Application of 10B-Lined Proportional Counters to Traditional Neutron Counting
Applications in International Safeguards*
R.D. McElroy Jr., and S. Croft
Oak Ridge National Laboratory
One Bethel Valley Road, PO Box 2008, MS-6166, Oak Ridge, Tennessee
United States of America
Abstract. Many neutron-detection techniques are under consideration as potential replacements for
3
He gas-filled proportional tubes, which are widely used in international nuclear safeguards
applications. The traditional 10B-lined proportional tube is a commercial off-the-shelf (COTS)
technology for neutron detection that predates the development of 3He detectors. There is also a long
history of its use in neutron counting facilities and in modeling, design, and testing of assay systems.
When compared to the familiar 3He tube, cylindrical 10B-lined proportional counters meet or exceed all
relevant criteria (e.g., stability, resistance to gamma-ray exposure) with the exception of neutrondetection efficiency per unit volume. The thin boron coating thickness and practical active detection
area per unit volume limit the measurement performance ultimately achievable with these detectors;
however, assay systems based on the 10B-lined proportional detector can be constructed with sufficient
measurement performance to achieve the International Target Values [1] for a significant subset of
traditional safeguards neutron counting applications. Additionally, these detectors are also well suited
to a number of active neutron interrogation applications such as the differential die-away and 252Cf
shuffler techniques, which can be accomplished with lower detection efficiency than when using the
detectors’ passive counterparts. As part of our studies of alternative 3He technologies, we examined
the performance of a set of commercially manufactured neutron-slab-counting assemblies configured
as a neutron coincidence collar, a passive neutron coincidence well counter, and the detector assembly
within a large-cavity 252Cf shuffler. Measurement performances are presented and compared with
those of the standard 3He-based counting counterparts. These performance levels represent a baseline
of what can be achieved using COTS neutron counting systems with no significant development
required. They are a minimum baseline against which the anticipated performance of any other
potential alternative technology, and the associated portion of the safeguards application space it could
address, can be judged.
1. Introduction
The high cost and limited availability of 3He have provided an impetus for development of alternative
neutron-detection technologies. These developments encompass a broad range of detection
technologies such as liquid and plastic organic scintillation detectors, coated solid state detectors, high
pressure noble gases, and 10B-lined proportional detectors. The development activities are often
launched without a full understanding of the measurement need or may address only an overly
specialized measurement assay situation such that while a novel detection technique may be
technically sound for a specific measurement scenario, it may be completely impractical as a
safeguards assay solution given the realistic breadth of challenges fielded systems face. The High
Level Neutron Coincidence Counter II (HLNCC-II) is often cited as a reference assay system for
development of alternative detection technologies because it is widely used and its detection
performance is well documented. However, the measurement performance of the HLNCC-II is not
very demanding relative to safeguards coincidence counting, and a development goal to meet the
performance targets for the HLNCC-II should be considered as the bare minimum performance at
which a technology becomes useful in any general way in safeguards applications.
*
Notice: This manuscript has been authored by UT-Battelle, LLC, under Contract No. DE-AC05-00OR22725
with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the
article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up,
irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to
do so, for United States Government purposes.
1
The 10B-lined proportional detector would seem an obvious option to consider as a replacement for
3
He proportional tubes in neutron coincidence counting (NCC) systems used in international
safeguards applications. These detectors represent an established technology first developed more than
50 years ago, with enriched 10B neutron-sensitive material expected to be readily available well into
the future. The traditional 10B-lined detector technology is less sensitive to gamma-ray interference
when operated with a high threshold than are 3He detectors, fits well into the existing safeguards
infrastructure, and (unlike many other alternative technologies such as BF3) poses no added safety
concerns relative to the 3He-based systems. Although the 10B-lined technology is a good match at the
level of an individual proportional tube, there are a number of performance requirements that must be
addressed before this technology can achieve widespread use in NCC systems. Of principal concern is
the low efficiency density achievable. Several development activities [2–9] seek to address this
limitation by various methods with varying levels of success. However, studies on the potential
performance of the traditional 10B-lined proportional tube have shown that even without any additional
developments, this commercial off-the-shelf (COTS) detector type should be able to address a
significant portion of the safeguards neutron-detection needs (Table I). The performance expectation
for these detectors to date had been primarily based on modeling exercises [10–12]. In this work, we
adapt a prototype counter developed by GE/Reuter-Stokes (GERS), the ABUNCL, to demonstrate the
potential application of the traditional 10B-lined tubes for nuclear safeguards applications [8, 9].
Table I. Standard neutron counting systems and the possibility of constructing a boron-lined, tubebased replacement system [10]. Also shown are the fission neutron detection efficiency, , principal
1/e neutron detection die-away time, , and the figure of merit ( ⁄√ ).
Neutron
Figure of
B-Lined
Die-Away
System
Detection
Merit
Variant
Time (µs)
Efficiency (%)
(FOM)
Feasible?
Slab, Well, and Collar Counters
Portable Slab Counter
UFBR-Universal Fast Breeder
UNCL—Coincidence Collar
PCAS – Canister Assay System
Curved Slab Counter
HLNCC-II
Flat Squared Counter
AWCC—Active Well
INSV—Inventory Sample
OSL—On-Site Laboratory
PSMC—Multiplicity Counter
Drum Counters
WCAS—Waste Crate Assay
252
Cf Drum Shuffler
Waste Drum Assay System
IWAS—Passive/Active Drum
HENC—Drum Assay System
HNMC—Hexagonal
Multiplicity Drum Counter
2.2
7.0
14.6
15.1
19.4
17.9
24.4
24.9
35.0
40.0
54.0
50
22
54
57
75
43
56
52
45
54
50
3
15
20
20
22
27
33
35
52
54
76
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
16.6
17.5
19.4
27.0
30.8
74
80
79
44
52
14
17
20
47
52
Yes
Yes
Yes
No
No
51.0
28
137
No
2. Baseline Well Counter Incorporating 10B-Lined Proportional Tubes
The GERS-NCC is a passive neutron coincidence well counting (NCC) system adapted from the
GERS ABUNCL [8] neutron coincidence collar for use in the evaluation of traditional 10B-lined
proportional tubes in quantitative safeguards neutron assay systems. The ABUNCL was developed and
fabricated by GERS as a demonstration unit and possible alternative to the traditional 3He-based
uranium neutron coincidence collar (UNCL) [13] and has been evaluated by both Pacific Northwest
2
National Laboratory and Los Alamos National Laboratory [8]. The observed measurement
performance parameters (Table II) for the ABUNCL were somewhat poorer in comparison to the 3Hebased neutron collar in use today by the International Atomic Energy Agency but are sufficient to
achieve the required performance specified by the International Target Values (ITVs) [1] with
increased measurement times (~50% longer).
Table II. Observed measurement performance (with a 252Cf fission source) for the
ABUNCL and the standard 3He-based neutron coincidence collar system
3
10
He Proportional Tube
B-Lined Proportional Tube
Neutron Collar (JCC-71)
Neutron Collar (ABUNCL)
Passive
Active
Passive [8]
Active [9]
Efficiency
11.3%
9.0%
11.60%
9.40%
Die-Away
52 μs
52 μs
75 μs
83 μs
Figure of Merit ( ⁄√ )
15.8
12.5
13.4
10.3
Weight
38 kg
~100 kg
Dimensions
42 × 42 × 52 cm (L × W × H)
36 × 51 × 80 cm (L × W × H)
3
Note: The He counter performances are with cadmium liners installed, while no liners
are present in the ABUNCL.
To examine the potential performance of the 10B-lined proportional tube for use in other safeguards
neutron coincidence counting systems, cadmium liners and high-density polyethylene end plugs were
fitted to the ABUNCL (Fig. 1). Although, the configuration was not optimal (e.g., rectangular rather
than cylindrical), the general arrangement was similar in most important respects to that of a typical
neutron coincidence well counter such as the HLNCC-II [14]. Testing examined general detection
response, stability, and temperature dependence.
FIG. 1. Photograph of the GERS NCC (ABUNCL) with end plugs fitted (left) and sketch of the tube
arrangement (right).
A summary of the performance parameters for the 10B-lined NCC is presented in Table III along with
the corresponding values for the HLNCC-II. To achieve the same level of statistical precision using
the 10B-lined NCC as is possible with the 3He-based HLNCC-II, it is necessary to count approximately
three times longer. The systematic error contributors (e.g., calibration error, sample uniformity) will,
however, be approximately the same for the two systems. From this we can conclude that the
performance of the highly non-optimized 10B-lined NCC system is sufficient to achieve the ITVs
(Table IV), albeit in 1000 s assay time rather than 300 s. Monte Carlo simulations suggest that the
measurement performance of the HLNCC-II could be met by a near optimized assay system
incorporating 160 traditional (25.4 mm outside diameter [OD]) 10B-lined proportional tubes [11].
3
Table III. Observed measurement performance for the ABUNCL configured for well
counting and a standard 3He-based NCC (HLNCC-II)
3
10
He Proportional Tube NCC
B-lined Proportional Tube
(HLNCC-II)
NCC
3
10
Detection
4 atm He
B coating
Number Tubes
18
72
Active Length
50 cm
61 cm
Efficiency (ε)
17.9%
10.7%
Die-Away (τ)
43 μs
65 μs
Figure of Merit ( ⁄√ )
27.3
13.4
240
Sensitivity (Reals/g Pueff)
18.1
6.6
Weight
43 kg
120 kg
Sample Cavity
17.5 × 41 cm (OD × H)
16 × 22 × 38 cm (L × W × H)
Dimensions
34 × 76 cm (OD × H)
36 × 51 × 80 cm (L × W × H)
Note: The ABUNCL results are without a cadmium liner.
Table IV. HLNCC Random (r) and Systematic (s) Uncertainty ITVs [1]
Material
μ(r)
μ(s)
Combined
Plutonium Oxide
1
0.5
1.1
Mixed Oxide (>10% Pu)
2
0.5
2.1
Mixed Oxide (<10% Pu)
4
1.5
4.3
Mixed Oxide (clean scrap)
5
2
5.4
3. Use of the 10B Proportional Tube in Other Safeguards Systems
Table I suggests that these detectors may also be useful for other safeguards applications such as the
252
Cf shuffler [15] or differential die-away (DDA) [16] active interrogation techniques. The 252Cf
shuffler has been used for both quantitative assay of product materials and waste assay applications.
Typical measurement performance of the 252Cf shuffler for cans of product materials containing
kilogram quantities of 235U is generally around 0.3% precision for a single 1000 s assay, making it one
of the highest-performing non-destructive assay (NDA) systems for safeguards. DDA systems, on the
other hand, are among the most sensitive NDA techniques, typically providing 5 mg detection levels
for 200 L waste containers in a few hundred seconds or less. Both the shuffler and the DDA technique
have also been used for high-level waste applications such as hull monitoring [17]. Both systems
require substantial personnel shielding to protect the operator from the interrogating neutron source (5
× 107 to 1 × 109 n/s) where a need for somewhat larger detector volumes would not have a substantial
impact on the overall system footprint if needed. As an example of the potential application of 10Blined proportional tube as a replacement neutron detector for these systems, the actual 10B-lined
detector modules discussed above were integrated into Oak Ridge National Laboratory’s (ORNL’s)
waste drum shuffler. (The waste drum shuffler was chosen for convenience because it was a simple
matter to integrate the new detector modules into the assay cavity and acquisition electronics.)
The 252Cf shuffler measures the delayed neutron emission rate from the sample following irradiation
by a large (typically 1 × 108 to 2 × 109 n/s) 252Cf source. The observed delayed neutron rate is
approximately proportional to the fissile mass of the sample (although generally the response function
is represented by a quadratic or cubic form). Because the shuffler relies on totals neutron counting
rather than coincidence techniques, high neutron detection efficiencies are not required, and precision
can be improved by increasing the number of shuffle cycles. To illustrate this point, a proof-ofprinciple measurement was performed using an existing 252Cf drum shuffler and portions of the GERS
ABUNCL system. The shuffler’s 3He-based neutron detector assemblies were replaced by three GERS
10
B-based detector slabs, as shown in Fig. 2. The GERS slab segments, containing 46 10B-lined
proportional tubes divided into six detector banks, were arranged in a U-shape facing the 252Cf
4
interrogating neutron source, creating a non-optimized but functional canister counting assembly with
the following properties:
Cavity Dimensions:
Detection Efficiency:
Die-Away Time:
24 × 17 × 38 cm (L × W × H)
4.7%
65 μs
FIG.2. The ORNL 252Cf drum shuffler depicted at left with three slabs of the GERS 10B-lined counter
installed in the assay cavity to replace the 3He detector assemblies shown at right.
To examine the measurement performance of the modified shuffler, several replicate assays were
performed on four containers of enriched uranium in the form of U3O8 powder. The response as a
function of 235U mass is shown in Fig. 3. For each container, nine replicate assays, each of 15 min
duration, were performed and the results averaged to compensate for the relatively weak 252Cf source
(~2.5 × 107 n/s) interrogation source used for this study. Averaging over the nine assays provided
measurement performance roughly equivalent to a single 15 min assay using a 100 μg 252Cf source (2.3
× 108 n/s) more typical of these systems. The observed measurement precision for the 200 g 235U
standard was 0.9%. For comparison, the typical measurement precision obtained in a standard active
well coincidence counter, based on Am(Li)-induced fission coincidence counting, is 3.8% for the same
200 g standard in 1000 s.
FIG 3.The measured 252Cf shuffle response as a function of 235U effective mass for the proof-of-concept
system based on three GERS 10B-lined detector slabs and a relatively weak 2.5 × 107 n/s 252Cf
interrogation source. (Error bars are included but are too small to be visible.) The sketch at right
shows the sample position within the detection assembly and the position of the GERS counter within
the shuffler. The measurement items were all similar in geometry but varied in enrichment with the
highest mass item being 93.2 wt% 235U.
5
4. General Properties of the 10B-Lined Counters
The 10B-lined and 3He proportional tubes have similar properties, facilitating the integration of the 10Blined detectors into the current infrastructure. Novel technologies in nuclear safeguards applications,
such as neutron scintillation detectors, can be challenged by environmental factors such as
temperature, sample properties such as gamma-ray sensitivity, and mechanical configuration such as
system footprint. These issues are already addressed by the 10B-lined proportional tubes.
4.1
Gamma-Ray Tolerance
One of the key features of a neutron detector in safeguards applications is the relative insensitivity of
the 3He tube to gamma-ray exposure. The typical 3He-based safeguards system is designed to
accommodate 5 mSv/h exposure rates, although many systems have been developed to tolerate
exposures on the order of 1 to 10 Sv/h. The 10B-lined tubes have similar gamma-ray sensitivity [18]
properties. For example, passive and active measurements of high-exposure-rate 233U samples (>2
mSv/h at the inner surface of the assay cavity) were performed at ORNL using the reconfigured
ABUNCL system. No excess counts were observed due to the high gamma-ray exposure rate.
4.2
Temperature Dependence
NDA measurements are often performed in non-environmentally controlled facilities, so it is important
that the detectors be relatively insensitive to changes in temperature. Two of the ABUNCL slabs were
placed into an environmental chamber, one with the amplifier modules located within the chamber and
the other with amplifiers located outside the chamber in constant temperature (Fig. 4). The response
with amplifiers external to the chamber decreased linearly with temperature as expected due to the 1/v
thermal cross-section dependence. The temperature dependence for the slab with the amplifiers also in
the environmental chamber became more complex, but the efficiency stayed within a +/–1% range
from –40 to +40°C. For comparison purposes a standard 3He slab placed in the environmental chamber
varied ±1% over the temperature range of –40 to +40°C.
FIG 4. Comparison of the temperature dependence of the 10B-lined slabs with the amplifiers inside and
outside of the environmental chamber. At right is a photo of the detector slabs within the
environmental test chamber.
4.3
Scalability
Detection modules based on the 10B-lined proportional tube are scalable in the same manner as the
3
He-based modules in the sense that it is possible to construct larger assay systems simply by
increasing the number and active length of the tubes. For example, the GERS NCC slabs could be
expanded to create a waste drum assay system (WDAS) [19] of the kind used in plutonium-fuel
facilities. The 10B-lined detector alternative to the WDAS would provide the same performance and
have the same overall system footprint but would require 486 proportional tubes arranged in three
rows about the rectangular assay cavity. The potential downside to this detector arrangement is cost
(compared to historical expectations) due to the large number of tubes required to achieve the existing
detection efficiency and die-away time.
6
An experience-based review of NDA system installations over a 15 year period suggests that about
32% of 3He gas used in safeguards-related neutron counting systems could be replaced by traditional
10
B-lined proportional tube-based systems. The applications not addressable by these technologies are
the higher-performance multiplicity counters, which consume a disproportionate share of the 3He gas;
however, 10B-lined tubes could address approximately 62% of NDA systems fielded. The fraction of
3
He gas consumption for safeguards applications that is replaceable increases to about 50% if the novel
10
B-lined technologies currently under development are included, accounting for 82% of systems
deployed. For comparison, 64% of the historical safeguards 3He consumption and almost 90% of
systems deployed could have been replaced by BF3 proportional counters (which have generally been
deemed an unacceptable replacement due to the hazardous nature of the BF3 gas).
5.
10
B-Lined Proportional Tubes as a 3He Baseline Alternative Technology
Alternative neutron detector developments include a broad range of candidate technologies such as
liquid, plastic and glass scintillators, solid-state detectors, high-pressure noble gas detectors, and novel
10
B-lined detectors (e.g., straw detectors). While many of these detectors perform well for a specific
application and material form, their overall utility to international safeguards may be limited due to
limitations of the detector such as gamma-ray sensitivity, safety considerations, and life-cycle cost.
The traditional 10B-lined proportional tube represents an existing neutron technology that can be
integrated into existing safeguards programs with minimal to no development required. NDA systems
based on these detectors require no modifications to existing software or data acquisition systems and
so place no burden on the existing support infrastructure. The detectors meet all of the criteria for a
replacement technology (e.g., simplicity, gamma-resistance [19], stability, reliability, scalability) with
the exception of efficiency. However, even with the limitation on efficiency, these detectors address a
significant fraction of the current safeguards neutron-detection needs (by our estimates about one-third
of the historic 3He consumption and two-thirds of systems deployed).
6. Conclusion and Discussion
The traditional 25.4 mm diameter cylindrical 10B-lined proportional tube technology can address a
broad range of neutron counting needs in international nuclear safeguards today without the need for
extensive infrastructure development. These tubes represent an essentially COTS solution that can be
applied immediately. The 10B-lined proportional tube provides many of the same proven and beneficial
characteristics as the 3He proportional tube (e.g., a long operating history spanning 70 years, gammaray insensitivity, stability, scalability). The overall efficiency density of the 10B proportional tube is
low in comparison to traditional 3He-based systems, and this limits the application space addressable
by this technology. However, the traditional 10B-lined proportional tube can, in principle, address a
significant fraction of the safeguards neutron counting applications. And when combined with other
safeguards measures (e.g., fast calorimetry instead of multiplicity counting or a change in inspection
procedure) or when used differently (e.g., in active systems), the fraction of solution space can be
expanded.
Research and development activities focused on replacing 3He in safeguards applications tend to
examine a single measurement scenario, without consideration for how the technology under
consideration fits with the overall safeguards needs and capabilities. Yet this should be a fundamental
question in the evaluation of neutron counting technologies. Additionally, developers of new
technologies targeted for safeguards should keep in mind that the traditional COTS 10B-lined
proportional tubes in their current form and without significant effort can address what are many
routine applications and set a minimum performance target. Evaluations should include consideration
of how a new technology addresses both the unchallenging as well as the challenging applications.
REFERENCES
[1] ZHAO, K., et al., International Target Values 2010 for Measurement Uncertainties in
Safeguarding Nuclear Materials, IAEA STR-368, IAEA, Vienna (2010).
7
[2] HENZLOVA, D., et al., Summary of Project on Design and Development of a Safeguards
Coincidence Counter Based on Boron-Lined Proportional Detector Technology - High Level
Neutron Counter Boron (HLNB), Los Alamos National Laboratory Technical Report, LA-UR12-24876, Los Alamos, NM (2012).
[3] LACY, J.L., et al., A Straw Based HLNCC Multiplicity Counter with Improved Figure of Merit
in Same Form Factor, presented at the 55th Annual Meeting of International Nuclear Materials
Management, Atlanta, GA (July 2014).
[4] LACY, J.L., et al., Performance of a Straw-Based Portable Neutron Coincidence/Multiplicity
Counter, 2011 IEEE Nuclear Science Symposium and Medical Imaging Conference (Proc. Conf.
Valencia, Spain, 2011), Curran Associates, Inc., Red Hook, NY (2011) 529–532.
[5] TSORBATZOGLOU, K, and MCKEAG, R.D., Novel and Efficient 10B Lined Tubelet Detector
as a Replacement for 3He Neutron Proportional Counters. 2010 Symposium on Radiation
Measurements and Applications, Univ. of Michigan, Ann Arbor, MI, Paper 0452 (May 2010).
[6] MCKINNEY, K.S., et al., An Investigation of 10B Lined Proportional Counters for Use in
Safeguards Applications, 52nd Annual Meeting of the Institute of Nuclear Materials Management
2011 (Proc. Conf. Palm Desert, CA, 2011), Curran Associates, Inc., Red Hook, NY (2011).
[7] ROGERS, J.L., et al, Uranium Neutron Coincidence Collar Model Utilizing B-10 Lined Tubes,
Technical Report PNNL-21750, Pacific Northwest National Laboratory, Richland, WA (2012).
[8] KOUZES, R.T., et al., Boron-10 ABUNCL Prototype Initial Testing, Technical Report PNNL22147, Pacific Northwest National Laboratory, Richland, WA (2012).
[9] KOUZES, R.T., et al., Boron-10 Active Testing, Technical Report PNNL-22567, Pacific
Northwest National Laboratory, Richland, WA (2013).
[10] MCELROY, R.D., et al., Potential Application of Boron-Lined Proportional Counters for
Safeguards Neutron Coincidence Assay, 54th Annual Meeting of the Institute of Nuclear
Materials Management (INMM 54) (Proc. Conf. Atlanta, GA, 2013), Curran Associates, Inc.,
Red Hook, NY (2013).
[11] MCELROY, R.D., and CROFT, S., The Role of Active Neutron Interrogation to Safeguards NonDestructive Assay (NDA) in the Post 3He World, 53rd Annual Meeting of the Institute of Nuclear
Materials Management 2012 (INMM 53) (Proc. Conf. Orlando, FL, 2012), Curran Associates,
Inc., Red Hook, NY (2012).
[12] MCELROY, R.D., et al., Performance Considerations for Alternatives to 3He-Based Neutron
Counters for Safeguards Applications, 52nd Annual Meeting of the Institute of Nuclear Materials
Management 2011 (INMM 52) (Proc. Conf. Palm Desert, CA, 2011), Curran Associates, Inc.,
Red Hook, NY (2011).
[13] MENLOVE, H.A., Description and Performance Characteristics for the Neutron Coincidence
Collar for the Verification of Reactor Fuel Assemblies, Los Alamos National Laboratory Report
LA-8939-MS, Los Alamos, New Mexico (1981).
[14] CANBERRA INDUSTRIES, JCC-31 Specification Sheet, http://www.canberra.com/products/
waste_safeguard_systems/pdf/JCC-31-SS-C36906.pdf.
[15] RINARD, P.M., Application Guide to Shufflers, Los Alamos National Laboratory Report LA13819-MS, Los Alamos, New Mexico (1991).
[16] CALDWELL, J.T., The Los Alamos Second-Generation System for Passive and Active Neutron
Assays of Drum Size Containers, Los Alamos National Laboratory Report LA-10774-MS, Los
Alamos, New Mexico (1996).
[17] TOUBON, H., et.al., Innovative Nuclear Measurement Techniques Used to Characterized Waste
Produced by Cogema’s New Compaction Facility, Proceedings of the WM’01 Waste
Management Conference, Tucson, AZ, http://www.wmsym.org/archives/2001/62/62-4.pdf
(2001).
[18] MCKINNY, K.S., Neutron Detector Sensitivity Measurements and Their Variation, 2011 IEEE
Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC 2011) (Proc. Conf.
Valencia, Spain, 2011), Curran Associates, Inc., Red Hook, NY (2011) 533–537.
[19] MENLOVE, H.O., et al., WDAS Operation Manual Including the Add-A-Source Function, Los
Alamos National Laboratory Report LA-12292-M, Los Alamos, New Mexico (1992).
8