional ions, particularly
aluminum, hinders the effectiveness of the technique.
Another immobilized biomass, Citrobacter, has also been studied for its use in
treating acidic uranium drainage waters [30]. Samples of the acidic solutions containing
35 ppm uranium were obtained from the National Uranium Enterprise (ENUSA, Spain)
and were treated by the biomass. The samples were supplemented with 5-mM glycero-2-
16
phosphate and adjusted to a pH of 4.5. The biomass was successful in removing up to
50% of the uranium by maintaining a residence time of 1.4 hours (resulting in a flow rate
of 50 mL/h). Although the technique is effective, the substrate used is expensive, making
the process economically unfeasible. A more feasible substrate has yet to be identified.
Brewery yeast has also proven successful in the removal of uranium from aqueous
wastes in a process that combines biosorption and precipitation [31]. Both unwashed and
washed biomass samples were evaluated. Testing of the unwashed yeast resulted in a
maximum capacity of 360 mguranium/gsorbent (dry weight). It was determined that washing
the yeast reduced the capacity to 150 mguranium/gsorbent. An additional experiment involved
the retention of the unwashed biomass by a semi-permeable membrane. It was
discovered that 40% of the uranium present precipitated outside the membrane, indicating
that a significant portion of the uranium removal was due to precipitation induced by low
molecular weight compounds loosely associated with the biomass.
Another group of immobilized biomass products has been analyzed using several
different polymers as a binder material [32]. These polymers included calcium alginate,
polyacrylamide, polysulfone, and polyurethane. Polyurethane-based materials were
determined to be superior to the other candidates and production of pseudomonas
aeruginosa CSU in the form of spherical beads of uniform size has already been
accomplished at the pilot plant scale. The immobilized biomass was evaluated using a
batch sorption isotherm approach including both a stirred-tank approach and continuous
loading/elution in an up-flow, packed-bed column. Breakthrough results indicate that P.
aeruginosa CSU is effective for removal of uranium from acidic, low-concentration
wastewater and achieved loading capacities of 97 mguranium/gdry biomass. Although the
technique is effective, it takes over an hour to become so.
17
There has been much discussion that the microorganisms play an indirect role by
producing reduced compounds that are then used to reduce U(VI) in an abiological
reaction or that cell walls provide a surface for abiological reduction. Despite these
criticisms, work continues in this area and is summarized below.
An initial study determined that Fe(III)-reducing organisms obtain energy for
growth from electron transport to U(VI) [33]. This type of reduction can occur much
faster than typical reduction techniques. The specifics of this mechanism may explain the
deposition of uranium in aquatic sediments and aquifers and is suggested as a biological
technique for the remediation of uranium in the environment.
Following initial studies, several more tests were completed on the potential use
of biological techniques to treat uranium. One study suggested that sulfate-reducing
microorganisms may contribute to the reduction of U(VI) in sediments [34]. The
experiments involved the reduction of U(VI) in the presence of washed cells of sulfate-
grown Desulfovibrio desulfuricans in a bicarbonate buffer using lactate or H2 as an
electron donor, without which no reduction occurred. Applying heat prior to reduction
also inhibited treatment due to termination of the cells. However, exposure to air did not
effect D. desulfuricans’ ability to reduce U(VI). Attempts were made to grow the
bacteria using U(VI) as the electron acceptor but were unsuccessful.
Additional studies were conducted on microbial reduction of U(VI) in an attempt
to describe the kinetics of the technique [35]. A pure culture of Shewanella alga strain
BrY was used to reduce U(VI) under non-growth conditions where lactate was used as
the electron donor. Initial U(VI) concentrations ranged from 13 to 1,680 ìmol/L.
Reduction rates were measured and used to determine the maximum specific U(VI)
reduction rate (2.37 ìmole -U(VI)/(mg biomass⋅h)) and the Monod half-saturation
coefficient (132 ìM -U(VI)). The reduction of U(VI) was maintained at a minimum of
60% of this rate throughout at least 80 hours of the experiment. Results also showed that
oxygen present initially delays but does not inhibit U(VI) reduction. Although BrY only
reduces U(VI) by 30% of the rate at which it reduces Fe(III), it is always comparable but
most often surpasses the performance of other metal reducing species.
18
CHAPTER 3
THEORETICAL BACKGROUND
19
3.1.1 Properties of Uranium Compounds
Uranium is an extremely dense metal, closely resembling steel in appearance. It
can be found with three different crystalline modifications including alpha, beta, and
gamma structures. A summary of the physical properties of elemental uranium is
presented in Table 3-1.
Three uranium oxides are known to exist for certain including the dioxide (UO2),
the trioxide (UO3), and the mixed oxide (U3O8). It has also been speculated that the
existence of a fourth oxide, the monoxide (UO), also exists but is not widely observed
[38]. The properties of the three common oxides are summarized in Table 3-2.
The affinity of uranium oxides for water increases from UO2 to U3O8 to UO3.
The hydrate of uranium dioxide (UO2⋅nH2O) is obtained from the reaction of ammonia or
20
alkalis with a solution of a U(IV) salt, or from the hydrolysis of dilute solutions of U(IV)
chloride or acetate. In air, the hydrate is easily oxidized to UO3⋅H2O. After drying over
sulfuric acid, the hydrate has the composition UO2⋅2H2O. The freshly prepared U(IV)
hydroxide is readily soluble in acids, but its solubility decreases upon standing [15].
Several hydrates of UO3 also exist including UO3⋅H2O (H2UO4), UO3⋅2H2O
(H4UO5), and 2UO3⋅H2O (H2U2O7). These hydrates are much more stable than the
hydrates of the dioxide and are only slightly soluble in water [15].
21
type of environment is more conducive to qualitative reduction. Uranium (IV) is
typically present in solution as the simple ion U4+ or possibly as U(H2O)n4+, where n = 6
or 8 [15].
The presence of U(IV) salts results in an acidic solution caused by hydrolysis.
The reaction of the hydrolysis is shown below.
This reaction, when extensive, can cause the formation of poly-nuclear ions of the type
U[(OH)3U]n4+n and also polymers [U(OH)4]x [40]. Reaction 3-1 is strongly temperature
dependent as shown in Table 3-4 which presents hydrolysis constants (Ka) at various
temperatures [41].
Uranium (III) is often formed throughout the reduction process but is easily and
quickly oxidized to U(IV) by atmospheric oxygen. A mercury cathode can be used to
reduce U(VI) to U(IV) electrolytically and quantitatively [42]. Reduction can also be
achieved through a photochemical reaction with alcohol [43], ether [44], or lactic acid
[45]. Finally, U(IV) can also be obtained by reduction of uranyl nitrate with rongalite
(CH2O⋅NaHSO2⋅2H2O) in nitric acid [46].
Cold, acidic solutions of U(IV) are fairly stable in darkness; however, it has been
shown that its oxidation in air is significantly increased with exposure to light,
particularly direct sunlight or ultraviolet light. The reaction proceeds as shown below.
22
A quantum of light is indicated by hv and an asterisk indicates an excited uranium ion
[47].
The equilibrium constant of the above-detailed process is (1.7 ± 0.3)×106 [50]. Uranium
(V) is most easily maintained in solution with little disproportionation when the pH is
kept between 2 and 40.
24
Table 3-6. Activity Coefficients of the Uranyl Ion as a Function of the Ionic Strength of the Solution [52]
Ionic Strength of Activity coefficient Ionic Strength of Activity coefficient
Solution, µ of UO22+, ã Solution, µ of UO22+, ã
0.002 0.868 0.500 0.430
0.003 0.858 0.600 0.440
0.004 0.830 0.900 0.500
0.007 0.794 1.200 0.600
0.012 0.699 1.500 0.740
0.022 0.608 1.800 0.920
0.042 0.567 2.100 1.160
0.080 0.488 2.400 1.490
0.109 0.473 2.700 1.950
0.139 0.451 3.000 2.570
0.201 0.425 3.600 4.610
0.261 0.422 4.200 8.350
0.335 0.462
25
[UO2(OH)]+. Upon further addition of alkali, increasing the ratio to above 1, a
condensation process sets in and causes an increase in the number of polymeric, singly-
charged ions (UO2(UO3)nOH+). Still further addition of alkali, increasing the ratio to
between 1.4 and 1.6, results in the formation of a metastable colloidal solution containing
unstable polymers of [UO2(UO3)nOH]+ that are unstable and therefore converted to
colloidal uranyl hydroxide. This colloid decomposes and UO3⋅nH2O is precipitated.
Once the ratio reaches values between 1.4-1.6 and 1.9-2.0, uranyl hydroxide is
precipitated. Finally, the further addition of alkali causes the hydroxide to be converted
to uranates and polyuranates. Table 3-8 shows precipitation pH as a function of UO22+
concentration [52].
[UO OH + ] [OH − ]
SP1 = = 1.3 ⋅ 10 −12
2
[UO (OH ) ]
3-8
2 2
[UO ] [OH ]
2+ − 2
[UO (OH ) ]
3-9
2 2
Recently, hydrolysis of the uranyl ion and the dependence of the hydrolysis on
both temperature and ionic strength of solution have been studied [57]. It is still
maintained that the phenomenon is explained by the formation of the monomer
[UO2OH]+ according to the reaction
26
The hydrolysis constants are also shown below
=
[(UO OH ) ] [H ]
2
+ +
[UO ]
K1 2+
3-12
2
K2 =
[(UO 2
⋅ UO 3 )
2+
] [H ]+ 2
[UO ]
3-13
2+ 2
2
and the temperature and ionic strength dependence are illustrated in Table 3-9.
Table 3-9. Temperature and Ionic Strength Dependence of Hydrolysis Constants of UO22+
Conditions (ionic strength and temp.) K1 K2
µ = 0.347; 25°C 4⋅106 1.5⋅106
µ = 0.0347; 25°C 1.5⋅106 0.7⋅106
µ = 0.0347; 40°C 8⋅106 1.2⋅106
27
effects of a change in oxidation due to stronger acids (concentrated nitric acid) is
neglected. These points are however discussed later in the section that addresses the
relationship between potential and pH.
28
Figure 3-2 illustrates that the chemistry of U(VI) is more complicated than that of
U(IV). At pHs between 4 and 4.5, nearly all species are present in equilibrium with each
other. At lower pH, the dominant species is UO22+ while at higher pH (UO2)3(OH)5+ is
prevalent. Also, although not noted in this figure, several uranium hydrates are formed at
pHs above 6 including UO3⋅H2O which precipitates from solution. These issues are also
addressed in the following section.
29
Table 3-10. Potentials for Oxidation/Reduction Reactions of Uranium Species [58]
Reaction Potential, V
Two Dissolved Substances
U3+ = U4+ + e- E0 = -0.607 + 0.0591 log (U4+/U3+)
U + H2O = UOH3+ + H+ + e-
3+
E0 = -0.538 – 0.0591 pH + 0.0591 log (UOH4+/U3+)
U3+ + 2H2O = UO2+ + 4H+ + e- E0 = 0.612 – 0.2364 pH + 0.0591log (UO2/UOH3+)
UOH3+ + H2O = UO2+ + 3H+ + e- E0 = 0.546 – 0.1773 pH + 0.0591 log (UO2+/U3+)
U3+ + 2H2O = UO22+ + 4H+ + 2e- E0 = 0.333 – 0.1182 pH + 0.0591 log (UO22+/U4+)
UOH3+ + H2O = UO22+ + 3H+ + 2e- E0 = 0.299 – 0.0886 pH + 0.0591 log (UO22+/UOH3+)
UO2+ + UO22+ = e- E0 = 0.052 + 0.0591 log (UO22+/UO2+)
Limits of the Domains of Predominance
3+ 4+
U /U E0 = -0.607
U3+ / UOH3+ E0 = -0.538 – 0.0591 pH
U4+ / UO2+ E0 = 0.612 – 0.2364 pH
UOH3+ / UO2+ E0 = 0.546 – 0.1773 pH
U4+ / UO22+ E0 = 0.333 – 0.1182 pH
UOH3+ / UO22+ E0 = 0.299 – 0.0886 pH
UO2+ / UO22+ E0 = 0.052
One Dissolved Substance and One Solid Substance
U4+ + 2H2O = UO2 + 4H+ NA
UOH3+ + H2O = UO2 + 3H+ NA
UO22+ + H2O = UO3 + 2H+ NA
a.) E0 = -0.382 – 0.2364 pH – 0.0591 log (U3+)
U3+ + 2H2O = UO2 + 4H+ + e-
b.) E0 = -0.019 – 0.2364 pH – 0.0591 log (U3+)
a.) E0 = 0.221 + 0.0295 log (UO22+)
UO2 = UO22+ + 2e-
b.) E0 = 0.040 + 0.0295 log (UO22+)
U3O8 + 4H+ = 3UO22+ 2H2O + 2e- E0 = -0.403 + 0.1182 pH + 0.0886 log (UO22+)
a.) Uranous oxide, b.) Uranous hydroxide
30
Figure 3-3 (a & b). pH/Potential Relationship for Uranium Species [58]
31
3.2 Electrochemistry
As stated previously, this study involves electrosorption, not electrodeposition;
however, an understanding of classical electrochemistry provides a basis for speculation
on the possible mechanism that is responsible for the success of this uranium removal
technique. Two possible mechanisms have been proposed in previous studies and will be
discussed in more detail in the following section.
An electrochemical cell consists of two electrodes, an anode and a cathode. Each
electrode is submersed in a suitable electrolyte solution. An electrochemical cell can be
used to measure the current being generated in the cell by a reaction that is taking place
or an external source of energy can cause a reaction to take place in the cell. Regardless
of the situation, the two electrodes must be connected to one another externally and
submersed in electrolyte solutions that are also in contact. This allows the flow of
electrons to occur by two separate techniques; one outside the cell via electrical
conductivity through wires and the other inside the cell through the migration of electrons
through the solution [59].
As stated previously, an electrolytic cell can be connected to a voltmeter that can
measure the amount of energy generated by a chemical reaction; or, as in the case with
the present study, the cell can be connected to a potentiostat that induces a potential
across the electrodes and initiates a chemical reaction. Regardless of the purpose or use
of the cell, an electric current will flow. This can happen by three distinct processes
including the movement of electrons through the external connection (i.e., a wire,
voltmeter, or potentiostat), migration of cations and anions within the electrolyte solution,
and an oxidation or reduction reaction that takes place at the surface of one of the two
electrodes. The electrode where the oxidation reaction (electrons released) occurs is
known as the anode while the reduction reaction (electrons consumed) occurs at the
cathode [59].
32
understood due to its case sensitive nature as a result of the interactive forces present at
the surface of the electrodes. It is however known that electrosorption or
electroadsorption as it is sometimes called is the result of an electrically enhanced
adsorption process. On this basis, two possible mechanisms have been proposed and are
summarized below [4].
The first of the two proposed mechanisms entertains the possibility of
electrosorption as the result of the formation of double-layer charging of the carbon
electrode. It is known that an electrical double layer exists at the interface between a
porous carbon electrode and an electrolyte solution. The layer has an abundance of
cations or anions on the solution side. These ions are adsorbed onto the electrode surface
or remain in the diffuse double layer; regardless, no Faradaic (i.e., no oxidation or
reduction) reactions occur during the adsorption process. Simple calculations can predict
the adsorptive capacity of a given electrode material based on surface area, applied
potential, and double-layer capacitance. Based on these calculations, the amount of
uranium expected to be adsorbed was negligible compared to that actually observed,
indicating another reaction is occurring either in place of or in addition to the double
layer formation.
The second proposed mechanism was one that mimics an ion-exchange type
sorption process in which surface functional groups play an important role. This
explanation proposes that the induced negative potential causes a Faradaic reaction,
producing hydrogen and hydroxyl ions, resulting in the dissociation of acid groups on the
carbon surface. The dissociated acid groups then react with the uranyl ions, causing them
to be bound to the carbon surface. This technique therefore depends on the presence of
surface functional groups on the carbon nanofiber electrode, supporting the highly
effective oxidized ASI fibers; however, this technique also has a limited capacity and
therefore does not explain removal efficiencies. Therefore, a separate or additional
phenomenon must be identified to explain the success of this process.
33
CHAPTER 4
EXPERIMENTAL APPARATUS AND METHODS
The following section describes in detail the apparatus and techniques that were
involved in performing the abovementioned studies.
34
Table 4 - 1. Inherent Fiber Properties
PR-23-HT -- -- -- -- -- --
PR-24 -- -- -- -- -- --
Values in parentheses estimated based on AG fiber values. Data provided by ASI, missing data not available.
4.2 Reagents
Various solutions of uranium were prepared by diluting a stock solution of
UO2(NO3)2 (10,000 ppm, Plasma Standard, SPEX Industries, Inc., Edison, NJ) with a
0.1-M solution of potassium nitrate and deionized water. The deionized water was
purified using a NANOpure ultra-pure water purification system (Barnstead-
Thermolyne, Dubuque, IA). All other chemicals were certified ACS grade. The pH of
all solutions was adjusted to 3.5 (Accumet pH meter, model AR20, Fisher Scientific),
using both potassium hydroxide (KOH) and nitric acid (HNO3). The pH was maintained
at 3.5 due to the results of a previous study showing the dependence of performance on
pH [4]. This relationship is discussed in detail in Chapter 5.
35
reactions occur at the working and auxiliary electrodes while the reference electrode
establishes a basis to which other potentials are referenced. The reference electrode used
here was a silver/silver chloride (Ag/AgCl) electrode (Bioanalytical Systems Inc., Part
No. MF-2021, West Lafayette, IN). The auxiliary electrode was constructed of platinum
wire bent into a coil. The carbon fibers were placed into the cell between a platinum
mesh connected to a platinum wire that served to provide proper electrical contact to the
fibers. A layer of filter paper was added above the fibers to keep them from being eluted
from the cell. A schematic of the cell illustrating all components can be seen in Figure 4-
1.
To minimize the voltage drop across the cell, the height of the nanofiber bed was
kept small (approximately 2 cm although this varies with each type of fiber) and the
fibers were kept under a mild compressive force. Also, the distance between the three
electrodes was kept to an absolute minimum.
36
After being loaded with the carbon nanofiber material, the cell was connected to a
potentiostat (Model PWR-3, Bioanalytical Systems, Inc., West Lafayette, IN) that applied
a constant voltage to the carbon electrode. It should be noted that all subsequent voltage
values are always referenced to the Ag/AgCl reference electrode. A peristaltic pump
(Model 7518-60 (driver model 7521-50), Cole Parmer Instrument Company, Vermont
Hills, IL) was used to transport the aqueous uranium solution through the cell at specified
flow rates. Samples were taken at various times that were predetermined according to the
type of experiment being performed. These experiments are discussed in the following
section. The voltage was monitored during the experiments by means of a digital
voltmeter (Keithly 2700 multimeter/data acquisition system, Integra Series, Keithly
Instruments, Inc., Cleveland, OH) to ensure that it remained constant and that the
reference electrode was operating properly. Both the pH and the electrical current drawn
throughout the experiment (Range doubler multitester, catalogue no. 22-215, Radio
Shack) were also recorded in an attempt to gain an understanding of the adsorption
mechanism. A photograph of the setup is shown in Figure 4-2.
It is both interesting and important to note that it is imperative that the cell be fed
in the manner indicated by Figure 4-1 (i.e., the feed should come into contact with the
working electrode prior to coming into contact with the auxiliary electrode). When the
direction is reversed, the process becomes ineffective. This infers that the reaction
37
occurring at the auxiliary electrode taints the reaction that occurs at the working
electrode.
38
be only two hours in length as the interesting phenomena typically occur during the first
30 to 60 minutes of the experiments. The test of cell potential involved testing potentials
starting from zero (pure adsorption) and slowly decreasing to –0.9 V. The evaluation of
flow rate involved the use of flow rates 2, 2.5, and 3 times that of the original flow rate.
Finally, several pHs were investigated in addition to the original of 3.5. These included
2.0, 5.0, and 7.0. Unless otherwise noted, all experimental parameters with the exception
of the one being tested were kept constant and equal to those used during the
performance experiments. Samples were taken every fifteen minutes until steady-state
effluent concentrations were obtained and less frequently thereafter.
The final experiments involved the effect of stripping and reloading on removal
efficiency and the ultimate capacity of the fibers. The cyclic loading experiment utilized
the same carbon sample which was repeatedly loaded for two hours and stripped for 30
minutes, then reloaded. The process was repeated six times and samples again were
taken every fifteen minutes until steady-state concentrations were reached and less
frequently thereafter.
The determination of ultimate capacity involved the use of a 1,000-ppm solution
as opposed to 100-ppm used for the abovementioned experiments. The solution was fed
at 0.7 mL/min. Once again, samples were taken every fifteen minutes until a steady-state
concentration was achieved and less frequently thereafter. Due to the length of this
experiment, it is possible that several hours passed between certain samples. However,
due to the nature of the experiment and the performance of the fibers, this did not effect
the interpretation of the results nor hinder the experiment in any way.
39
of the uranium. The voltage scan began at -0.5 V and increased to 0.0 V at a scan rate of
10.0 mV per second (versus Ag/AgCl reference electrode). Uranium (VI) was reduced to
U(V) during the scan causing a current peak to appear at approximately -0.17 V (versus
Ag/AgCl reference electrode). This agrees with the information presented in Chapter 3,
which states that the reduction of U(VI) to U(V) occurs at 0.052 V. After accounting for
the use of the reference electrode, which deducts 0.22V from what should be observed,
the peak should occur at -0.168 V.
The height of the peak is proportional to the concentration of U(VI) present in the
sample and can be used to determine the amount of U(VI) remaining in solution. A
typical U(VI) voltammogram is shown in Figure 4-3. Separate calibration curves were
constructed for use in both high and low concentration ranges. As stated above, the
calibration curves were constructed prior to testing and are shown below as Figures 4-4
and 4-5. It should be noted that the current peak can shift as the KCl-AgCl solution in
the reference electrode is diluted through diffusion. The solution was therefore changed
regularly (RDE0022, EG&G Instruments, Princeton Applied Research, Princeton, NJ).
0 -50 -100 -150 -200 -250 -300 -350 -400 -450 -500
0.0
Peak Height
-0.5
-1.0
Current (uA)
-1.5
-2.0
-2.5
-3.0
Potential (mV)
40
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0
-20
y = -21.434x
R 2 = 0.9905
-40
Peak Height (nA)
-60
-80
-100
-120
Uranium Concentration (ppm)
Figure 4-4. Calibration Curve for the Concentration Range 0.0 to 5.0 ppm
0 20 40 60 80 100 120
0
-500
y = -25.637x
R 2 = 0.999
-1,000
Peak Height (nA)
-1,500
-2,000
-2,500
-3,000
Uranium Concentration (ppm)
Figure 4-5. Calibration Curve for the Concentration Range 5.0 to 100.0 ppm
41
applying a positive potential of +1.00V (versus a Ag/AgCl reference electrode). The
uranium was recovered as a bright yellow solid. Although the form of the solid has been
identified as a hydrate, little information is known about its exact structure. It is worth
restating that it is this phenomenon that makes this technology unique when compared to
current treatment methods. Despite the industry’s current ability to concentrate uranium
wastes, thereby drastically reducing their volume, an emerging technology such as this
provides a means by which the uranium can be recovered in the solid form. This process
also provides an advantage to those already treating uranium with anion exchange (i.e.,
Fernald, Ohio, mentioned previously) as it can be used to treat the high concentration of
regenerants obtained after elution.
42
CHAPTER 5
RESULTS AND DISCUSSION
ASI produced eight types of nanofibers with varying physical properties for
testing at WVU. After the completion of initial testing, two additional fibers were also
fabricated and incorporated into the testing regime. In an attempt to identify those
properties that make the fibers effective in uranium removal, the fibers were initially
compared based upon a series of performance experiments. Additional testing to
determine the dependence of fiber performance on pH, flow rate, and cell potential then
followed these experiments. Finally, a fiber was chosen as a representative sample and
used to test for capacity and effects of regeneration on removal efficiency. FTIR analysis
to identify surface functional groups on the carbon fibers and X-ray diffraction
techniques to identify the solid uranium substance obtained upon stripping were also
completed and the results are included.
43
noted. PR-1-ox500 was also effective in uranium removal and the performance curve for
this fiber is shown in Figure 5-2.
100
90
100
90
80
80
70
70
60
Uranium Concentration (ppm)
50
60 40
30
50 20
10
40 0
0 15 30 45 60
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
44
It is also interesting to note the trend of both the current drawn due to the reaction
taking place inside the cell throughout the duration of the experiment and the pH of the
effluent. The current and pH trends are shown in Figures 5-3 and 5-4, respectively. As
can be seen from Figure 5-3, the current is initially high and then slowly drops to a
steady-state level. In opposition, the pH initially drops and then slowly rises to a steady-
state effluent pH of approximately 2.8. Although the trend of the two parameters is
opposite, both demonstrate a relationship similar to that of the concentration versus time
graphs. All parameters demonstrate an initial spike at the onset of the experiment then a
constant decrease until a steady-state value is achieved. Because of the similar behaviors
that mimic concentration trends, it is likely that changes in both pH and current are
related to the electrosorption mechanism.
3.5
3.0
2.5
2.0
Current (mA)
1.5
1.0
0.5
0.0
0 60 120 180 240 300 360 420 480
Time (minutes)
The third fiber tested, also an oxidized fiber, was PR-19-ox400 and it performed
identically to that of the baseline fiber, PR-1-ox400. This fiber was fabricated identically
to that of PR-1-ox400; however, natural gas was used as the carbon source as opposed to
methane resulting in a slightly larger diameter. Because both fibers were oxidized in air
at 400°C, they are believed to have the same extent of oxidation on the surface; hence,
their identical performance was not unexpected. The results are shown in Figure 5-5.
45
4.00
3.50
3.00
2.50
2.00
pH
1.50
1.00
0.50
0.00
0.0 60.0 120.0 180.0 240.0 300.0 360.0 420.0 480.0
Time (minutes)
100
90
80
Uranium Concentration (ppm)
70
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
46
As opposed to the intial set of three fibers which were all oxidized, the second set
of fibers tested were unoxidized or “as grown” fibers meaning they underwent no post-
treatment processing. This group included two fibers, PR-1-AG and PR-19-AG. PR-1-
AG was fabricated from methane and PR-19-AG from natural gas. Both fibers were very
effective in uranium removal. The results of the performance tests completed using the
PR-1-AG and the PR-19-AG fibers are shown in Figures 5-6 and 5-7, respectively.
It should be noted that in previous testing conducted at WVU, the unoxidized
fibers prepared and supplied by ASI proved to be unsuccessful [4]. However, when
similar fibers were supplied and tested again (PR-1-AG), the fibers proved to be effective
in the removal of uranium. The cause of this discrepancy is unknown at this time due to
the lack of knowledge of the electrosorption mechanism. It is possible that the fibers
changed over time; it is also possible that due to continuing improvements in the
production process, an unidentified characteristic was altered, causing the fibers to be
more effective (e.g., surface PAH content). Regardless, to ensure the importance of
utilizing the fibers provided by ASI, two samples of carbon black were obtained and
tested for their effectiveness at uranium removal. The results of these tests are presented
at the end of this section.
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
47
100
90
80
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
The next set of fibers tested included the last three fibers initially supplied by ASI
including PR-19-HT, PR-21-PS, and CO2-950. The PR-19-HT fiber is a fully graphitized
version of PR-19-AG and has a higher surface energy and higher graphitization index
than the other fibers. The fibers are heated to above 3,000°C in an inert atmosphere for
between 1 and 4 hours. It is also possible that the fibers have a higher edge plane density
but this has yet to be substantiated. This fiber, like the previous fibers tested, was
successful in removing uranium and the results of the testing can be seen in Figure 5-8.
PR-21-PS is a production fiber that has approximately the same diameter as the
PR-19-AG fibers but has a higher surface area and surface energy. The fiber is produced
using natural gas with an addition of CO2 and the post-treatment involves heating in
argon for between 4 and 6 hours. This fiber was also effective in uranium removal, and
the results can be seen in Figure 5-9.
The CO2-950 fiber has the highest surface area and surface energy of all the fibers
supplied. They performed very well also and the results of this test can be seen in Figure
5-10.
48
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
49
100
90
80
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
The last set of fibers included the two additional fibers not within the original
scope of testing. The ninth fiber, PR-23-HT, was similar to the PR-19-HT fiber, but has a
smaller diameter and possibly more edge planes. This fiber was just recently developed
and was therefore not included in the initial scope of the testing. This fiber, like PR-19-
HT is post-treated by heating to above 3,000°C in an inert environment. PR-23-HT
performed very well and the results are shown in Figure 5-11.
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
50
The final fiber tested, PR-24, was post-treated by soaking in peracetic acid in a
successful attempt (verified by ASI via XPS) to saturate the surface with oxygen
functional groups. This fiber also proved to be very effective in the removal of uranium
and the results of this test are shown in Figure 5-12.
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 60 120 180 240 300 360 420 480
Time (minutes)
As stated previously, two samples of carbon black were evaluated to ensure the
unique capability provided by using the fibers produced by ASI. The samples obtained
were conductive in an attempt to obtain different types of carbon that are nominally used
for the same types of applications. The carbon black samples were tested in the same
manner as that of the fibers. The first sample, PG-195-XB, was relatively unsuccessful in
removing uranium. The concentration initially dropped from 100 ppm to approximately
70 ppm but then began to increase toward the initial 100-ppm concentration of the feed
solution in a very short time. Although the sample did remove some uranium, when
compared with the fibers it was considered unsuccessful. The second sample, Superior
Graphite (BG-34), was more successful than the PG-195-XB. The concentration initially
dropped and then increased until reaching a steady state of sorts where it remained for a
short time before also increasing toward 100 ppm. The results from both experiments are
shown in Figures 5-13 and 5-14, respectively.
51
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 15 30 45 60 75 90 105 120
Time (minutes)
100
90
80
70
Uranium Concentration (ppm)
60
50
40
30
20
10
0
0 15 30 45 60 75 90 105 120
Time (minutes)
52
5.2 Effect of Cell Potential on Fiber Performance
It was both presumed and demonstrated through previous testing that several
factors contribute to the effectiveness of the nanofibers including pH, flow rate, and cell
potential. The first factor investigated was the potential applied to the electrolytic cell.
An experiment was conducted using the baseline fiber (PR-1-ox400) during which the
effect of several cell potentials was established. (It should be noted that this is the same
fiber used for the remainder of the experiments discussed in this chapter.) The potentials
ranged from zero (pure adsorption) to -0.9 V, the potential applied during the
performance experiments discussed in the previous section. The results of the variation
in cell potential are shown in Figure 5-15.
The dashed line, which for the most part is constant at 100 ppm, represents zero
potential and therefore resulted in very little removal. Following the observation of zero
potential, both -0.1 V and -0.2 V were tested and although a small amount of removal
was noted during the first fifteen minutes, the uranium concentration in the effluent
immediately returned to the inlet concentration of 100 ppm. Once the applied potential
was increased to -0.3 V, the fibers became more effective; however, the steady-state
effluent concentration still remained between 85 ppm to 95 ppm. The major increase in
removal efficiency occurred when the potential was raised from -0.3 V to -0.4 V. The
effluent concentration initially decreased to below 10 ppm then increased to a steady-
state concentration of approximately 15 ppm. The final three potentials tested included
-0.5 V, -0.7 V, and -0.9 V all resulted in a steady-state concentration of below 1 ppm. It
is obvious from this experiment that there is a change that occurs between -0.3 V and -0.4
V at which point the nanofibers become particularly more effective. This phenomenon is
most likely due to the required potential associated with the precipitation reaction/s.
It is interesting to look at the relationship between applied potential and steady
state effluent concentration in an attempt to predict the limiting potential or potential at
which the effluent remains at or below 1 ppm. These data are plotted and shown in
Figure 5-16. The plot shows that although the most significant change occurs between –
0.3 and –0.4 V, the limiting potential at which the effluent is maintained at or below 1
ppm is between –0.4 and –0.5 V.
53
100
80
70
Uranium Concentration (ppm)
-0.3 V
60
50
40
-0.4 V
30
20
-0.7 V
10
-0.5 V
-0.9 V
0
0 15 30 45 60 75 90 105 120
Time (minutes)
54
100
90
80
60
50
40
30
20
10
0
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
55
100
90
80
70
Uranium Concentration (ppm)
60
50
40
2.1 mL/min
Res. Tm.: 1.7-2.4 min.
30
0.7 mL/min
Res. Tm.:
20 5-7 min.
1.8 mL/min
Res. Tm.: 1.9-2.8 min.
10
Time (minutes)
56
Although it can be deduced from Figure 5-17 that the limiting flow rate occurs
between 1.4 and 1.8 mL/min, a separate plot was constructed in an attempt to predict the
exact value and is shown in Figure 5-18. The data were determined to have an
exponential relationship and predict a limiting flow rate of approximately 1.5 mL/min. It
should be noted that this fiber is a representative sample and it is likely that the limiting
flow rate would vary from fiber to fiber. It is also a strong function of cell size and
geometry.
20
4.9249x
y = 0.0005e
R2 = 0.9898
15
SS Effluent Concentration (ppm)
10
0
0.0 0.5 1.0 1.5 2.0 2.5
57
100
90
pH = 2.0
80
70
Uranium Concentration (ppm)
60
50
40
pH = 3.5
30
20
pH = 5.0
10
0
0 15 30 45 60 75 90 105 120
Time (minutes)
58
It should be noted that an attempt to test the performance at a pH of 7.0 resulted in
the precipitation of a uranium solid, making the experiment impossible to complete.
Because of the presence of the solid, very high removal efficiencies were noted in the
presence of a positive current, indicating that the solid was most likely being filtered by
the fiber.
59
100
90
80
70
Uranium Concentration (ppm)
Initial Pass
60
50
Subsequent Passs
40
30
20
10
0
0 15 30 45
Time (minutes)
60
balance, it was determined that a loading of 5.45 guranium/gfiber was achieved over the 40
hour period. A plot of the effluent concentration as a function of time is shown in Figure
5-21.
It should be noted that the mass balance was performed by collecting the effluent
from the cell for the entire duration of the experiment in 500-mL increments. The
concentration of the collected effluent was then determined. Because the influent
concentration was known, once the effluent concentration was determined, the amount of
uranium deposited on the carbon could be determined for that 500-mL increment.
A final point worth noting in relation to this experiment is the fact that as the
uranium deposits within the cell, it decreases the working volume inside the cell and
thereby increases the local flow rate, in effect, increasing the fluid velocity and
decreasing the residence time. The diminishing capacity demonstrated in the later stages
of this experiment could be due to an increase in flow rate that has already been shown to
reduce the fibers’ capacity (see Section 5.3). Future experiments involving scaled-up
models will be better apt at addressing this issue.
1,000
900
800
700
Uranium Concentration (ppm)
600
500
400
300
200
100
0
0 240 480 720 960 1,200 1,440 1,680 1,920 2,160 2,400
Time (minutes)
61
5.7 Possible Effect of Residual Iron
Because the fibers are produced using an iron catalyst, it was thought that possible
residual iron present in the fibers might have been contributing to the effectiveness of the
uranium removal process by acting as a magnet. Therefore, it was of interest to
determine the amount of iron present in the ten fiber samples. Ash tests were performed
by heating 0.5-g samples of fiber to 750°C in air and maintaining this temperature for at
least four hours. This process burns off all carbon and leaves only the oxides of the
metals in the original fiber. Several of the fibers were determined to contain significant
amounts of residual ash. Although it is possible that other metal oxides are present, it is
likely that the majority of the ash is composed of iron oxide as it is known that the fibers
are fabricated using an iron catalyst and the ash residue appeared to be rust colored upon
visual inspection. No other metals should be present since none were added initially.
The carbon feed, whether CH4 or methane, were pure gases. The results of the ash tests
are shown in Figure 5-22.
5.00%
4.50%
4.00%
3.50%
3.00%
Percent Ash
2.50%
2.00%
1.50%
1.00%
0.50%
0.00%
G
4
T
T
0
00
00
PS
0
-2
95
-A
40
A
H
x4
x5
1-
PR
9-
9-
3-
2-
ox
-1
-2
-o
-o
-1
-1
-2
CO
PR
9-
PR
-1
-1
PR
PR
PR
-1
PR
PR
PR
Sample Identification
Although all samples contained significant amounts of ash, the fibers produced
from natural gas had a higher ash content on average than the other fibers. The reason for
62
this is unknown; perhaps the carbon layer is thinner and results in a higher iron
concentration by weight or perhaps a higher concentration of catalyst was employed.
Despite the high ash content in the majority of the samples, the two heat-treated
graphitized fiber samples, PR-19-HT and PR-23-HT, contained no residuals. This is due
to the volatilization of the iron at the high temperatures used to treat the fibers (i.e., over
3,000°C). Because all the fibers were effective in uranium removal and two fibers
contained no residuals, it can be concluded that the presence of iron does not contribute
to the effectiveness of the fibers.
63
5.9 Identification of Uranium Solid By X-Ray Diffraction
The solid obtained upon stripping the fibers was collected and analyzed via x-ray
diffraction techniques. The solid was air dried at room temperature prior to analysis.
The results showing the various diffraction peaks can be seen in Figure 5-23. The results
were somewhat inconclusive and indicated that the solid contained a barium atom, a
substance that is not present anywhere in the system. It is likely that the actual uranium-
bearing solid was not present within the database with which the spectrum was compared.
Regardless, the analysis indicates that the solid stripped from the carbon fibers is some
type of complex uranium hydroxide hydrate (i.e., X(UO2)6O4(OH)6⋅8(H2O)).
64
Figure 5 – 23. Results of X-Ray Diffraction Analysis
65
Figure 5 – 24. SEM of PR-1-AG Showing Smooth Surface
66
CHAPTER 6
CONCLUSIONS AND RECOMMENDTIONS
The purpose of this study was to determine the effect of varying the surface
properties of catalytically grown carbon nanofibers on uranium removal. The fibers were
subjected to performance experiments that were used to compare them. Following these
initial experiments, the experimental parameters including cell potential, flow rate, and
pH of influent were varied to determine their effect on the uranium removal phenomenon.
A final set of experiments was conducted to determine the effect of repeatedly loading
and stripping the fibers on removal efficiency and the capacity of the fibers for
electrosorption. Attempts were made to identify surface functional groups and to
determine the identity of the uranium solid obtained upon stripping the fibers.
67
reached -0.4 V that the uranium removal significantly increased, maintaining a steady-
state effluent concentration of approximately 15 ppm. It is likely that there is a limiting
applied potential at which the precipitation mechanism is initiated. Further increases in
potential including -0.5 V, -0.7 V, and -0.9 V resulted in maintaining an effluent
concentration of below 1 ppm. Therefore, the limiting applied potential is reached
somewhere between -0.4 and -0.5 V.
The second parameter, the flow rate, affected the uranium removal rate
significantly. It was determined that the flow rate could be set to twice that of the initial
flow rate, resulting in a feed of 1.4 mL/min and still maintain an effluent concentration of
below 1 ppm uranium. However, upon tripling the flow rate, the effluent concentration
increased to between 15 and 20 ppm. Operating the cell at a flow rate two and a half
times the original flow rate resulted in a steady-state effluent concentration of 5 ppm. It
was therefore determined that the limiting flow rate is located somewhere between 1.4
and 1.8 mL/min, 1.5 mL/min approximately.
The final test parameter was pH. A range of pHs was considered including 2.0,
3.5 (used in the performance experiments), 5.0, and 7.0. Decreasing the pH to 2.0
resulted in decreased removal efficiency and is therefore not desirable. Increasing the pH
from 3.5 to 5.0 achieved the same high removal efficiency. The final increase in pH from
5.0 to 7.0 resulted in the formation of a uranium precipitate that was filtered by the fiber
even when a positive potential was applied and therefore the experiment could not be
completed. It appears from this information that the technique is not effective at pHs at
or below 2.0. Because no pHs were tested between 2.0 and 3.5, the limiting pH was not
clearly determined. It appears that increases in pH, increase the removal efficiency of the
process until precipitation is achieved at which point the uranium can simply be filtered
and removed.
68
uranium precipitates that are not expelled from the fiber surface during stripping. These
precipitates then provide sites for more immediate precipitation of additional uranium,
resulting in quicker removal during subsequent loadings.
The fiber capacity was also determined through an extensive loading experiment
involving a run that lasted for several days. The feed concentration was increased to
1,000 ppm to speed up the saturation process. The flow rate and pH were maintained at
0.7 mL/min and 3.5, respectively. The cell broke through after approximately 15 hours
when the fibers apparently became saturated with uranium. Breakthrough lasted for
several hours and the experiment was halted after 40 hours. The capacity was determined
to be 5.45 guranium/gfiber for the entire 40-hour period.
It should be noted that it is not clear whether the loading of the fiber is a function
of nucleation sites. Deposition of the uranium throughout the process of the experiment
causes a decrease in the working free volume of the cell. Therefore, as more uranium is
deposited, it causes an increase in the local flow rate and could therefore result in an
apparent decrease in removal efficiency. In addition to being affected by these
conditions, the capacity is also expected to be fiber specific although this was not studied
here.
69
Two fiber samples were analyzed using SEM techniques in an attempt to identify
edge plane density and other surface characteristics; however, the edge planes were not
apparent at the magnification used (25,000X). There was a minor difference between the
surface of the fibers in that the “as-grown” fibers appeared to be significantly smoother
than the heat-treated fibers. This is believed to be due to the pyrolytic materials that are
stripped from the surface of the heat-treated fibers during post-treatment.
70
REFERENCES
1. Hatfield, T.L., T.L. Kleven and D.T. Pierce, Journal of Applied Electrochemistry,
26, 1996, p. 567-574.
6. Lee, S.Y. and E.A. Bondietti, “Removing Uranium from Drinking Water by
Metal Hydroxides and Anion-exchange Resins”, J. AWWA, 75(10):536, 1983.
7. Hanson, S.W., D.B. Wilson, N.N. Gunaji and S.W. Hathaway, “Removal of
Uranium from Drinking Water by Ion-exchange and Chemical Clarification”,
EPA/600/S2-87/076, Dec. 1986.
8. Huxstep, M.R. and T.J. Sorg, “Reverse Osmosis Treatment to Remove Inorganic
Contaminants from Drinking Water”, US EPA Report/600/52-87/109, March
1987.
9. Clifford, D.A. and Z. Zhang, Removing Uranium and Radium from Groundwater
by Ion Exchange Resins, Ed. A.K. Sengupta, Ion Exchange Technology,
Technomic, Lancaster, PA, 1995.
10. Hathaway, S.W., Uranium Removal Processes, In Pros. AWWA Ann. Conference,
Las Vegas, NV, 1983.
71
13. Lee, S.Y., S.K. Hall and E.A. Bondetti, “Methods of Removing Uranium from
Drinking Water: II, Present Municipal Water Treatment and Potential Removal
Methods”, EPA 570/9-82-003, Washington, DC, 1982.
14. Hathaway, S.W., “Current Research on Uranium Removal from Drinking Water”,
US EPA Small Water Systems Technology Seminar, April 1982.
16. Farmer J.C., S.M. Bahowick, J.E. Harar, D.V. Fix, R.E. Martinelli, A.K. Au, and
K.L. Carroll, Energy and Fuels, 11, 1997, p. 337-347.
18. Grebenyuk, V.D., N.A. Linkov and V.M. Linkov, Water SA, 24, No. 2, 1998, p.
123-127.
19. Hazourlis S., G. Bonnecaze and M. Astruc, Environmental Technology, 17, No.
12, 1996, p. 1275-1283.
20. Hazourlis S., G. Bonnecaze and M. Astruc, Environmental Technology, 17, No.
12, 1996, p. 1285-1295.
21. Hazourlis S., G. Bonnecaze and M. Astruc, Environmental Technology, 17, No.
12, 1996, p. 1297-1312.
23. Guler, H., N. Sahiner, G.A. Aycik and O. Guven, Journal of Applied Polymer
Science, 66, No. 3, 1997, p. 2475-2480.
24. Benes, P., K. Kratzer, S. Vlckova and E. Sebestova, Chemicke Listy, 91, No. 9,
1997, p. 774-775.
25. Sato, T., T. Murakamit, N. Yanase, H. Isobe, T.E. Payne and P.L. Airey,
Environmental Science and Technology, 31, No. 10, 1997, p. 2854-2858.
26. Kaplan, D.I., T.L. Gervais and K.M. Krupka, Radiochimica Acta, 80, No. 4, 1998,
p. 201-211.
72
28. Zhang, X.Z., S.G. Luo, Q. Yang, H.L. Zhang and J.Y. Li, Journal of Applied
Phycology, 9, No. 1, 1997, p. 65-71.
30. Roig, M., G. Manzano and M. Diaz, Water Research, 31, No. 8, 1997, p. 2073-
2083.
31. Riordan, C., M. Bustard, R. Putt, A.P. Michale, Biotechnology Letters, 19, No. 4,
1997, p. 385-387.
32. Hu, M.Z.C. and M. Reeves, Biotechnology Progress, 13, No. 1, 1997, p. 60-70.
33. Lovely, D.R., E.J.P. Philips, Y.A. Gorby and E.R. Landa, Nature, 350, April
1991, p. 413.
34. Lovely, D.R. and E.J.P Philips, Applied and Environmental Microbiology, Mar.
1992, p. 850-856.
35. Truex, M.J., B. Peyton, N.B. Valentine and Y.A. Gorby, Biotechnology and
Bioengineering, 55, No. 3, August 5, 1997, p. 490-496.
36. Katz, J. and E. Rabinowitch. The Chemistry of Uranium, Nila, Wiley, 1951.
37. Katz J. and G.T. Seaborg, The Chemistry of Actinide Elements, 1957.
39. Seaborg, G.T. and J. Katz, The Actinide Elements, New York, McGraw-Hill,
1954.
41. Kraus, K.A. and F. Nelson, J. Am. Chem. Soc., 77, 1955, p. 3721.
42. Tyurikov, G.S., K.I. Rosental and V.I. Veselovskii, ZhFKh, 32, 1490, 1958.
43. Rao, G.G., V.P. Rao and N. Venkatamma, -Z. anal. Chem., 150, 178, 1956.
44. Rao, V. and G. Rao, -Z. anal. Chem., 160, 190, 1958.
45. Rao, G.G. and V.P. Rao, Anal. Chim. Acta., 15, 97, 1956.
73
46. Grenberg, A.A., G.I. Petrazhak, L.E. Nikol’Skaya, B.V. Pittsyn and F.M. Filinov,
Trudy Radievogo Instituta AN SSSR, 8, 166, 1958.
47. Ushatskii, V.N. and Y.M. Tolmachev, Trudy Radievogo Instituta AN SSSR, 7, 8,
1956.
48. Ratner, A.P. and P.I. Chaikin, Sbornik rabot po radiokimii, Moscow, Izd. LGU,
1955.
49. Kern, D.M.H. and E.F. Orleman, J. Am. Chem. Soc., 71, 2102, 1949.
50. Nelson, F. and K.A. Kraus, J. Am. Chem. Soc., 73, 2157, p. 1951.
55. Komar, N.P. and Z.A. Tret’yak, ZhAKh, 10, 236, 1955.
60. Brennsteiner, A., J. Zondlo, A.H. Stiller, P.G. Stansberry, D. Tian and Y. Xu,
Energy and Fuels, 11, 1997, p. 348-353.
61. Pavia, D.L., G.M. Lampman and G.S. Kriz, Introduction to Spectroscopy, 2nd Ed.,
Saunders College Printing, New York, 1996.
74
APPENDIX
75
PERFORMANCE EXPERIMENTS
PR-1-ox400
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
0 0.0 NA 100.00 0 0.0 NA 100.00
13 6.5 -1,007.40 39.29 15 6.5 -892.80 34.82
25 19.0 -26.24 1.02 30 22.5 -57.81 2.25
37 31.0 -1.51 0.06 45 37.5 -13.48 0.53
48 42.5 -1.48 0.06 60 52.5 -2.04 0.08
59 53.5 -8.12 0.32 75 67.5 -5.16 0.20
73 66.0 -4.45 0.17 90 82.5 -1.26 0.05
87 80.0 -2.65 0.10 105 97.5 -2.16 0.08
101 94.0 -2.19 0.09 120 112.5 ND <0.05
117 109.0 -6.17 0.24 420 NA ND <0.05
131 124.0 -4.29 0.17 480 NA ND <0.05
160 145.5 -0.96 0.04
175 167.5 ND <0.05
480 480.0 ND <0.05
PR-1-ox500
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
0 0.0 NA 100.00 0 0.0 NA 100.00
15 7.5 -690.30 26.93 15 7.5 -475.80 18.56
30 22.5 -21.60 0.84 30 22.5 -22.13 0.86
45 37.5 -7.93 0.31 45 37.5 -5.75 0.22
60 52.5 ND <0.05 60 52.5 -15.96 0.62
75 67.5 -2.36 0.09 75 67.5 -2.18 0.09
90 82.5 -3.01 0.12 90 82.5 -4.77 0.19
105 97.5 ND <0.05 105 97.5 -2.15 0.08
135 120.0 -6.04 0.24 120 112.5 ND <0.05
195 NA -2.58 0.10 480 NA ND <0.05
300 NA ND <0.05
480 NA ND <0.05
76
Current and pH
0 NA 3.50
15 3.05 2.54
30 2.05 2.60
45 1.85 2.71
60 1.80 2.76
75 1.80 2.80
90 1.75 2.83
105 1.75 2.85
135 1.70 2.89
195 1.60 2.87
300 1.55 2.87
480 1.50 2.87
PR-19-ox400
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
77
PR-1-AG
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
0 0.0 NA 100.00 0 0.0 NA 100.00
15 7.5 -933.80 36.42 15 7.5 -924.10 36.05
30 22.5 -52.85 2.06 30 22.5 -26.75 1.04
45 37.5 -13.11 0.51 45 37.5 -11.22 0.44
60 52.5 -7.05 0.28 60 52.5 ND <0.05
75 67.5 -7.90 0.31 75 67.5 ND <0.05
90 82.5 -3.41 0.13 90 82.5 -2.03 0.08
105 97.5 -18.94 0.74 105 97.5 -4.57 0.18
120 112.5 -2.01 0.08 120 112.5 -3.14 0.12
390 NA ND <0.05 420 NA -1.21 0.05
480 NA -2.89 0.11 480 NA -27.44 1.07
PR-19-AG
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
78
PR-19-HT
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
PR-21-PS
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
79
CO2-950
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
PR-23-HT
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
80
PR-24
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
0 0.0 NA 100.00 0 0.0 NA 100.00
15 7.5 -860.50 33.56 15 7.5 -869.00 33.90
30 22.5 -42.66 1.66 30 22.5 -27.95 1.09
45 37.5 -14.66 0.57 45 37.5 -7.05 0.27
60 52.5 -4.14 0.16 60 52.5 ND <0.05
75 67.5 ND <0.05 75 67.5 ND <0.05
90 82.5 ND <0.05 90 82.5 ND <0.05
105 97.5 ND <0.05 105 97.5 -5.34 0.21
120 112.5 ND <0.05 120 112.5 ND <0.05
480 NA ND <0.05 150 NA -4.75 0.19
180 NA ND <0.05
240 NA -8.94 0.35
360 NA ND <0.05
480 NA -2.51 0.10
Time Ave. Time Peak Ht. Conc. Time Ave. Time Peak Ht. Conc.
(min.) (min.) (::A) (ppm) (min.) (min.) (::A) (ppm)
0 0.0 NA 100.00 0 0.0 NA 100.00
15 7.5 -2,329.00 93.10 15 7.5 -813.20 32.51
30 22.5 -1,959.00 78.31 30 22.5 -111.80 4.47
45 37.5 -1,756.00 70.20 45 37.5 -1,073.00 42.89
60 52.5 -2,185.00 87.35 60 52.5 -1,455.20 58.17
90 82.5 -2,424.00 96.90 75 67.5 -1,552.10 62.05
90 82.5 -1,544.20 61.73
105 97.5 -1,582.50 63.26
120 112.5 -1,666.00 66.60
81
VARIATION OF EXPERIMENTAL PARAMETERS
Applied Potential
82
Flow Rate
pH
83
CYCLIC LOADING/UNLOADING
0 NA 100.00 0 NA 100.00
15 -745.63 29.08 15 -799.23 31.17
30 -7.32 0.29 30 -9.56 0.37
45 ND <0.05 45 -5.46 0.21
60 ND <0.05 60 ND <0.05
75 -1.27 0.05 75 ND <0.05
90 ND <0.05 90 -3.21 0.13
105 -3.33 0.13 105 ND <0.05
120 ND <0.05 120 -1.89 0.07
0 NA 100.00 0 NA 100.00
15 -676.33 26.38 15 -714.56 27.87
30 -8.73 0.34 30 -12.34 0.48
45 -4.87 0.19 45 -6.74 0.26
60 ND <0.05 60 ND <0.05
75 -3.25 0.13 75 ND <0.05
90 ND <0.05 90 -2.67 0.10
105 -5.61 0.22 105 ND <0.05
120 ND <0.05 120 ND <0.05
84
CAPACITY
Time (min.) Ave. Time (min) Peak Ht. (::A) Conc. (ppm)
0 0.0 NA 100.00
15 7.5 -16,680.00 650.62
30 22.5 -9,827.00 383.31
45 37.5 -8,541.00 333.15
60 52.5 -7,811.00 304.68
90 75.0 -7,246.00 282.64
120 105.0 -6,985.00 272.46
240 NA -5,693.00 222.06
405 NA -5,455.00 212.78
585 NA -5,720.00 223.12
780 NA -5,749.00 224.25
1,560 NA -8,259.00 322.15
2,115 NA -18,770.00 732.14
2,370 NA -21,316.00 831.45
RESIDUAL ASH
Percent Ash
Percent Ash
Wt. of Both
Wt. of Both
after Ash
after Ash
Crucible
Crucible
Wt. of
Wt. of
Wt. of
Wt. of
Duplicate
Fiber
Fiber
Initial Test
Test
PR-1-ox400 8.45 0.63 8.46 1.73% PR-1-ox400 9.28 0.54 9.28 1.79%
PR-1-ox500 9.44 0.52 9.45 2.67% PR-1-ox500 9.38 0.59 9.39 2.95%
PR-19-ox400 8.52 0.61 8.53 2.38% PR-19-ox400 9.71 0.68 9.72 2.75%
PR-1-AG 9.28 0.72 9.29 1.36% PR-1-AG 8.99 0.63 9.00 1.37%
PR-19-AG 8.76 0.64 8.78 3.70% PR-19-AG 8.60 0.61 8.63 3.67%
PR-19-HT 9.45 0.71 9.45 0.00% PR-19-HT 8.94 0.73 8.94 0.00%
PR-21-PS 8.76 0.58 8.78 3.85% PR-21-PS 9.36 0.54 9.37 2.57%
CO2-950 18.74 0.52 18.75 2.28% CO2-950 20.56 0.58 20.57 2.43%
PR-23-HT 9.43 0.63 9.43 0.00% PR-23-HT 9.21 0.69 9.21 0.00%
PR-24 8.72 0.54 8.73 1.81% PR-24 9.09 0.52 9.10 1.87%
85