Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 87
4
Waste Processing and
Waste Form Production
T
he focus of this chapter is on the fourth charge of the statement of
task for this study (see Box 2.1 in Chapter 2), which calls for the
identification and description of “potential modifications of waste
form production methods that may lead to more efficient1 production of
waste forms to meet their performance requirements.” Waste form produc-
tion involves many complex operations, a comprehensive review of which is
well beyond the scope of this study. Instead, this chapter presents some key
unit operations/technologies that have been or could potentially be used to
produce nuclear waste forms.
The following waste processing technologies are described in this chap-
ter, and their key attributes are summarized in Table 4.1:
• Joule-heated melters
• Cold crucible induction melters
• In-container vitrification
• Self-sustaining vitrification
• Cold pressing and sintering
• Hot uniaxial pressing
• Hot isostatic pressing
1 A given waste processing technology is “more efficient” compared to a baseline technology
when, for example, it enables higher material throughputs or higher waste loadings; accom-
modates higher levels of feed stream variability or liquid feed streams; accommodates higher
levels of incompatible elements; results in reduced secondary wastes; is more operationally
robust; or is less expensive.
87
OCR for page 88
88 WASTE FORMS TECHNOLOGY AND PERFORMANCE
TABLE 4.1 Waste Form Processing Technologies
Processing Treatment and
Processing Mode B=Batch Waste Stream
Scalea
Technology C=Continuous Waste Forms Produced
Joule-Heated, Melter C Large Borosilicate glass other
(JHM) glass types (LaBs, FeP, AIP,
chalcognide, and others)
Advanced Joule C Large Borosilicate glass, glass-
Heated Melter ceramic materials, other
(AJHM) glass types (LaBs, FeP, AIP,
chalcognide, and others)
Cold Crucible C Large Borosilicate glass, glass-
Induction Melter ceramic materials, other
(CCIM) glass types (LaBs, FeP, AIP,
chalcognide, and others),
crystalline ceramics/simple
oxides, metal matrix
In-Container B Depends on Borosilicate glass, glass-
Vitrification (ICV) container ceramic materials, other
(also known as bulk size (could be glasses (LaBs, FeP, AIP,
vitrification) medium to large) chalcognide, and others)
Cold Press and B Small Glass-ceramic materials,
Sinter crystalline ceramic/simple
oxides, metal matrix,
zeolites, hydroceramics
Self-Sustaining B Small Glass-ceramic materials
Vitrification (SSV)
Hot Uniaxial B Small Glass-ceramic materials,
Pressing (HUP) crystalline ceramics/simple
oxides, metal matrix,
zeolites, hydroceramics
Hot Isostatic B Small Borosilicate glass (lab
Pressing (HIP) demonstration only), glass-
ceramic materials, crystalline
ceramics/simple oxides,
metal matrix, zeolites,
hydroceramics
OCR for page 89
89
WASTE PROCESSING AND WASTE FORM PRODUCTION
Advantages Disadvantages
Proven technology; typically operates with Materials of construction can be problematic
a “cold cap” to minimize volatility of for some wastes; solubility control of certain
species of concern species (e.g., chromium spinels) critical;
excessive spinels may seize-up melter;
electrode erosion may be a problem
Increases throughput and melt rate Operates with minimal or no “cold cap” with
compared to JHM associated increases in volatility of species of
concern
Allows the processing of corrosive glasses; Higher temperature operation can increase
no refractories; no metallic or oxide volatilization of species of concern but “cold
electrodes; water cooled; self-cleaning; high cap” coverage minimizes these impacts
purity; can be stirred if needed; increases
capacity compared to JHM and AJHM;
can operate at higher temperatures than
JHM and AJHM; operates with a “cold
cap” to minimize volatility
Relatively cheap and simple for low- Inhomogeneous waste forms produced;
activity wastes or contaminated soils; not no temperature control so radionuclide
applicable to high-level radioactive waste vaporization is high; little to no convection in
melt causes processing problems
Higher waste loadings; minimum disposal Usually small scale; may require precalcining
volumes or pretreating waste to an oxide form to
avoid shrinkage of form
Low capital requirements, can be used to May require some pre-processing, for
process small amounts of wastes at remote example, grinding of the waste and
locations pre-mixing
Higher waste loadings; minimum disposal Usually small scale; may require precalcining
volumes; mature flexible technology; or pretreating waste to an oxide form for
mature industrial process shrinkage control
Zero off-gas emissions; higher waste Processes small quantities; can overpressurize
loadings; minimum disposal volumes; if large amounts of volatiles (e.g., nitrates/
mature flexible technology; no major hydrates) are present; may require
secondary wastes; mature industrial precalcining or pretreating waste to an oxide
process form (shrinkage handled by bellows like
canisters)
continued
OCR for page 90
90 WASTE FORMS TECHNOLOGY AND PERFORMANCE
TABLE 4.1 Continued
Processing Treatment and
Processing Mode B=Batch Waste Stream
Scalea
Technology C=Continuous Waste Forms Produced
Fluidized Bed Steam C Large Crystalline ceramics/simple
Reforming (FBSR) oxides, zeolites as formed,
hydroceramics, geopolymers
(as encapsulated)
a All processes characterized as “small” are batch processes. The size of the batch varies by
technology. For example, HIP and SSV batches can range from less than a liter to several tens
of liters in volume. The processing capacity depends on a multiplicity of factors, including the
time scales of various physical process (e.g., time to charge, melt, and cool). Processes charac-
terized as “large” are continuous processes that have the potential to handle several kilograms
to several hundreds of kilograms of waste per hour. In HLW applications, for example, JHMs
can typically process 3 tonnes per day per melter or 125 kilograms per hour; AJHMs can
process up to about 4 tonnes per day. For LAW the production rate could be up to 25 tonnes
per day. Similarly, a CCIM can double the capacity of waste processing. A fluidized bed, on the
other hand, does not have the geometrical physical limitations imposed by JHMs or CCIMs
and can process several hundred kilograms to several tonnes of waste per hour.
• Fluidized bed steam reforming
• Other thermal technologies
• Mix and set technologies
Cold crucible induction melters, hot isostatic pressing, and fluidized bed
steam reforming technologies were discussed in the committee’s interim
report (NRC, 2010).
This list of technologies is not exhaustive. Rather, the committee
selected technologies that it judged were potentially most relevant to the
U.S. Department of Energy, Office of Environmental Management (DOE-
EM) cleanup program.
4.1 JOULE-HEATED MELTERS
The term Joule heating refers to heating obtained by passing an electri-
cal current through a resistively conducting material. The electrical resis-
tance of the material causes the electrical energy to be converted to heat,
with power dissipation following Ohm’s Law.
A Joule-Heated Melter (JHM) is a refractory-lined container with
nickel-chromium alloy electrodes, usually Inconel™. It is loaded with a
OCR for page 91
91
WASTE PROCESSING AND WASTE FORM PRODUCTION
Advantages Disadvantages
Industrially proven technology; acidic or Product is granular and requires a high-
basic tank wastes can be processed without integrity container or encapsulation in a
neutralization; destroy organics and binder to make a glass ceramic material, a
nitrates; convert aqueous components in geopolymer, or a hydroceramic; radionuclide
a waste to stable, water-insoluble mineral partitioning among the phases needs to be
products in single step; immobilize sulfur, further studied
chlorine, and fluorine in a stable mineral
form with no secondary waste
calcine or slurry containing waste and glass frit and is melted by applying
voltage across the electrodes. The nominal melt temperature is 1,150°C,
which is only 200°C lower than the melting point of the Inconel™ elec-
trodes. A waste form is produced by pouring the molten material into a
container and allowing it to cool. The melting and pouring processes can
be operated in continuous or semi-continuous modes.
JHM technology is being tested or used successfully to immobilize
high-level radioactive waste (HLW) in several countries, including Belgium,
France, Japan, Russia, UK, and the United States (see Table 3.3 in Chapter
3). JHMs are part of the current DOE-EM baseline for immobilizing HLW
at West Valley, Savannah River Site (SRS), and Hanford Site. Details on
melter designs in use in different countries can be found in recent com-
pendiums by Caurant et al. (2009), Jain (1998), Jantzen (in press), Ojovan
(2011), and Ojovan and Lee (2005, 2007).
The size of a JHM is usually limited only by capacity of the cranes that
are used to install and, if necessary, replace it; structural support is provided
by a stainless steel shell of the melter, which contains the refractory. The
Defense Waste Processing Facility (DWPF) at the Savannah River Site is the
largest production JHM ever built. Larger JHMs are under construction for
use at the Hanford Site; these large melters will be mounted on rails for ease
of replacement.
OCR for page 92
92 WASTE FORMS TECHNOLOGY AND PERFORMANCE
Several improvements have been made to conventional JHM designs
to increase throughputs and waste loadings. These improved designs are
referred to as Advanced Joule Heated Melters, or AJHMs. These designs have
been proven at the pilot scale with simulated wastes, as well as operated at
the plant scale with mixed radioactive/hazardous waste at SRS where main-
tenance and repairs could be handled manually. They have not been operated
with HLW where maintenance and repairs have to be performed remotely.
JHMs have production rates that are approximately proportional to the
surface area of the melt; convection caused by the Joule heating is enhanced
as the size of the melter is increased. So larger melters, with larger surface
areas, have proportionately higher melt rates. The melt temperature is limited
by the materials of construction of the electrodes, generally Inconel™ 690.
Melt rates can also be increased without increasing melter size through
the following means:
• Adding lid heaters to increase the temperature of the melter plenum
and enhance melting of the cold cap (i.e., the unmelted feed mate-
rial on top of the melted mass).
• Adjusting the proportions of frit and cold chemical additions.
• Increasing the use of reducing agents (e.g., formic acid/sugar) to
control oxygen foaming.
• Adding surface-active materials such as sulfates and halides.
• Increasing melter convection by using lower-viscosity glass, power
skewing the bottom electrodes, or mechanical agitation (stirrer/
bubblers/airlift pumps).
• Dry feeding instead of slurry feeding.
• Increasing the operating temperature.
• Bubbling to enhance feed to glass conversion.
Melt rates can also be increased by increasing melt temperatures. JHMs
can be operated at ~1,200°C before different materials of construction are
required. However, increased melt pool volatility, refractory corrosion, and
electrode corrosion are to be expected at higher operating temperatures.
In order to limit volatilization of radionuclides such as techentium-99 and
ruthenium-104 in borosilicate melts, the melter can be operated with a
cold cap and under reducing conditions to keep the technetium as sodium
pertechnatate (NaTcO4) and ruthenium as an oxide (RuO2). The cold cap
also helps minimize volatilization of cesium (as CsBO2) and other alkali
salts such as sodium chloride (NaCl), sodium fluoride (NaF), and sodium
iodide (NaI) (Jantzen, 1991). The increasing production capability is offset
by increasing complexity of the melter system.
The DWPF is considered to be a JHM with a few AJHM design fea-
tures. It is round with a slightly sloped floor to avoid cold corners and
OCR for page 93
93
WASTE PROCESSING AND WASTE FORM PRODUCTION
improve melt pool convection. Wet feeding is used to avoid dusting and the
need for a dust collection system. This melter also has lid heaters, adjusts
the proportions of frit to waste, uses reducing agents to control oxygen
foaming and convert nitrates to N2, uses lower melting viscosity glasses,
power skews the bottom electrodes, and uses an airlift pump to improve
melt rate.
The planned Hanford HLW melter is a square AJHM melter.2 It will
allow the proportions of cold chemical additions to be adjusted, use of
reducing agents to convert nitrates to N2, and use of lower viscosity glasses.
Additionally, it will allow use of power skewing of the bottom electrodes
and use of bubblers3 to improve melt rate and melt pool convection. Higher
temperatures will be used. Partial dry feeding will also be used to improve
melt rate.
The primary advantages of JHMs for waste immobilization are their
high production rates and ability to produce waste form material of consis-
tent quality. They can be designed with sloped bottoms and bottom drains
with or without mixing to facilitate periodic draining of noble metals that
may precipitate (as is done with the JHM in use at Tokai, Japan). The
overall problems with noble metals in waste streams have been addressed
by Bibler (2005) and some specifics of noble metals in a JHM have been
treated by Jantzen and Lambert (1999).
JHMs also have several disadvantages: They are large compared to
some other types of melters (see Section 4.2), and the electrodes and refrac-
tories require maintenance and periodic replacement. They are also intoler-
ant of crystal growth in the melt, which causes slag formation. For example,
the induction heating melter at Sellafield (Riley et al., 2009), which is
not a JHM design, has shown the ability to increase waste loading from
25 weight percent to 38 weight percent by allowing spinel formation in the
melt.4 In comparison, the melter planned for Hanford will allow for only
1-2 percent crystallization of spinels; it is anticipated that these crystals will
remain buoyant because of melt pool agitation by the bubblers. This strat-
egy will likely work unless the crystals grow larger than can be sustained
by the bubblers—as might happen, for example, during long maintenance
outages.
2 Square melters have cold corners where glass can crystallize unless the melt pool is agi-
tated. Round melters such as the DWPF and Japanese designs require no melt pool agitation
except as needed to improve melt rate.
3 The melter will be outfitted with eight twin-orifice melt bubblers that bubble air and
thus improve melt rate and convection. The use of multiple bubblers with the capability for
frequent replacements is considered to be an AJHM design.
4 Transition metal-containing spinels form from the iron generated during fuel reprocessing
and from corrosion products. Usually the least soluble component in the wastes is chromium,
which readily precipitates in a melt as (Fe,Ni)(Fe,Cr)2O4 (Hrma et al., 2002).
OCR for page 94
94 WASTE FORMS TECHNOLOGY AND PERFORMANCE
4.2 COLD CRUCIBLE INDUCTION MELTING
The Cold Crucible Induction Melter (CCIM) consists of water-cooled
tubes that are arranged to form a crucible, the contents of which are heated
by induction. An inductor surrounding the crucible is energized by a high-
frequency alternating current that induces eddy currents (and resultant Joule
heating) of materials contained in the crucible. The melting process is usu-
ally initiated by inserting a resistive heating element into the crucible. This
element couples with the fields to form an initial melt; the initial melt then
couples with the electromagnetic field. At that point, the resistive element
can be removed so that no foreign materials are in contact with the melt.
A solid “skull” of quenched waste material, typically a few millimeters in
thickness, forms along the inside crucible wall, protecting it from degrada-
tion and corrosion. The shell isolates the melt from the crucible so that the
latter can be maintained at ambient conditions. This prevents molten mate-
rials from bonding to the crucible, allowing residuals to be removed at the
end of the melting campaign. CCIMs are potential replacements for JHMs
and AJHMs. As noted previously, JHMs with some AJHM features are now
part of the current DOE baseline for production of high-activity and low-
activity waste glass (see Chapter 2).
CCIMs have several advantages over both current-generation and
advanced JHMs. They allow for higher throughputs and waste loadings.
They are operationally simpler and allow for faster recoveries from sys-
tem upsets.5 The absence of internal electrodes and refractories allows for
increased melter longevity and permits higher-temperature operation com-
pared to current-generation JHMs. As a consequence, CCIMs can be used
to process a wider range of waste compositions, including corrosive wastes
that are incompatible with current-generation JHMs. Additionally, they
can more easily accommodate differing glass compositions, including iron
phosphate glasses, that are incompatible with many types of electrodes used
in JHMs. CCIMs can be cycled frequently with varying feed compositions
without thermal damage or loss of compositional control. And they are
capable of producing crystalline ceramics through controlled or spontane-
ous crystallization. They can be fitted with stirrers to facilitate melt pool
convection and homogenization so that crystals do not accumulate in one
location in the melt pool.
CCIMs can also be used in conjunction with other technologies. For
example, an integrated process that combines an oxygen plasma and induc-
tion-heated cold crucible is reported by Vernaz and Poinssot (2008). This
process, which is still under development, is referred to as the Advanced
5 Simpler and more robust processing technologies are generally preferable because system
upsets can pose critical bottlenecks for operations that must be conducted in hot-cell environ-
ments to protect workers from high radiation fields.
OCR for page 95
95
WASTE PROCESSING AND WASTE FORM PRODUCTION
Hybrid System for Incineration and Vitrification (SHIVA). It consists of a
single reaction vessel that has three functions: (1) incineration, (2) vitrifica-
tion, and (3) gas post-combustion. This is a promising technology for pro-
cessing mixed wastes containing radioactive, organic, and other hazardous
chemical constituents that are difficult to separate by other processes. The
plasma decomposes organic material, significantly reducing its volume, and
produces a high-quality containment material (glass in this instance).
CCIM development began in France and Russia in the 1970s (Elliott,
1996). Russia is currently using CCIMs to process radioactive waste at the
Mayak Plant (Demine et al., 2001; Kushnikov et al., 1997; Lifanov et al.,
2003; Polyakov et al., 1997; Stefanovsky, 2009; Toumanov, 2003), and
the French are using a CCIM to vitrify HLW at an industrial scale at the
La Hague plant. DOE-EM is currently investigating CCIM technology for
possible use in its HLW immobilization programs.
CCIM is an emerging technology for the vitrification of fission product
solutions and decontamination waste streams. It also has potential applica-
tions for processing metallic waste streams (Vernaz, 2009). The underlying
technology is proven, but operational experience in large-scale waste stream
processing environments is limited in comparison to JHMs. Its deployment
in DOE-EM applications may require some up-front development work
to ensure its compatibility with specific process flow sheets, but no basic
research is likely to be required. Of course, a CCIM system must be engi-
neered for safety, for example, to prevent the loss of coolant. Gombert and
Richardson (2001) have discussed such safety aspects for CCIM design.
Because CCIMs are smaller per unit of throughput and operation-
ally more robust than JHMs, they could potentially be back-fitted to the
Defense Waste Processing Facility at SRS (Barnes et al., 2008) and the Waste
Treatment Plant at Hanford. For example, the DOE Independent Project
Evaluation Team (IPET, 2003) examined the feasibility of replacing the
JHMs in the Waste Treatment Plant at Hanford with CCIMs. It concluded
that two CCIMs could be retrofitted into each of the two melter cells in
the plant. If the melters were installed before the plant was hot commis-
sioned, only about four months would be required to modify the melter
cells and install the new equipment (IPET, 2003, p. 4.70). Additional time
would be required to install the melters after hot commissioning—either to
decontaminate the melter cells prior to installation of the new equipment
or to construct a new melter facility.
Although CCIMs have a track record of successful deployment inter-
nationally, the experience and understanding for DOE applications are
limited. Furthermore, the design of these melters is challenging in that,
for example, one needs to understand the material properties, particularly
the electrical resistivity of the melt, for determining the optimal operating
conditions for a given crucible diameter. Some of these design challenges
OCR for page 96
96 WASTE FORMS TECHNOLOGY AND PERFORMANCE
are summarized by Gombert and Richardson (2001). However, these are
the kinds of challenges that are normally faced when implementing any new
technology, and they can be overcome with appropriate development work.
4.3 IN-CONTAINER VITRIFICATION
In-Container Vitrification (ICV) (Thompson et al., 2003) is a batch
process by which contaminated soil, liquid waste mixed with soil, and glass
formers are vitrified in situ in a refractory-lined steel vessel. The energy
for melting is generated by passing an electrical current through graphite
electrodes installed in the vessel. The vitrified waste can be disposed of with
or without the container; in the latter case the container can be reused. A
similar concept for immobilizing contaminated soil in the ground, In-Situ
Vitrification, was developed earlier (Thompson et al., 1992).
ICV requires soil- or glass-formers to establish the melt and create a
stable, vitrified waste form. Therefore, soil or soil-like materials must be
added to the waste stream being processed. (The silica and alumina contained
in the soil are the glass formers for the process.) Because neither soil nor frit
is a good electrical conductor at ambient conditions, melting is initiated by
placing a conductive pathway between the electrodes.6 The heat generated
from graphite electrodes provides the electrical current in the vicinity of the
conductive pathway and melts the soil or frit, which increases its electrical
conductivity and establishes Joule heating. The melted zone gradually grows
outward toward the refractory liner, and eventually the entire vessel contents
become molten. The temperature of the melt varies between 1,400°C and
1,800°C depending on the composition of the soil and waste materials. It
usually takes up to three days to melt and process a single batch.
The melting process destroys organic contaminants contained in the
soil/waste mixture by pyrolysis or dechlorination. ICV can be equipped
with an off-gas treatment system to capture any residual organic or other
volatile materials released from the melt. The removal efficiency of the
organics is typically greater than 99.9999 percent (Thompson et al., 2003).
The retention efficiency of most metals and radionuclides in the melt
is greater than 99.99 percent for non-volatile species (Thomspon et al.,
2003). The residuals released with the off-gas stream are captured either
by filtration or scrubbing; radionuclide volatility is high because of the high
processing temperatures and the inability to monitor melt temperatures.
This is one of the major drawbacks of the process.
The vitrified product usually consists of a mixture of glass and crystal-
line materials. Convective currents generated during melting help mix the
6 A thin layer of graphite is added to the top of the waste-glass former mixture and touches
the graphite electrodes to initiate Joule heating.
OCR for page 97
97
WASTE PROCESSING AND WASTE FORM PRODUCTION
melt. In theory, if the melt is well mixed, the high content of glass-former
materials such as SiO2 and Al2O3 contained in the starting mixture can
produce a durable waste form. However, compositional non-uniformities
may exist that may greatly reduce durability.
The ICV technology was considered to treat low-activity waste (LAW)
at Hanford and has been piloted at the Demonstration Bulk Vitrification
System (DBVS), a full-scale test facility at the site. The process used soil
from the Hanford Site and glass-former additives as the starting mixture.
Several problems were encountered during the demonstration: High melting
temperatures caused vaporization of all of the technetium into the off-gas
stream. The Fe2O3 in the soil was reduced to Fe metal, which penetrated the
ceramic liners and the outer metal shell of the container. These operational
challenges are currently under review (Gerdes et al., 2007), and the project
is on hold (GAO, 2007).
4.4 SELF-SUSTAINING VITRIFICATION
Self-Sustaining Vitrification (SSV) utilizes exothermic chemical reac-
tions to produce glass, glass-ceramic, or ceramic waste forms. This technol-
ogy was initially developed in the Soviet Union for producing high-quality
ceramics and other refractory compounds (Borisov et al., 2002). The appli-
cation of this technology to radioactive waste immobilization is discussed
in Ojovan and Lee (2007).
The exothermic chemical reaction can be obtained by mixing waste
and powder metal fuels (PMFs) such as Mg, Al, Si, or Ca. The oxidation
of these fuels releases heat. The process is controlled by waste and PMF
compositions, which are established based on computer simulations. A
number of pre-processing steps may be required to develop the appropri-
ate starting compositions and properties. For example, water removal is
required to avoid excess gas generation and ensure the uniformity of the
resulting waste form.
Because it does not require expensive equipment, SSV can be particu-
larly useful for immobilizing small-volume waste streams or wastes that are
difficult to immobilize by other methods. The feasibility of producing glass-
ceramic waste forms using this technology has been demonstrated for a
number of waste materials, for example, contaminated clay soils and ashes,
spent inorganic ion exchangers, and calcined HLW (see Ojovan and Lee,
2007; Ojovan et al., 1999).
4.5 COLD PRESSING AND SINTERING
Cold Pressing and Sintering (CP&S) is one of the earliest technologies
used for forming technical crystalline ceramics. The technology itself is
OCR for page 108
108 WASTE FORMS TECHNOLOGY AND PERFORMANCE
4.10.4 Bitumen
Bitumen, a viscous hydrocarbon and a major component of asphalt,
has been used in Europe and Canada and to a lesser extent in the United
States to solidify and stabilize radioactive materials. Bitumen immobilizes
waste by encapsulation; it does not bind the waste chemically. The advan-
tages of bitumen as a waste form are simplicity of production, low operat-
ing cost, and leach resistant characteristics. However, bitumen does not
perform well with dehydrated salts, such as sodium sulfate, sodium nitrate,
magnesium chloride, and aluminum sulfate (Noyes, 1995). It can also be
a fire hazard (Zakharova and Masanov, 2000), especially when oxidizing
wastes like nitrates are involved.
There are several processes for solidifying bitumen with waste streams
(Ojovan and Lee, 2005). The most common are the use of a screw extruder
or a rotary thin-film evaporator. The screw extruder, which is commonly
used to mix pastes and plastics, consists of a screw, barrel, drive mecha-
nism, bitumen, and waste feed point. The constantly turning screw augers
the waste and bitumen through the heated barrel where it is heated to form
a homogeneous melt. The volatile gases are allowed to vent.
The thin-film evaporator consists of vertical vessel with a rotated shaft
at its center equipped with wiper blades, which help create a thin film on
the wall and help mix the bitumen-waste mixture. Preheated bitumen and
partially evaporated liquid waste are fed to the top of the evaporator, a
distributor spreads the mixture around the inner wall of the vessel where
heat is supplied, while the action of the wiper assemblies create a thin liquid
film on the wall and also help to mix the molten material. The evaporated
liquid passes through a series of condensers, and a molten bitumen waste
mixture is discharged from the bottom of the evaporator.
4.11 DISCUSSION
This chapter has provided a brief overview of processing technologies
used to produce a variety of waste forms, emphasizing technologies that
are currently used, planned to be used, or in development stage that may be
considered advancements. These technologies are generally well established
through their use in other industries. The attributes, the advantages, and
disadvantages of these processes are summarized in Table 4.1.
The committee highlighted the advantages of three technologies in its
interim report (NRC, 2010): fluidized bed steam reforming, cold crucible
induction melting, and hot isostatic pressing. These waste form production
technologies are being used commercially in both nuclear and/or non-
nuclear applications, as described elsewhere in this chapter, and appear to
be applicable for processing and immobilizing a range of DOE-EM waste
OCR for page 109
109
WASTE PROCESSING AND WASTE FORM PRODUCTION
streams, especially HLW streams, which are summarized in Table 4.2.
DOE-EM is already planning to apply these technologies to some of its
waste streams, as discussed in the interim report and elsewhere in this chap-
ter. The committee concurs with DOE-EM about the applicability of these
technologies in the cleanup program. Even though these are mature tech-
nologies and have applications in different industries, some development
work will be needed to use them in processing nuclear waste as described
in the interim report (NRC, 2010) and in this chapter.
OCR for page 110
110 WASTE FORMS TECHNOLOGY AND PERFORMANCE
TABLE 4.2 Waste Form Production by Various Processing Technologies
Other
Glasses Minerals/
(LaBs, FeP, Glass- Crystalline
AlP, chalco- Ceramic Ceramics/
Borosilicate gnide, and Materials Simple
Glass others) (GCMs) Oxides
Joule-Heated Melters Y Y Y N
(JHMs)
Advanced JHMs Y Y Y N
Heated Melters
(bubblers, etc.)
Cold Crucible Y Y Y Y*
Induction Melters
THERMAL
(CCIM)
Press + Sinter N N Y Y
Hot Isostatic Pressing N N Y Y
(HIPing)
Fluidized Bed Steam N N Y Y
Reforming (FBSR)
Calcining N N N Y
Hydrothermal N N Y Y
Processing
Mix and Set N N N N
NONTHERMAL
OCR for page 111
111
WASTE PROCESSING AND WASTE FORM PRODUCTION
Cements
Zeolites (OPC,
Metal Hydro- Ceramicrete, and
Matrix ceramic Others) Bitumen Geopolymers
N N N N N
N N N N N
Y N N N N
Y Y N N N
Y Y N N N
N Y N N N
N N N N N
N Y Y N Y
N N Y [to include Y (heating Y [to include
use of as waste necessary to use of as waste
forms and as melt bitumen) forms and as
binders (macro- binders (macro-
encapsulation)] encapsulation)]
OCR for page 112
112 WASTE FORMS TECHNOLOGY AND PERFORMANCE
APPENDIX 4.A
FLUIDIZED BED STEAM REFORMING
The Fluidized Bed Steam Reforming (FBSR) of nuclear waste is a rela-
tively new technology, though the fluidization phenomenon and steam reform-
ing are well established in the chemical engineering field. Steam reforming
is a method for generating hydrogen by reacting fossil fuels with water. For
example, for natural gas:
CH4(g) + H2O(g) → CO(g) + 3H2(g)
If coal is used as a carbon source, it first undergoes pyrolysis or devola-
tilization then the char (C) reacts with steam according to the following
reaction:
C(s) + H2O(g) → CO(g) + H2(g)
The H2 is combined with O2 so that no excess H2 exists in the system
at any one time. This combination is exothermic and provides energy in the
form of heat for the autocatalytic operation of the FBSR.
The FBSR consists of two fluidized beds. The first one operates in a
reducing environment, and its function is to evaporate the liquid nuclear
waste stream; destroy organics; reduce nitrates, nitrites, and nitric acid to
nitrogen gas; and form a stable solid waste product. The first-stage fluid-
ized bed of the FBSR process is referred to as the Denitration and Min-
eralization Reformer, or DMR. The DMR uses superheated steam as the
fluidizing media. The bed material consists of granular solid additives and
co-reactant(s), such as carbon, clay, silica, and/or catalysts. Liquid waste is
directly fed to the fluidized bed after minor pretreatment (e.g., to concen-
trate or dilute solubles) except the addition of clay.
By analogy to the above steam reforming chemistry, the carbon fed to
FBSR (coal in this instance) produces H2 and CO. For organic compounds
in the waste stream, which undergo pyrolysis to form various hydrocar-
bons, the reducing environment is generated by the following reaction:
CnHm(g) + nH2O(g) → nCO(g) + (n + m/2)H2(g)
Similarly, the nitrates contained in the liquid waste are reduced to
2NaNO3(g) + 3C(s) → 2NO(g) + 3CO(g) + Na2O(s)
In the steam environment, the sodium oxide is transferred to sodium
hydroxide:
OCR for page 113
113
WASTE PROCESSING AND WASTE FORM PRODUCTION
Na2O(s) + H2O(g) → 2NaOH(s,l)
yielding the overall reaction
2NaNO3(g) + 3C(s) + H2O(g) → 2NO(g) + 3CO(g) + 2NaOH(s,l)
2NaNO3(g) + 2C(s) + H2O(g) → 2NO2(g) + 2CO(g) + 2NaOH(l,s)
The NO and NO2 are further reduced to nitrogen gas by the reaction
of CO, C, or H2 generated from the reaction of the organic material with
steam as shown above. The nitrates can also be reduced by the addition of
a catalyst or a metal. For example:
2NaNO3(g) + 5Fe(s) + H2O(g) → N2(g) + 5FeO(s) + 2NaOH(s,l)
The second fluidized bed of the FBSR process operates in an oxidizing
environment and is referred to as the Carbon Reduction Reformer, or CRR.
The fluidizing gases are the off-gas from the first stage and added oxygen.
Its function is to gasify carbon fines carried over in the process gases from
the DMR, oxidize CO and H2 to CO2 and water, and convert trace acid
gases to stable alkali compounds by reacting these acids with the bed media
consisting of calcium carbonate and/or calcium silicate particles.
The addition of bulk aluminosilicates to the fluidized bed results in the
production of various phases including anhydrous feldspathoid mineral
analogue phases such as sodalite. The sodalite family of minerals (includ-
ing nosean) are unique because they have cage-like structures formed of
aluminosilicate tetrahedra. The remaining feldspathoid mineral analogues,
such as nepheline, have a silica “stuffed derivative” ring-type structure. The
cage structures are typical of sodalite and/or nosean phases where leach
testing has indicated that the cavities in the cage structure retain anions
and/or radionuclides, which are ionically bonded to the aluminosilicate
tetrahedra and to sodium cation.
Sodalite has the formula Na8[Al6Si6O24](Cl2). In sodalites and ana-
logues with sodalite topologies, the cage is occupied by two sodium and
two chlorine ions. When the 2NaCl are replaced by Na2SO4, the mineral
phase is known as nosean, (Na6[Al6Si6O24](Na2SO4)). Because the Cl, SO4,
and/or S2 are chemically bonded and physically restricted inside the soda-
lite cage structure, these species do not readily leach out of the respective
FBSR waste form mineral phases. Thus, FBSR waste forms can be useful
for immobilizing these species to prevent their leaching into groundwater.
Other minerals in the sodalite family, namely hauyne and lazurite,
which are also cage-structured minerals, can accommodate either SO42– or
S2–. They are potential products of the steam reforming depending on the
OCR for page 114
114 WASTE FORMS TECHNOLOGY AND PERFORMANCE
REDOX of the sulfur during the process. Sodalite minerals are known to
accommodate Be in place of Al and S2 in the cage structure along with Fe,
Mn, and Zn, e.g., helvite (Mn4[Be3Si3O12]S), danalite (Fe4[Be3Si3O12]S),
and genthelvite (Zn4[Be3Si3O12]S). These cage-structured sodalites were
minor phases in HLW supercalcine waste forms and were found to retain
Cs, Sr, and Mo into the cage-like structure, e.g., Mo as Na6[Al6Si6O24]
(NaMoO4)2. In addition, sodalite structures are known to retain B, Ge,
I, and Br in the cage-like structures. Indeed, waste stabilization at Idaho
National Laboratory currently uses a glass-bonded sodalite ceramic waste
form (CWF) for disposal of electrorefiner wastes for sodium-bonded metal-
lic spent nuclear fuel from the EBR II fast breeder reactor.
OCR for page 115
115
WASTE PROCESSING AND WASTE FORM PRODUCTION
REFERENCES
ASME [American Society of Mechanical Engineers]. 1985. The Evolution of HIP: Commemo-
rating the First Hot and Cold Isostatic Processing Vessels, ASME Landmarks Program,
Available at http://files.asme.org/ASMEORG/Communities/History/Landmarks/5569.pdf.
Bao, Y., M. W. Grutzeck, and C. M. Jantzen. 2005. “Preparation and Properties of Hydro-
ceramic Waste Forms Made with Simulated Hanford Low Activity Waste (LAW),” J.
Amer. Ceram. Soc. 88(12), 3287-3302.
Barnes, A. B., D. C. Iverson, A. J. Brannen, and E. Tchemitcheff. 2008. “Feasibility Evalu-
ation and Retrofit Plan for Cold Crucible Induction Melter Deployment in the Defense
Waste Processing Facility at Savannah River Site–8118,” WM2008 Conference, Phoenix,
Ariz. (WSRC-STI-2007-00669, Rev 1), Available at http://sti.srs.gov/fulltext/WSRC-STI-
2007-00669R.pdf.
Bibler, N. E. 2005. Measuring and Predicting Fission Product Noble Metals in Savannah
River Site High Level Waste Sludges, WSRC-TR-2005-00098, Savannah River National
Laboratory, Aiken, S.C.
Borisov, A. A., L. DeLuca, and A. Merzhanov. 2002. Self-Propagating High-Temperature Syn-
thesis of Materials, Combustion Science & Technology Book Series 5, Sheridan Books,
Ann Arbor, Mich.
Buckham, J. A., A. L. Ayers, and J. A. McBride. 1966. “Fluidized Bed Calcination of High-
Level Radioactive Waste in a Plant-Scale Facility,” Chemical Engineering Progress, Sym-
posium Series, Nuclear Engineering Part XV, 62, 52-63.
Caurant, J. D., P. Loiseau, O. Majerus, V. Aubin-Chevaldonnet, I. Bardez, and A. Quintas.
2009. Glasses, Glass-Ceramics and Ceramics for Immobilization of Highly Radioactive
Nuclear Wastes, Nova Science Publishers, Inc., New York.
Clarke, D. R. 1981. “Preferential Dissolution of an Intergranular Amorphous Phase in a
Nuclear Waste Ceramic,” J. Amer. Ceram. Soc. 64, C89-C90.
Cooper, J. A., D. R. Cousens, J. A. Hanna, R. A. Lewis, S. Myhra, R. L. Segall, R. St. C.
Smart, P. S. Turner, and T. J. White. 1986. “Intergranular Films and Pore Surfaces in
Synroc C: Structure, Composition, and Dissolution Characteristics,” J. Amer. Ceram.
Soc 69(4), 347-352.
Demine, A. V., N. V. Krylova, P. P. Polyektov, I. N. Shestoperov, T. V. Smelova, V. F. Gorn, and
G. M Medvedev. 2001. “High Level Liquid Waste Solidification using a “Cold” Crucible
Induction Melter,” Mat. Res. Soc. Symp. Proc. 663, 27-33.
DOE [U.S. Department of Energy]. 1998. Innovative Technology Summary Report: Plasma
Hearth Process, Washington, DC: U.S. Department of Energy, Office of Environmental
Management.
DOE. 2010. Amended Record of Decision: Idaho High-Level Waste and Facilities Disposi-
tion Final Environmental Impact Statement, Revised by State 12/21/09, Federal Register
75(1), 137-140. Available at http://edocket.access.gpo.gov/2010/pdf/E9-31151.pdf.
Elliott, M. L. 1996. Letter report: Cold Crucible Melter Assessment, PNNL-11018, Pacific
Northwest National Laboratory, Richland, Wash.
Fuhrmann, M., R. M. Neilson Jr., and P. Colombo. 1981. A Survey of Agents and Tech-
niques Applicable to the Solidification of Low-Level Radioactive Wastes, BNL-51521,
Brookhaven National Laboratory, Upton, New York.
GAO [Government Accountability Office]. 2007. DOE Should Reassess Whether the Bulk Vit-
rification Demonstration Project at Its Hanford Site Is Still Needed to Treat Radioactive
Waste, GAO-07-762, GAO, Washington, D.C. (June 12).
OCR for page 116
116 WASTE FORMS TECHNOLOGY AND PERFORMANCE
Gerdes, K., J. O. Honeyman, J. Longnecker, N. Machara, and B. M. Mauss. 2007. Demon-
stration Bulk Vitrification System (DBVS) ExternaI Review, CH2M-32706-FP (Rev 0),
CH2MHILL Hanford Group, Richland, Wash. (February), Available at http://www.osti.
gov/bridge/servlets/purl/899770-b42JCN/899770.pdf.
Gombert, D. and J. G. Richardson. 2001. Cold Crucible Induction Melter Technology:
Results of Laboratory Directed Research and Development, INEEL/EXT-01-01213,
Idaho National Engineering and Environmental Laboratory, Idaho Falls.
Harker, A. B., E. D. Morgan, and F. Flintoff. 1984. “Hot Isostatic Pressing of Nuclear Waste
Glasses,” J. Amer. Ceram. Soc. 67(2), C-26–C-28.
Hermann, E., C. Kunze, R. Gatzweiler, G. Kiesig, and J. Davidovits. 1999. “Solidification of
Various Radioactive Residues by Géopolymère with Special Emphasis on Long-Term
Stability,” Géopolymère ‘99 Proceedings, 2nd International Conference on Geopolymers,
Saint-Quentin, France.
Hrma, P., P. Schill, and L. Nemec. 2002. Settling of Spinel in a High-Level Waste Glass Melter,
PNNL-13747, Pacific Northwest National Laboratory, Richland, Wash., Available at
http://www.pnl.gov/main/publications/external/technical_reports/pnnl-13747.pdf.
IPET [Independent Project Evaluation Team]. 2003. Technical Evaluation of Hanford HLW
Vitrification Process Alternatives: Report of the Independent Project Evaluation Team.
U.S. Department of Energy, Office of River Protection, Richland, Wash.
Jain, V. 1998. Survey of Solidification Process Technologies, CNWRA 98-005 (NUREG/
CR-6666), Center for Nuclear Waste Regulatory Analyses, San Antonio, Texas. (April).
Jantzen, C. M. 1991. “Glass Melter Off-gas System Pluggages: Cause, Significance, and Reme-
diation,” Proceedings of the 5th International Symposium on Ceramics in Nuclear Waste
Management, G. G. Wicks, D. F. Bickford, and R. Bunnell (Eds.), American Ceramic
Society, Westerville, Ohio, 621-630.
Jantzen, C. M. 2003. Disposition of Tank 48H Organics by Fluidized Bed Steam Reforming
(FBSR), WSRC-TR-2003-00352, Westinghouse Savannah River Company, Aiken, S.C.
Jantzen, C. M. 2006. Fluidized Bed Steam Reformer (FBSR) Product: Monolith Formation
and Characterization, WSRC-STI-2006-00033, Savannah River National Laboratory,
Aiken, S.C. (June).
Jantzen, C. M. and D. Lambert. 1999. Inspection and Analysis of the Integrated DWPF Melter
System (IDMS) After Seven Years of Continuous Operation (U), WSRC-MS-99-00336,
Westinghouse Savannah River Company, Aiken, S.C.
Jantzen, C. M. and M. R. Williams. 2008. “Fluidized Bed Steam Reforming (FBSR) Miner-
alization for High Organic and Nitrate Waste Streams for the Global Nuclear Energy
Partnership (GNEP),” WM2008, Phoenix, Ariz., Available at http://www.wmsym.org/
archives/2008/pdfs/8314.pdf.
Jantzen, C. M. In press. “Development of Glass Matrices for High Level Radioactive Wastes,”
In Handbook of Advanced Radioactive Waste Conditioning Technologies, M. I. Ojovan
(Ed.), Woodhead Publishing, U.K., Chapter 9.
Juoi, J. M., M. I. Ojovan, and W. E. Lee. 2008. “Microstructure and Leaching Durability of
Glass Composite Wasteforms for Spent Clinoptilolite Immobilisation,” J. Nucl. Mat.
372, 358-366.
Keppler, H. 1990. “Ion Exchange Reactions between Dehydroxylated Micas and Salt Melts
and the Crystal Chemistry of the Interlayer Cation in Micas,” Amer. Mineral. 75,
529-538.
Kushnikov, V. V., Y. I. Matiunin, and T. V. Smelova. 1997. “Induction Melting in a Cold
Crucible for Immobilization of Plutonium-Containing Wastes,” Atomic Energy 83(5),
801-806.
Lifanov F. A., M. I. Ojovan, S. V. Stefanovsky, and R. Burcl. 2003. “Cold Crucible Vitrification
of NPP Operational Waste,” Mat. Res. Soc. Symp. Proc. 757, 201-206.
OCR for page 117
117
WASTE PROCESSING AND WASTE FORM PRODUCTION
Marshall, D. W., N. R. Soelberg, and K. M. Shaber. 2003. THORsm Bench-Scale Steam
Reforming Demonstration, INEEL/EXT-03-00437, Idaho National Engineering & Envi-
ronmental Laboratory, Idaho Falls.
Mason, J. B., T. W. Oliver, M. P. Carson, and G.M. Hill. 1999. “Studsvik Processing Facility
Pyrolysis/Steam Reforming Technology for Volume and Weight Reduction and Stabiliza-
tion of LLRW and Mixed Wastes,” Waste Management ’99, Tucson, Ariz., Available at
http://www.wmsym.org/archives/1999/60/60-3.pdf.
McFarlane, H. F., K. M. Goff, F. S. Felicione, C. C. Dwight, and D. B. Barber. 1997. “Hot
Demonstrations of Nuclear-Waste Processing Technologies,” JOM 49(7), 14-21, 83,
Available at http://www.tms.org/pubs/journals/JOM/9707/McFarlane-9707.html.
Meyers, B., G. A. Armantrout, C. M. Jantzen, A. Jostsons, J. M. McKibben, H. F. Shaw, D. M.
Strachan, and J. D. Vienna. 1998. Technical Evaluation Panel Summary Report: Ceramic
and Glass Immobilization Options, UCRL-ID-129315, Lawrence Livermore National
Laboratory, Livermore, Calif.
Newby, B. J. and B. H. O’Brien. 2000. Summary of Waste Calcination at the Idaho Nuclear
Technology and Engineering Center, INEEL/EXT-2000-01206, Idaho National Engineer-
ing and Environmental Laboratory. (October).
Noyes, R. 1995. Nuclear Waste Cleanup Technology and Opportunities, Noyes Publications,
Park Ridge, N.J.
NRC [National Research Council]. 2010. Waste Forms Technology and Performance: Interim
Report, National Academies Press, Washington, D.C.
O’Connor, W. K. and P. C. Turner. 1999. Thermal Treatment of Solid Wastes Using the Electric
Arc Furnace, DOE/ARC-99-010, Presented at Global Symposium on Recycling, Waste
Treatment and Clean Technology, Sebastian, Spain, Available at http://www.osti.gov/
bridge/servlets/purl/821858-VgI7Sa/native/821858.pdf.
Ojovan, M. I. (Ed.). 2011. Handbook of Advanced Radioactive Waste Conditioning Tech-
nologies, Woodhead Publishing Series in Energy, 12, Woodhead Publishing Limited,
Cambridge, U.K.
Ojovan, M. I., G. A. Petrov, S. V. Stefanovsky, and B. S. Nikonov. 1999. “Processing of Large-
scale Radwaste-containing Blocks using Exothermic Metallic Mixtures,” Mat. Res. Soc.
Symp. Proc. 556, 239-246.
Ojovan, M. I. and W. E. Lee. 2005. An Introduction to Nuclear Waste Immobilisation,
Elsevier, Amsterdam.
Ojovan, M. I. and W. E. Lee. 2007. New Developments in Glassy Nuclear Wasteforms. Nova
Science Publishers, New York.
Oversby, V. M. and E. R. Vance. 1994. Comparison of Ceramic Waste Forms Produced by
Hot Uniaxial Pressing and by Cold Pressing and Sintering, UCRL-JC-117039, CONF
-941075-8, Lawrence Livermore National Laboratories, Livermore, Calif. (September
1994).
Polyakov, A. S., L. A. Mamaev, O. L. Masanov, and K. P. Zakharova. 1997. “Strategy of
Active Waste Management in the Russian Federation,” Nucl. Eng. and Design 173,
269-275.
Riley, A., S. Walker, and N. R. Gribble. 2009. “Composition Changes and Future Challenges
for the Sellafield Waste Vitrification Plant,” Scientific Basis for Nuclear Waste Manage-
ment XXXIII, Materials Research Society, Pittsburgh, Penn., 267-273.
Ringwood, A. E., S. E. Kesson, K. D. Reeve, D. M. Levins, and E. L Ramm. 1988. “Synroc,”
In Radioactive Waste Forms for the Future, W. Lutze and R. C. Ewing (Eds.), North
Holland, Amsterdam, Chapter 4.
Soelberg, N. R., D. W. Marshall, S. O. Bates, and D. D. Taylor. 2004. Phase 2 THOR Steam
Reforming Tests for Sodium Bearing Waste Treatment, INEEL/EXT-04-01493, Rev. 1,
Idaho National Engineering and Environmental Laboratory, Idaho Falls.
OCR for page 118
118 WASTE FORMS TECHNOLOGY AND PERFORMANCE
Staples, B. A., T. P. O’Holleran, D. A. Knecht, P. C. Kong, S. J. Johnson, K. Vinjamuri, A.
Schoies, A. W. Erickson, S. M. Frank, K. Meyer, M. Hansen, E. L. Wood, and H. C.
Wood. 1997. Preparation of Plutonium Waste Forms with ICPP Calcined High-Level
Waste, INEEL/EXT-—97-00496, Idaho National Engineering and Environmental Labo-
ratory, Idaho Falls. (May).
Stefanovsky, S. V. 2009. “Ceramic and Phosphate Glass Waste Forms and Cold Crucible
Melting Technology,” presented at the Committee on Waste Forms Technology and
Performance, Meeting #3, November 4, 2009, Washington, D.C.
Sutton, A. H., J. C. Bishop, M. H. Cohen, and K. J. Stahman. 1966. “Reduction and Hydro-
fluirination of Uranium Concentrates by Fluid Bed Techniques,” Chemical Engineering
Progress, Symposium Series, Nuclear Engineering Part XV, 62, 20-30.
Thompson, L. E., S. O. Bates, and J. E. Hansen. 1992. Technology Status Report: In Situ Vitri-
fication Applied to Buried Wastes, PNL-8219, Pacific Northwest Laboratory, Richland,
Wash.
Thompson, L. E., P. S. Lowery, H. W. Arrowsmith, T. Snyder, and J. L. McElroy. 2003.
Development of the Bulk Vitrification Treatment Process for the Low Activity Fraction
of Hanford Single Shell Tank Wastes, WM’03, Tucson, Ariz., Available at http://www.
wmsym.org/archives/2003/pdfs/368.pdf.
Toumanov, I. N. 2003. Plasma and High Frequency Processes for Obtaining and Processing
Materials, Nova Publishing, New York.
Vernaz, E. Y. (Ed.). 2009. Nuclear Waste Conditioning. Commissariat à l’Énergie Atomique,
Paris, France.
Vernaz, E. Y. and C. Poinssot. 2008. “Overview of the CEA French Research Program on
Nuclear Waste,” Mat. Res. Soc. Symp. Proc. 1107, 21-32.
Wahlquist, D. R. 1996. Design and Fabrication of a Glovebox for the Plasma Hearth Process
Radioactive Bench-Scale System, ANL-TD-CP-90478, Argonne National Laboratory-West,
Idaho Falls, Idaho, Available at http://www.osti.gov/bridge/servlets/purl/266620-Nj74mo/
webviewable/266620.pdf.
Zakharova, K. P. and O. L. Masanov. 2000. “Bituminization of Liquid Radioactive Wastes.
Safety Assessment and Operating Experience,” Atomic Energy 89(2), 646-649.
Zhang, Z. and M. L. Carter. 2010. “An X-ray Photoelectron Spectroscopy Investigation of
Highly Soluble Grain-Boundary Impurity Films in Hollandite,” J. Amer. Ceram. Soc.
93(3), 894-899.
