IOP PUBLISHING
JOURNAL OF PHYSICS: CONDENSED MATTER
J. Phys.: Condens. Matter 22 (2010) 443201 (32pp)
doi:10.1088/0953-8984/22/44/443201
TOPICAL REVIEW
The physics of solid-state neutron detector
materials and geometries
A N Caruso
Department of Physics, 257 Flarsheim Hall, University of Missouri–Kansas City,
5110 Rockhill Road, Kansas City, MO 64110, USA
E-mail: [email protected]
Received 11 June 2010
Published 22 October 2010
Online at stacks.iop.org/JPhysCM/22/443201
Abstract
Detection of neutrons, at high total efficiency, with greater resolution in kinetic energy, time
and/or real-space position, is fundamental to the advance of subfields within nuclear medicine,
high-energy physics, non-proliferation of special nuclear materials, astrophysics, structural
biology and chemistry, magnetism and nuclear energy. Clever indirect-conversion geometries,
interaction/transport calculations and modern processing methods for silicon and gallium
arsenide allow for the realization of moderate- to high-efficiency neutron detectors as a result of
low defect concentrations, tuned reaction product ranges, enhanced effective omnidirectional
cross sections and reduced electron–hole pair recombination from more physically abrupt and
electronically engineered interfaces. Conversely, semiconductors with high neutron cross
sections and unique transduction mechanisms capable of achieving very high total efficiency are
gaining greater recognition despite the relative immaturity of their growth, lithographic
processing and electronic structure understanding. This review focuses on advances and
challenges in charged-particle-based device geometries, materials and associated mechanisms
for direct and indirect transduction of thermal to fast neutrons within the context of application.
Calorimetry- and radioluminescence-based intermediate processes in the solid state are not
included.
(Some figures in this article are in colour only in the electronic version)
4. Summary
Acknowledgments
References
Contents
1. Introduction
1.1. Scope, limitations and goals
1.2. Applied motivation
1.3. Neutron transduction in the solid state
2. General considerations for electron–hole pair generation and separation
2.1. Detector geometry and electric field
2.2. General heterostructure considerations
3. Future solid-state neutron detector devices and applications
3.1. Zero external bias
3.2. Resolving incident neutron kinetic energy
3.3. Concept for a spintronic neutron detector for
enhanced signal-to-noise
0953-8984/10/443201+32$30.00
1
1
2
2
20
21
21
1. Introduction
1.1. Scope, limitations and goals
4
4
8
Advances in materials and device architectures toward passive
or active high-efficiency charged-particle-based solid-state
neutron detection, building from the foundation provided by
Bell [1], Knoll [2], Leroy [3], Lutz [4], McGregor [5, 6],
Petrillo [7], Peurrung [8, 9], Spieler [10] and references
therein are reviewed. Processes or mechanisms that utilize
phonon [11, 12] or photon transduction [1, 13–21] are not
reviewed. The following assumes that the reader has a working
17
17
18
20
1
© 2010 IOP Publishing Ltd Printed in the UK & the USA
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
knowledge of the electronic structure of solids in the bulk
and at heterostructure interfaces including the description of
electronic band structure [22], space charge formation and
electrical carrier transport [23, 24], as well as moderateenergy ion loss processes in the solid state [25, 26]. The
detectors reviewed here are for neutron kinetic energies
related to terrestrial production or applications from 1 meV
to 20 MeV; the mechanisms responsible for interaction at or
above 100 MeV are outside the scope of this review. As
regards temporal resolution, the focus here is toward realtime versus time-integrated detectors (e.g. track detectors,
whose event information is determined sometime much later
and typically outside of a neutron field) [27]. Of the real-time
detection methods, passive detection is primarily reviewed,
although detection of reflected or prompt neutrons from active
interrogation [28, 29] equally applies [30]. Basic knowledge
of neutron interactions and reactions relevant to their detection
is not assumed of the reader. The goal of this work is to
provide a comprehensive summary of advances and challenges
in charged-particle-based solid-state neutron detection such
that the knowledge base from experimental, computational,
theoretical and phenomenological condensed-matter physics,
materials science and nuclear physics communities may
gain greater overlap so as to stimulate and encourage nextgeneration ideas in this emerging interdisciplinary field.
1.3. Neutron transduction in the solid state
In any detection scheme, a transduction of energy must
occur. The general problem is to determine, in real time,
the presence of a (net) neutral fermion at some volume
in space, defined by the physical detector element, whose
incident energy and direction may not be known. Direct
detection of the neutron from its properties alone would require
utilizing an interaction based on its classical translational
momentum, proposed electric dipole or quantum mechanical
spin. For a single neutron, the direct detection of a change
in translational momentum from a scattering event may be
possible [51], but is impractical; for example, the force would
be of the order of tens of piconewtons for full energy and
momentum transfer from a 1 MeV neutron collision. The
detection of a neutron from its predicted electric dipole [52],
or possibly through a weak interaction [53], is at the
threshold of contemporary measurement capabilities [54, 55],
where decades may be required before useful transduction
information becomes available from these interactions; the
fervent search for dark matter will most likely dominate
improvements in the understanding and use of the weak
interaction mechanism in the near term. Determining the
presence of a neutron through a spin–orbit or magneticexchange interaction [56, 57], utilizing a coupling interaction
to the neutron’s spin angular momentum, may be possible
through the Mott [58], Rashba [59], Dresselhaus [60],
Schwinger [61, 62] or spin-echo [63] mechanisms, but
information derived through such interactions is presently
too subtle or too ambiguous to be pragmatic for neutron
detection. However, utilization of the above mechanisms as
alternatives to 3 He- and crystal-based spin-polarized neutron
scattering [64, 65], in rough analogy with spin-polarized
electron scattering [66] with a separate neutron detector, for
low total efficiency spin-resolved neutron detection may still be
possible. Ultimately, because the detection of neutrons through
direct neutron interaction is not pragmatic, we look to primary
or higher-order neutron strong interaction products that can
provide a more sensible form of energy conversion. The
downside to this approach is that the incident kinetic energy
and spatial position of the neutron in question must be inferred,
as the detection mechanism is indirectly representative of the
original neutron four-vector:
1.2. Applied motivation
Neutron detection with x y (i.e. spatial position) resolution
on the order of one to hundreds of microns is important to
the advance of neutron-based nuclear medicine and diffraction
techniques. Specifically, in neutron capture therapy [31], more
accurate spatial neutron dosimetry allows for greater control
over the irradiation of healthy/unhealthy tissues [32–34]. In
elastic neutron diffraction [35], where the physical and/or
magnetic structure of a solid may be determined, tremendous
gains [36] in structural refinement and reduced collection times
(with usually less sample damage) can be made with a twodimensional channel plate detector [37], analogous to the
advances in x-ray diffraction and photoemission techniques
with the discovery and use of the two-dimensional channel
plate with delay-line detectors [38–40].
High-efficiency neutron detection with spatial, temporal
and/or energy resolution is important to the amelioration
of high-energy physics [41, 42], neutron forensics [43],
non-proliferation of special nuclear materials [44], nuclear
energy [45, 46], oil-well logging [47], H2 O/CO2 exploration [48], the search for dark matter [49] and inelastic
neutron scattering [50]. In these applications, cumbersome
and costly post-scattering monochromators, time-of-flight or
low-resolution traditional Bonner spectrometer methods could
be replaced by compact, portable and low-voltage solidstate detectors that operate at room temperature. While the
challenges to realizing detectors with the above capabilities
are daunting [8], success is absolutely physically possible in
the next 20 years with the proper materials, heterostructure
geometries and unfolding procedures.
1
0n
+ xy Z → γ + 00 v f + 10 n + 0−1 e− + ri W+ + sj X− + tk Y.
(1)
Primary reaction products possible from neutron scattering
or capture interactions, as illustrated in equation (1),
include gamma-rays, neutrinos, more neutrons, fast electrons,
moderate-energy heavy and light ions, and neutral atoms.
Thus, how is one to use such products to determine that a
neutron has been captured?
Gamma-rays produced from neutron capture are not
directly detectable, but can be utilized efficiently in higherorder reactions, while neutrinos are even more difficult to
detect than neutrons [67]. The production of more neutrons
through their multiplication [68] and/or average kinetic energy
reduction (i.e. effective moderation) could potentially lead to a
larger effective cross section for capture or scattering, but those
considering the use of such multiplication (n, x n) reactions
2
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
Figure 1. Cross section comparison for Compton and photoelectric
(PE) electron production in gallium arsenide (GaAs), silicon (Si) and
boron carbide (B5 C) as a function of gamma-ray energy. From [569].
Figure 2. Comparison of calculated (a) nonionizing and (b) ionizing
energy loss of moderate-energy ( ) nitrogen ions, ( ) alpha
particles, ( ) protons and ( ) electrons in silicon. From [569].
•
should carefully consider the isotope and neutron kinetic
energy-specific microscopic cross sections (vide infra). Direct
detection of the charge carriers (i.e. fast electrons or moderateenergy ions), even at atto- [69] and sub-atto-ampere [70]
thresholds, is possible under strict laboratory conditions with
proper noise rejection, but is generally unrealistic, especially
in the limit of single-neutron detection from a low incident
neutron flux. However, accumulation of the primary reaction
product charge in a capacitive heterostructure, similar to that
used in transfer neutron radiography [71], is one reasonably
practical means of detecting neutrons with a solid-state
device, despite the loss of time resolution. The neutralatom product, generated by the Wigner effect [72], is not
particularly useful, but the defect or vacancy left behind is;
this momentum-conserving defect mechanism is expanded
upon in the nonionizing energy loss description of moderateenergy ion loss discussed below. Because transduction of
the primary reaction products is only sometimes possible,
and in such cases impractical for the applications under
consideration, we turn to the secondary reaction products
available from interactions of gamma-rays, fast electrons and
moderate-energy ions (i.e. the primary reaction products)
with a solid to provide a more comprehensible means of
energy transduction. This consequently causes information
concerning the original incident neutron to be further diluted,
convoluted and folded [2, 73].
Gamma-ray interaction with matter can also result
in the emission of moderate-energy electron(s) [74] by
either the Compton [75], photoelectric [76] or electron–
positron pair production [77] processes. The cross section
for electron production from photoelectric and Compton
interactions are compared for common solid-state neutron
detector materials, GaAs, Si and B5 C, in figure 1; the
relative cross section for electron–positron pair production
at low gamma energy (<2 MeV) is so small that it is not
included for comparison. Gamma-rays, or more probably xrays, may also cause interband (i.e. core-to-bound) electron
transitions via absorption [78], wherein a next-order process
can result in the emission of a photon [79] or low-energy
Auger electron [80]. Here, only moderate-energy electrons
converted from the gamma-ray primary reaction products of
neutron capture by isotopes such as 72 Ge, 113 Cd, 155 Gd, 157 Gd
or 199 Hg are considered a significant means through which
to cause a sufficient energy transfer (enough generation of
charge) for detection [13–21, 81–83]. However, discussion
of this gamma-ray transduction mechanism is outside the
scope of this review. In general, understanding gammaray energy conversion, especially the time over which the
linear energy transfer occurs in the material of interest, is
extremely important for applications which require rejection
or differentiation of gamma-ray-induced signals from those
produced by moderate-energy charged particles.
The interaction of moderate-energy charged particles (i.e.
electrons and fully or partially ionized atoms with kinetic
energies of the order of 1 keV to 1 MeV) with a solid can result
in ionizing or nonionizing energy loss [84]. The energy loss
profile for four different charged particle types (N+ , α, p, β)
as a function of their kinetic energy in silicon is shown in
figure 2, where ionizing energy loss (IEL) is usually much
greater than nonionizing energy loss (NIEL), and increases
with increasing particle mass. To help illustrate the IEL
and NIEL mechanisms, figure 3 schematically represents a
simplified version of both ionizing and nonionizing energy
loss following neutron capture or scattering from 10 B and
1
H. Table 1 lists the neutron scattering, capture or fissionbased reactions for the isotopes that produce moderate-energy
charged particles that are discussed throughout this review.
NIEL from moderate-energy charged particles can include atomic displacement through momentum conservation [85–91], promotion of a core-level electron to the
conduction band [92], a capture reaction process [93] or a
radiative loss process. The capture reaction of a moderateenergy charged particle can lead to the generation of other
moderate-energy charged particles [94–96] or even neutron
multiplication [97, 98]; both of these mechanisms have rarely
3
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
Figure 4. Three heterostructure electric field configurations through
which e–h hole pairs may be separated: (a) resistive junction; (b) p–n
junction and (c) Schottky junction. E F is the Fermi level, and E C and
E V are the conduction and valence band edges, respectively. The
resistive junction requires an external voltage to establish an internal
electric field within the semi-insulating material, while the p–n and
Schottky junctions have a built-in scalar potential capable of
separating e–h pairs without external voltage. The graded area in (b)
and (c) represents a space charge region created at the p–n and
metal–semiconductor interface.
Figure 3. Reactions and interactions encountered from the capture
and scattering of incident neutrons with 10 B and 1 H in amorphous
hydrogenated boron carbide (a-B5 C:Hx ). All dashed lines produce
electron–hole pairs through ionizing energy loss, while the blue
squiggles represent energy given up to phonons during the e–h pair
creation. With an energy cost of ∼3.5 eV per e–h pair created, the
moderate-energy particles 1 H, 7 Li, 4 He or β (electrons) can produce
from 103 to 105 e–h pairs as they inelastically transfer kinetic energy.
Note that the dual 7 Li and 4 He reaction products depart from the
reaction center antiparallel to each other.
During ionizing energy loss, the Coulombic interaction
of a moderate-energy charged particle with a solid results in
the generation of free electrons and bound electron–hole (e–h)
pairs [130]. At a cost of tens to thousands of electron volts per
electron, electrons are liberated past the vacuum level, where
they can provide a detectable signal as a current through their
replacement from ground or through electron multiplication.
Although straightforward, detection of such low current levels
from electron emission may only be possible under very
high flux (as in total yield x-ray absorption spectroscopy)
and, as such, this mechanism is discounted by the author
for its very low sensitivity. The generation, separation and
collection of e–h pairs from IEL of moderate-energy charged
particles, however, has been and may continue to be the
most efficient means of transducing incident neutrons and is
a main focus of this review. Near-term advances in real-time
passive or active solid-state neutron detection are discussed in
section 2; in analogy with the photovoltaic community, such
advances will depend on improvements to neutron-sensitive
materials and their heterostructure geometries that will lead
to more effective and efficient electron–hole pair creation and
separation with reduced recombination and leakage. Some farterm capabilities are discussed in section 3.
been reported in the design of modern detectors [99] but may
be important in the design of future detector materials and geometries in the context of effective cross-section enhancement
through flux increase and kinetic energy reduction. The coreto-bound electron transition process initiated by a moderateenergy electron [100], as occurs in electron-energy-loss nearedge-structure (ELNES) spectroscopy, is as ill-pragmatic for
primary reaction product transduction as utilizing relatively
low-energy (<10 keV) Auger electrons (vide supra)—although
one recent and unique example of the K-shell Auger electron
resonance for gadolinium (>10 keV) from neutron capture has
been reported [101]. However, the momentum-conserving or
recoil mechanism of a moderate-energy charged particle that
transfers kinetic energy in the form of atomic displacement
or primary knock-on, where a Frenkel defect [102] (as may
also be generated by the Wigner effect) may be formed, is
useful. These defects—or even transient formation of new
interface states—when created in heterostructures which are
very sensitive to voltage, resistivity, carrier lifetime or leakage
current change [103–106], may be used as a poorly efficient but
reasonable means of indicating neutron interaction [107–121],
especially in applications that allow for long integration
time. As the monetary cost of random-access memory is
reduced, use of the memory upset mechanism for passive
time-integrated detection of fast and slow neutrons for area
monitoring may become ubiquitous. Finally, the products
(mostly photons) of bremsstrahlüng [122], synchrotron [123],
Čerenkov [124] and cyclotron [125] radiative loss and higherorder processes [126, 127] from these NIEL pathways are not
considered by the author to be significantly useful towards
detection, although reverse Čerenkov [128, 129] energy loss in
novel metamaterial geometries may be an untapped mechanism
worth investigating. Therefore, by elimination, we are led to
determine whether the energy conversion of moderate-energy
ions through ionizing energy loss provides a more reasonable
means by which to use a charged particle reaction product from
neutron capture to form the basis for a solid-state detector.
2. General considerations for electron–hole pair
generation and separation
2.1. Detector geometry and electric field
All solid-state devices that generate and separate e–h pairs do
so through a heterostructure geometry [131]. In the context
of a solid-state neutron detector, a heterostructure device may
be as simple as a single neutron-sensitive dielectric sandwiched
between two metallic contacts [46, 132–158] or as complicated
as a stack of gadolinium-coated p–n junctions with back
surface field space charge regions. Figure 4 illustrates three
heterostructure classes commonly used in solid-state neutron
4
moderate-energy ion reaction product will be produced relative to the probability for all other moderate-energy ion products for that isotope at an incident neutron kinetic energy of 1 eV
or 1 MeV. Abs. (%) is the absolute probability that a moderate-energy ion reaction product will be produced relative to the probability for all possible neutron-induced products for that
isotope at an incident neutron kinetic energy of 1 eV or 1 MeV. The normalized and absolute probabilities were determined from the reaction-specific cross-section information available
from Chadwick et al [568].
Norm. (%)
Isotope
Helium
Lithium
Reaction
2
1n
0
+ 32 He →
5
1n
0
+ 10
5 B→
+ 28
14 Si
→
1n
0
+ 29
14 Si
(1 eV)
(1 MeV)
(10 MeV)
+ 21 H
—
—
29.6
—
—
4.7
+ 11 H
100
100
70.4
81.7
29.2
11.3
100
100
78.0
97.7
19.0
2.2
4.783
—
—
—
—
—
—
−2.363
—
—
22.0
—
—
0.6
99.9
99.6
37.1
42.3
8.8
3.4
⎧7 ∗
⎪
3 Li (1.472 MeV) + α(0.840 MeV)
⎪
⎪7
⎪
⎨ Li(1.776 MeV) + α(1.013 MeV)
3
9
2
⎪
⎪
4 Be + 1 H
⎪
⎪
⎩ 10
1
4 Be + 1 H
1n
0
+ 30
14 Si →
12 Mg + α
28
1
13 Al + 1 H
26
→
12 Mg + α
29
1
13 Al + 1 H
27
Germanium
Q -value
(MeV)
(10 MeV)
25
1
0n
Nat. Iso. Abd.
(%)
(1 MeV)
⎧3
⎪
H(2.73 MeV) + α(2.05 MeV)
⎪
⎨1
5
2
1
6
n
+
Li
→
0
3
2 He + 1 H
⎪
⎪
⎩6
1
2 He + 1 H
Boron
Silicon
1H
3
1H
Abs. (%)
(1 eV)
12 Mg + α
30
1
13 Al + 1 H
7.59
−3.268
0.764
−2.725
2.312
2.790
—
—-
27.8
—
—
2.6
2.2 × 10−7
0.4
35.1
9.5 × 10−8
3.8 × 10−4
3.2
37.2 (12 MeV)
10.1 (12 MeV)
60.7 (12 MeV)
16.5 (12 MeV)
57.9 (12 MeV)
13.4 (12 MeV)
42.1 (12 MeV)
9.7 (12 MeV)
67.7 (12 MeV)
2.8 (12 MeV)
32.3 (12 MeV)
2.1 (12 MeV)
0.7 (18 MeV)
1.3 × 10−2 (18 MeV)
28.0 (18 MeV)
0.5 (18 MeV)
73.3 (18 MeV)
1.3 (18 MeV)
19.9
−4.361
0.226
−2.654
92.2297
−3.840
−0.034
4.6832
−2.897
−4.200
3.088 72
−7.778
—
20.84
2.963
−0.873
Topical Review
⎧ 0
⎪
e(1.215 MeV) + γ
⎪
⎨ −1
1
70
67
n
+
Ge
→
Zn
+α
0
32
30
⎪
⎪
⎩ 70
1
31 Ga + 1 H
1.4 × 10−4
J. Phys.: Condens. Matter 22 (2010) 443201
Table 1. Examples of isotope-specific neutron capture or fission reactions that generate moderate-energy charged particles. Norm. (%) is the normalized probability that a
Norm. (%)
Isotope
Reaction
(1 eV)
⎧0
e(0.692 MeV) + γ
⎪
⎪
⎨ −1
69
1 n + 72 Ge →
Zn
+α
0
32
30
⎪
⎪
⎩ 72
1
31 Ga + 1 H
70
30 Zn + α
73
1
31 Ga + 1 H
1n
0
+ 73
32 Ge →
1n
0
+ 74
32 Ge →
1
0n
+ 76
32 Ge
1
0n
+ 155
64 Gd →
71
30 Zn + α
74
1
31 Ga + 1 H
73
→
6
Gadolinium
30 Zn + α
76
1
31 Ga + 1 H
152
62 Sm + α
155
1
63 Eu + 1 H
⎧ 0
e(0.079 MeV) + γ
⎪
⎪
⎨ −1
1
157
154
0 n + 64 Gd → ⎪ 62 Sm + α
⎪
⎩ 157
1
63 Eu + 1 H
Uranium
1n
0
+ 158
64 Gd →
1n
0
⎫
⎬
+ 235
92 U
1 n + 238 U⎭
0
92
0
−1 e(0.261
MeV) + γ
(1 MeV)
Abs. (%)
(10 MeV)
(1 eV)
1.1 × 10−2 (18 MeV)
22.5 (18 MeV)
0.2 (18 MeV)
76.5 (18 MeV)
0.8 (18 MeV)
21.3 (18 MeV)
0.2 (18 MeV)
78.7 (18 MeV)
0.9 (18 MeV)
28.2 (18 MeV)
0.2 (18 MeV)
71.8 (18 MeV)
0.5 (18 MeV)
22.1 (18 MeV)
0.1 (18 MeV)
77.9 (18 MeV)
100
(10 MeV)
1.0 (18 MeV)
100
22.9
9.7 × 10−8
6.9 × 10−8
27.54
−0.811
−0.454
36.28
−4.586
−2.157
7.61
2.5 × 10−5
77.1
—
—
8.3 × 10−5
97.9
99.9
88.5
0.3
1.4
3.2 × 10−2
4.2
5.4 × 10−5
2.1 × 10−9
1.5
1.476
3.909
7.73
—
2.1
Q -value
(MeV)
−3.217
—
× 10−7
Nat. Iso. Abd.
(%)
—
0.3 (18 MeV)
1.5
× 10−5
× 10−5
—
—
7.3
—
—
2.6
—
—
—
6.1
1.4
3.3 × 10−2
See figure 10
→ X 1 (Z 1 , A1 ) + X 2 ((Z 2 , A2 ) + x01 n
(1 MeV)
J. Phys.: Condens. Matter 22 (2010) 443201
Table 1. (Continued.)
−6.228
8.339
14.80
−0.530
—
15.65
7.278
−0.581
24.84
—
0.7200
99.2745
199.9
Topical Review
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
detection, defined by the method through which their internal
electric field is established; the merits and difficulties
associated with each class are discussed below in the
context of application. The charge-coupled device (CCD),
although reported in the solid-state neutron detection literature
[144, 159, 160] (not to be associated with scintillation-based
CCD transduction [161, 162]), is not included as its own
class here due to its limitations in radiation hardness, temporal
resolution and ability to detect over a continuous area [163];
however, some of the CCD limitations have recently been
rectified with the charge injection device (CID) which may
be worth further investigation [164]. For the detector class
in which the internal electrical field is produced only by
an external voltage (as shown schematically in figure 4(a))
the dielectric is a direct-conversion material that acts to
both capture neutrons as well as transduce the primary
capture reaction products. These direct-conversion resistive
or dielectric devices (figure 4(a)) are limited in their ability
to suppress leakage current and separate charge generated
by primary reaction products [142], such that the remaining
focus here will be on the p–n junction and Schottky-based
heterostructures (figures 4(b) and (c), respectively), which
self-establish their internal electric field and can benefit over
the dielectric devices (e.g. reduced capacitance, increased
depletion width) by the application of external bias as well.
Within the p–n junction and Schottky-based classes of
solid-state neutron detector heterostructures, two subclasses
are delineated: (1) indirect-conversion (a.k.a. thin-filmcoated or conversion layer) and (2) direct conversion (a.k.a.
solid form). Although both subclasses were first reported
over 50 years ago, the terms ‘thin-film-coated’ and ‘solid
form’ were not coined until recently [165]. Neutron and
primary reaction product transduction for indirect- and directconversion heterostructures are compared in figure 5. For
indirect-conversion heterostructure geometries (figure 5(a)),
a thin film of a neutron-sensitive material is placed within
range so that the scattering or reaction capture product(s) (i.e.
moderate-energy ions) may create e–h pairs in an adjacent
space charge device, where some of the ion energy is inevitably
absorbed or transferred to the conversion material without
creating electronically motive e–h pairs. In contrast, for directconversion heterostructures (figure 5(b)), the neutron-sensitive
material and space charge layer are the same, so that nearly
all of the energy of the charged particle reaction products
may be or is available for transduction. The direct-conversion
heterostructures are consequently the most straightforward to
construct and are capable of the highest total efficiencies.
However, materials in which the majority constituent is
a high-neutron-capture-cross-section isotope that also yields
moderate-energy charged particles upon neutron capture and is
capable of producing, sustaining, and separating e–h pairs (i.e.
direct-conversion semiconductors) are rare and usually poorly
understood in terms of both their processing and fundamental
electronic and electrical carrier transport properties. Therefore,
achieving very high conversion and total efficiency from directconversion solid-state neutron detectors has been and continues
to be a grand challenge. Conversely, indirect-conversion
heterostructures are based on well-understood semiconductors
Figure 5. Representation of neutron transduction from (a)
indirect-conversion (or conversion layer) and (b) direct-conversion
(or solid-form) solid-state p–n junction heterostructures.
and mature semiconductor processing technologies, and can
thus be fabricated with extremely high conversion efficiencies
for secondary reaction capture products [166, 167] as well
as the reproducibility needed for commercial manufacturing.
However, such thin-film-coated devices are innately limited
in their total efficiency because of the decrease in primary
scattering product energy that is converted into e–h pairs
as a result of unavoidable self-absorption processes in the
indirect-conversion heterostructure geometries. Therefore, the
persisting challenge in the competition between the indirectand direct-conversion solid-state neutron detector classes
for highest total efficiency may be presently summarized,
unfortunately, not in terms of pushing the physics frontier (vide
supra) but in terms of the maturity of the materials processing
technologies, in addition to the advances in heterostructure
geometry optimization and understanding of the electronic and
physical properties of the heterostructure materials [168–171].
Ultimately, indirect-conversion detector designers are striving
to optimize the interaction between the conversion and space
charge material regions to obtain effective direct-conversion
devices, while direct-conversion detector designers are focused
on developing and understanding the processing and electronic
properties of the conversion materials towards achieving high
total efficiency given the simplicity of the heterostructure
designs. The next two sections address the general design
considerations and challenges associated with the indirectand direct-conversion heterostructures of the p–n junction and
Schottky heterostructure classes.
Efficiency is used often in this review in the context of
solid-state detectors, and a point of clarification should be
made to remove ambiguity. Conversion efficiency refers to
7
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
2.2. General heterostructure considerations
For both direct and indirect heterostructure types, the
choice of materials and geometry, towards optimization, are
application-specific. Some of the important applicationrelevant characteristics to consider include:
• incident neutron flux and direction (i.e.
parallel,
omnidirectional);
• neutron mean free path or macroscopic cross section as a
function of kinetic energy;
• range and energy of moderate-energy IEL and NIEL
neutron reaction products;
• majority and/or minority carrier diffusion lengths;
• electronic bandgap, work function and band edge
positions of all materials;
• space-charge-induced electric field (direction and magnitude) between each interface.
Figure 6. Comparison of the neutron scattering or capture cross
section (only for the moderate-energy primary reaction
products—from table 1) as a function of incident neutron energy for
1
H ( ), 3 He ( ), 6 Li ( ), 10 B ( ), 28 Si ( ), 157 Gd ( ) and
235
U ( ). Cross-section data from [568].
Advancing both direct- and indirect-conversion heterostructures, in the context of significantly improving spatial,
temporal and/or energy resolution towards applications, requires simultaneously maximizing neutron stopping, e–h pair
production, and e–h pair separation and sweepout, while
suppressing nonsignal carriers and trapping/recombination.
Nonsignal carriers are electrons, holes, polarons or other
charged particles that transport under drift or diffusion and
do not originate from primary or secondary neutron reactions,
but contribute, as leakage current, to part of the total detected
signal. Trapping and recombination limit the total charge
that may be converted to a voltage pulse by the preamplifier;
obtaining a voltage pulse whose magnitude is clearly in
excess of the leakage/noise is paramount. Enhanced neutron
stopping is traditionally accomplished by increasing the
isotopic stoichiometry (enriching), density and/or volume of
high-neutron-capture- or scattering-cross-section centers. For
indirect-conversion devices, enhancement of neutron stopping
using these traditional methods for two-dimensional planar
heterostructures includes the trade-off of a decreased charged
particle product energy loss inside the adjacent space charge
region. For direct-conversion devices, enhanced neutron
stopping is usually commensurate with a decrease in the
number of e–h pairs that can be separated as a result of the
poor or unoptimized electrical transport properties (especially
carrier diffusion length) of conventional high-neutron-capture
cross-section direct-conversion materials (e.g. boron-rich
solids). For ideal and timely e–h pair separation, a large
mobility-carrier-lifetime product (μτ ), a low heterostructure
capacitance and a low concentration of recombination centers
(i.e. traps, deep levels) is sought [2, 23, 24]. For both
the direct- and indirect-conversion devices operated in high
flux or demanding environments [172] radiation hardness is
also required to ensure that leakage current levels remain
lower than the signal—this includes monitoring the leakage
current and e–h pair separation effects from resistivity and
carrier lifetime degradation [173]. General considerations
toward the optimization of neutron stopping as well as signal
and nonsignal carrier generation, separation, collection and/or
blocking for each heterostructure type are discussed below.
the efficiency of separation of e–h pairs after they have been
produced from one or multiple moderate-energy ion(s), while
total efficiency is defined in this work as the quotient of
detected-to-incident neutrons, which encompasses both crosssection-related interactions of the neutron and primary neutron
reaction products as well as e–h pair generation and separation
efficiency. For an incident neutron to be detected, a scattering
or capture reaction event must occur and must lead to the
generation of e–h pairs that are separated and collected with
a total charge-to-voltage pulse conversion distinctly above
the noise level. Because the energy of the incident neutron
dramatically affects the probability that a scattering or capture
reaction event takes place, the incident neutron energy and
probability for a particular reaction (see table 1) should be
qualified when stating the total efficiency. For example, the
total efficiency for a thermal neutron beam incident on a 10 B
thin-film-coated planar p–n junction will be much greater than
that of a reactor spectrum neutron beam incident on the same
device. In the latter case, if a moderator is used to reduce the
incident neutron energy of the reactor neutrons, the attenuation
in the moderator and conversion layers should be accounted for
when reporting the total reactor detection efficiency.
Finally, it is important to note that it is the average number
of e–h pairs that can be generated and separated in the space
charge region per incident neutron, not just the number or
energy of the primary reaction products that cross into the
space charge region, that determines whether enough charge
will be generated and swept out for an event to be counted.
For example, despite the order-of-magnitude gain in thermalneutron-capture cross section for 157 Gd versus 10 B over the
one to hundreds of meV energy range (figure 6), the average
number of e–h pairs generated and separated per incident
neutron is similar for the two because it requires approximately
thirty 79 keV conversion electrons from the gadolinium capture
to equal one ∼1 MeV helium or lithium ion from boron capture
assuming a 2π single-layer geometry (vide infra).
8
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
Figure 7. A schematic overview of indirect-conversion heterostructure geometries where black represents the electrodes, gray the region
responsible for e–h pair separation (from a p–n junction or Schottky heterostructure) and orange or green the neutron-sensitive conversion
material. (a) Conventional, single-sided planar conversion layer; (b) two-conversion-layer two-junction stack; (c) two-sided conversion layer
single junction; (d) single-conversion-layer double-junction sandwich; (e) single-sided porous conversion single junction; (f) single-layer
coating of pillar junctions; (g) single-sided trench conversion single junction; (h) double-sided conversion single junction with conventional
top/bottom contacts; (i) single-sided dual/compound conversion layer single junction; (j) single-sided sine wave trench conversion single
junction; (k) double-sided conversion single junction with trench conformal junction and contacts and (l) three-dimensional sine wave with
conformal junction and contacts. In going from (a) to (l), the semiconductor and conversion layer deposition processing complexity increases
but also allows for greater control over the production and collection of secondary reaction products from neutron capture.
which occurred at the conversion-to-space–charge interface
produced moderate-energy charged particles that resulted in
near-full energy conversion into e–h pairs. Further, given that
for most isotopes the two primary reaction products (table 1)
conserve momentum by their nearly 180◦ ejection (figures 3
and 5) from the capture reaction center, the 2π transduction
geometry restriction allows for only up to half of the primary
reaction products to be used for generating e–h pairs in the
space charge region (figures 5(a) and 7(a)).
Despite the above limitations, optimization of these planar
designs, especially towards specific applications [7, 186–199],
included the use of alternative conversion isotopes or com-
2.2.1.
Indirect-conversion heterostructures.
The first
indirect-conversion heterostructures, reported in the 1950s and
1960s, used a planar design (figure 7(a)) with thick (i.e. one
to tens of microns) conversion films or foils (see the list of
isotopes in table 1) to generate primary reaction products (i.e.
moderate-energy charged particles) [47, 174–182]. Given the
resulting bulk-like interactions for the reaction products and 2π
transduction geometry in the planar configuration, total device
efficiencies were inherently limited, which rendered this class
of devices unpopular despite their relative promise [183–185].
As the mean free path of the capture reaction products was of
the order of the film thickness, only those capture reactions
9
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
to the scale of the carrier diffusion length with limited selfabsorption in the converter, whereupon much higher total
efficiencies were made possible. Specific design improvements
included the systematic [5, 6, 270–272] consideration of the
conversion-material thicknesses, densities and isotope types,
as well as the aspect ratios of the neutron-conversion and space
charge materials, which, through modeling and extensive testing [5–7, 47, 159, 163, 165, 171, 174, 201–203, 224, 226, 239,
245, 270–284], resulted in the geometries shown in figures 7(e)–(h) [143, 266, 270, 285–304]. Ultimately, it was the
clever marriage between high-aspect-ratio heterostructures—
inspired by solid-state gamma-ray. [305] and charged particle
detector advances [137]—with the efficiency increases calculated for multiple p–n junction/converter sandwiches [290]
that led to the idea for [306] and demonstration of [287]
enhancing the interface area between the space charge region
and conversion material in a monolithic heterostructure. That
is, a single heterostructure in which the interfacial area between
the conversion material and space charge region is significantly
increased, allowing for a substantial decrease in the primary
reaction product self-absorption. Such devices are referred to
as grooved or perforated or, more generally, as microstructured
or three-dimensional.
In 2001, McGregor presented the first systematic
calculations [275] of the upper limit on total efficiency
for thermal neutrons from various hybrid heterostructure–
moderator geometries, with results much more conservative
than previous calculations. These geometries ranged from
a single-sided conversion layer (<4.5%—figure 7(a)), to a
double-sided conversion layer (<9%—figure 7(c)), to a single
layer sandwiched between two heterostructures capable of
coincidence (<4%—figure 7(d)), to a single-sided doublecoated (i.e. compound) conversion layer (<12%—figure 7(i)),
to filled holes in a single-sided heterostructure (<10%—
figure 7(e)), and, finally, to filled holes with compound layering
and staggered heterostructure sandwiching (>20%). The
natural progression for the highest total efficiency device,
regardless of the conversion reaction utilized, is that with the
highest interfacial area between space charge and conversion
material up to the threshold at which (1) the space charge
region is of the order of the primary reaction product mean
free path; (2) the carrier diffusion length in the active
semiconductor is exceeded (contact placement is crucial) or
(3) the neutron direction incident on the heterostructure does
not change the efficiency as a result of channeling. Designing
the heterostructure geometry with these points in mind allows
for both the energy loss of the primary reaction products and
their average take-off angle relative to the long axis of the
conversion material to be optimized [279, 281]. It should
be noted that, although stacking of planar heterostructures
increases total device efficiency, attenuation of the primary
reactions products in the bulk of each layer results in
total efficiencies of each heterostructure that do not add
linearly [300] nor as efficiently [5]; this is an explicit and
important point to be considered within the class of threedimensional indirect-conversion devices (i.e.
monolithic
heterostructures and stacks thereof), especially in comparing
those with conversion-material filled holes (figure 3(e)) versus
plexes thereof [177, 200–214], large-bandgap and various
resistivity semiconductors [30, 45, 190, 200, 215–223],
gamma-ray subtraction/discrimination [224–235], reduced
areal dimensions [163, 201, 236–244], heterostructure and heterostructure/moderator stacking [159, 245–258] (figures 7(b)–
(d)), and/or the exploitation of advances in e–h pair sweepout [117, 166, 167, 217, 226, 243, 259–265]. The use
of alternative conversion isotopes and/or their complexes
allowed for the optimization of film thicknesses and the ability
to tune material properties (vide infra) for cross section,
conversion particle type and mean free path. The use of
high-bandgap materials helped to suppress thermally induced
carriers contributing to leakage current, but decreased the total
signal as the energy cost per e–h pair increased [25]. Reducing
areal dimensions enabled pixilation, so that a real-time x y
neutron detector, with pixel sizes of the order of hundreds of
square microns, could be produced that was competitive with
3
He and track detectors. Heterostructure stacking (figures 7(b)
and (d)) led to 4π collection and increased the total opacity to
neutron capture without increasing the average self-absorption
for the primary reaction products; as the transduction of
both primary reaction products (from isotopes that yield dual
moderate-energy ions from table 1) became possible, a nearly
twofold increase in total efficiency was observed. Further,
heterostructure/moderator stacking (figures 7(b) and (d)) allowed for detection of incident neutrons over a large energy
range (i.e. one keV to tens of MeV), in some cases provided
effective collimation [182] and increased the total efficiency for
monoenergetic fast-neutron detection. A special version of the
heterostructure/moderator stacking design was also employed
for neutron spectroscopy (i.e. incident energy and incident
angle analysis) which incorporated transmission detectors for
energy loss rate over a defined distance (d E/dx), total energy
loss and/or scattering angle probability all in one detector [248]; these devices and proposed improvements thereof
are discussed in detail in section 3.2. Gamma-ray or other
background convolution with the neutron signal was effectively
suppressed or removed through coincidence subtraction (also
an advantage of 4π collection) or optimization [229] of the
size of the active space charge region (using the p–n junction,
but especially Schottky geometries) to minimize the separation
and sweepout of gamma-ray-generated e–h pairs (i.e. to
enhance the recombination of gamma-ray-based e–h pairs).
Overall, while total efficiencies for incident thermal neutrons
in the planar geometry were calculated to be as high as
25% [201] (figure 7(a)) or 73% [215, 245] (figure 7(b)),
empirical efficiencies were reported as no greater than 5%.
Spatial resolution of <100 μm was measured [163], although
500 μm was more typical [201, 240] with this generation of
device.
The major advances in the total detection efficiency of
indirect-conversion heterostructures, other than a forwardthinking 1983 patent [266], did not arrive until the mid1990s. Here, a combination of the right tools—progressive
lithographies, novel material-deposition methods, and advanced civilian/international versions of the Monte Carlo
particle transport codes [267–269]—opened a pathway to be
able to tune the capture reaction product scattering length
10
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
trenches (figure 3(g)), despite their comparable interfacial
areas. For comparison, calculations for stacks of 15 planar
devices, with 1.4 or 15 μm thick films of 10 B or 6 LiF,
have yielded total thermal neutron detection efficiencies in
excess of 25% [5], while calculations for the three-dimensional
designs and stacks thereof (figures 7(e)–(j)) have yielded total
efficiencies of >30% [307, 308], >75% [292], >50% [271],
>40% [298] and ∼90% [286]. However, empirical results
for total thermal neutron detection efficiency of the advanced
three-dimensional indirect-conversion heterostructure devices
have been reported as 21% [281] (figure 7(e)), 35% [281]
(figure 7(j)) and 12% [296] (figure 7(j)), with no published
values for the stacked devices. These lower experimental
efficiencies may, in part, be due to the implications of
creating high interfacial areas.
That is, despite the
elegance of monolithic indirect-conversion heterostructures,
the creation of a substantial interfacial area enhances the total
surface/interface recombination and leakage current [296] (as
well as electric field uniformity), which limits the conversion
efficiency (and hence the total efficiency), especially when
the interface consists of sensitive and irreparable crystal faces
(i.e. those with innate dangling bonds and the potential
to create new interface states with the conversion material
or its precoating). Much effort has been put into the
single-junction silicon solar cell industry to reduce surface
recombination loss of minority carriers through the growth
of epitaxial SiO2 with limited metal–semiconductor interface
contact [309]. Optimization of leakage current, contact
placement, space charge direction and interfacial defects
(in the context of wasted or unusable primary reaction
product energy loss) in advanced three-dimensional indirectconversion heterostructures are open challenges whose
solutions will allow empirically determined total thermal
neutron detection efficiencies to approach those predicted
theoretically.
Toward the goal of advancing both spatial and temporal
resolution with the manufacturability maturity needed for
a turnkey detector system—but at the cost of lower
total detection efficiency—planar indirect-conversion detectors
were recently revived [241, 310–315] based on advances
from the medical x-ray detector community (e.g. Medipix
detectors). Spatial resolutions of 50 μm for thermal [311]
and 99 μm for fast [316] neutrons have been achieved, which
provided the basis for neutron tomography of low- Z structures,
not possible with x-ray-based radiographies. The enhanced
spatial and time resolution has allowed for the identification of
unknown incident particles through the unique determination
of the primary reaction capture product energy loss [317].
Further progress in spatial resolution is limited by the stopping
range and/or scattering of the incident neutron, as the spatial
resolution of the primary reaction products can already be
determined within hundreds of nanometers [318]. Thus,
advances in spatial resolution will require a clever means of
corralling the neutron normal to the surface point of entry
into the conversion layer. Further enhancements will aim
to cut channels and load conversion material (figure 7(g))
into the medical x-ray detector analogs to marry the high
total efficiency, spatial and time resolution, and radiation
hardness [319] attributes.
It is also important to contrast the above detectors with the
recent microchannel plate (MCP)-based indirect-conversion
devices [320–327]. In the MCP devices, a glass plate with
closely spaced pores is loaded with neutron-sensitive isotopes
from which primary and secondary reaction products may
be generated [328]. The reaction products lead to electron
multiplication/avalanche in the microchannel pores which are
then subsequently detected; the final two steps of this process
(i.e. electron multiplication and detection) are beyond the
scope of this review. However, here again, benefiting from
the advances of the x-ray and charged particle detection
communities, the MCP-based neutron detectors are poised
to provide excellent two-dimensional spatial (<15 μm) and
temporal (1 μs) [324, 325] or one-dimensional temporal
(50 ns) [326] resolution on a platform that can theoretically
compete in total thermal neutron detection efficiency (from 2–
12% [323] to 50% [326] calculated) with the planar and threedimensional detector classes described above. Unfortunately,
the direct incident neutron energy information is made nearly
undecipherable by the charge multiplication process, but with
such excellent timing resolution, time-of-flight studies in the
proper geometry (vide infra) could be possible [326]. The high
spatial resolution comes from the intrinsic collimation of the
microchannels (i.e. their high aspect ratio of 250:1) [322, 324]
and the timing resolution from advances in delay-line readout.
The most recent experimental verifications of the above
calculated values have shown total efficiencies of 43% for
cold and 16% for thermal neutrons with <15 μm spatial and
<5 μs temporal resolution [327]. While the plates of the
MCP-based device presently provide an advantage in spatial
resolution and the electronics capability can yield resolutions
of the order of tens of picoseconds, the electron multiplication
step may ultimately limit the temporal resolution. Once the
lithographic techniques capable of embedding boron/lithium
into the medical x-ray (i.e. Medipix) or analogous detectors
are mature, it will be difficult for the MCP-based devices to
compete in time, efficiency and spatial resolution—that is,
unless hybrid device concepts which integrate the MCP and
p–n junction are realized [329].
The next generation of indirect-neutron-detection devices
will likely be developed by four groups/communities: (1)
those with the interdisciplinary skills to mesh the highefficiency three-dimensional devices with excellent spatial
resolution and fast readout electronics; (2) those with the
materials science know-how to fabricate heterostructures with
geometries yielding even higher surface areas; (3) those who
will exploit the advances from group 1 toward new applications
or toward the discovery of new neutron-based interactions and
(4) those with the insight to determine and rectify deficiencies
in the nuclear and condensed-matter physics of neutron and
primary reaction product transduction. The first and second
groups, while making incremental advances, are important
to the realization of commercial products that will benefit
research and general dosimetry end users in the next ten
years. The second group will likely strive to design detectors
that can simultaneously provide information regarding all
degrees of freedom of the incident neutron. The third and
fourth groups, however, are those cutting the road; a recent
11
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
example is the work by McGregor and co-workers on advanced
indirect-conversion neutron detector straight and sine wave
trenches [282, 296, 299] with conformal or three-dimensional
space charge configurations (figure 7(k)). Here, deficiencies
in the macroscopic interaction of neutrons with the conversion
material (i.e. channeling) and the uniformity of the space
charge region were identified. The channeling problem (in
two dimensions) was alleviated by etching two-dimensional
trenches in the form of sine waves or chevrons into silicon
p–n junctions (figure 7(j)), while the poor space charge region
uniformity was amended by diffusing dopant after the trenches
were etched to create a three-dimensional space charge
region rather than cutting into the traditional predoped planar
(i.e. two-dimensional) space charge monolith (figure 7(h)
versus figure 7(k)). Lithographically cutting or etching a
sine wave or other similar geometry that changes in three
dimensions (figure 7(l)), where neutrons incident from any
direction yield the same total efficiency, utilizing preferential
crystal etching, decreasing interface/surface recombination
and selectively depositing individual contacts within those
three dimensions would take this class of devices a step
further. Finally, despite the steady and marked advances
in indirect-conversion heterostructure design, it is unlikely
that any of the above four groups will stray from the use
of conventional bulk conversion and space charge materials
in advanced detector design. Section 2.2.2 departs from
the discussion of clever geometries, focusing rather on the
development and improvement of semiconductor materials
whose majority constituent is a neutron capture or scattering
center that is capable of producing moderate-energy charged
particle products. The class of devices integrating such
neutron-sensitive semiconductors may be effectively likened
to indirect-conversion heterostructures with infinite interfacial
area and zero self-absorption in the conversion material.
Figure 8. Neutron mean free path as a function of incident neutron
energy for (C2 H2 )n ( ), Li2 B6 ( ), B5 C ( ), Gd2 O3 ( ) and
UO2 ( ). Total cross sections to account for opacity were used in
the calculations here and 100% enrichment of 1 H, 6 Li, 10 B, 157 Gd and
235
U to generate a best case scenario. Cross-section data from [568].
Because self-absorption (i.e.
unconverted primary
reaction product energy) is not a limitation for directconversion heterostructures (see figure 5(b)), their geometry
can be limited to that of a simple planar junction. Advanced
applications for spatial, temporal or energy resolution of
incident neutrons may require the adjustment of contact
placement and diode stacking, but the planar design persists,
with transistors and other unique space charge directconversion heterostructures being exceptions to this type
of device development at present. This does not mean
that advanced active region direct-conversion heterostructures,
such as those with a back surface field geometry, are
less efficient but that they are simply too complex given
the deficiencies in our understanding of the known highmacroscopic-cross-section solids. The thickness of each planar
component in the heterostructure is designed primarily as a
function of the desired total stopping power (a function of
the neutron mean free path), carrier diffusion length and, to a
lesser extent, the junction capacitance. For most applications,
fully stopping the neutron is preferred, with differential energy
analysis applications and high flux beam monitors being
exceptions. For heterostructures based on low-macroscopiccross-section constituents, this means the planar layers must be
thick (i.e. one to tens of cm in thickness), which immediately
compromises both the leakage current and carrier diffusion
length. Therefore, the ideal direct-conversion heterostructure
is composed of materials that have both a large microand macroscopic cross section (to bring the semiconductor
thickness to the order of tens of microns for thermal neutrons;
figure 8) and a long carrier diffusion length (i.e. of the
order of tens to hundreds of microns) up to the limit, ideally,
at which the thickness is so low that junction capacitance
becomes the limiting factor. To some extent, leakage current
and capacitance can be controlled by tuning the heterostructure
series resistance and applied bias as long as requirements
pertaining to the desired stopping distance and carrier diffusion
length (assuming zero external bias) are first met. It is in this
context that the inability to simultaneously achieve both a short
2.2.2. Direct-conversion heterostructures. Solid-state directconversion neutron detection heterostructures are devices
whose space charge generating material also generates primary
reaction products from neutron capture or scatter. This space
charge material may be one or both sides of a p–n junction
(or the base layer of a conventional diffused junction) or
the sole semiconductor in a Schottky device. There are
three subclasses of direct-conversion heterostructures: (1)
those based on mature semiconductors with low microscopic
neutron cross sections (i.e. low-macroscopic-cross-section
heterostructures); (2) those based on mature semiconductors
that incorporate high-microscopic-neutron-cross-section constituents at low concentrations (i.e. low-macroscopic-crosssection heterostructures) and (3) those based on immature
semiconductors with high-microscopic-neutron-cross-section
constituents dominating the stoichiometry (i.e. large macroscopic cross-section heterostructures). Microscopic cross
section speaks here to the incident-neutron-energy-dependent
capture, scattering or fission cross section for moderate-energy
primary reaction product generation, whereas macroscopic
cross section refers to the incident-neutron-energy-dependent
stopping power or mean free path (see figure 8). Mature
semiconductors are taken as silicon or germanium.
12
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
neutron mean free path and long carrier diffusion length has
plagued the success and popularity of the direct-conversion
detector class and led to the evolution of several subclasses.
Each subclass presents its own advantages and challenges,
commensurate with the character of the communities that
champion the material’s growth, unique heterostructure design
and/or nuclear physics associated with that subclass. The
subclass represented by mature low micro- and macroscopic
cross-section semiconductors was first introduced in the early
1960s [179, 330–336] following a string of successes for
surface-barrier and p–n-junction-based solid-state charged
particle detectors [337–339]. Neutron capture reactions for
silicon- and germanium-based direct-conversion solid-state
neutron detectors are shown in table 1, with energy-dependent
cross sections represented in figure 9. During this early
period, special emphasis for the silicon devices was placed
on mapping the incident-neutron-energy-dependent isotopic
cross sections for the (n, α) and (n, p) reactions. It was
subsequently determined that energy analysis of the recoiling
nucleus (e.g. the 25 Mg from the 28 Si(n, α)25 Mg), in addition
to the energy determination of the alpha particle or proton,
could be used to back out the incident neutron energy;
on this basis, a spectrometer with low total efficiency but
relatively high resolution (of the order of tens of keV) could
be fabricated (see section 3.2 for more on energy analysis
using this technique). Building on the high-energy threshold
(>5 MeV for 28 Si, >3 MeV for 29 Si and >8 MeV for
30
Si) and resonant cross-section structure, applications for
determining fusion plasma ion temperatures [340] or 14 MeV
neutron flux from tokamak triton burnup [341, 342], to fastneutron dosimetry [247, 343], to total fluence determination
for colliders [310] were studied. Although there has been
recent work utilizing the alpha particle and proton products
from silicon neutron capture, such studies are tangential to
the intended application (e.g. gamma-ray or muon detection)
of the silicon-based detectors. Overall, silicon-based directconversion heterostructures may be of interest for future
development, not only because of their excellent electrical
transport properties and device integration, but also their
comparable cross section to 1 H, 3 He, 6 Li and 10 B (157 Gd is
an exception) for moderate-energy charged particle generation
at 1 MeV incident neutron energy (figure 6). That is, the
use of high-macroscopic-cross-section solids is usually only
advantageous for thermal neutrons. Upon consideration of the
cost of attenuation (i.e. intensity reduction) and delocalization
(in energy and real space) incurred during the process of
neutron thermalization (i.e. moderation), combined with the
unoptimized electronic properties of high-macroscopic-crosssection solids, the pursuit of silicon detectors may prove
worthwhile.
The second subclass of active materials in directconversion heterostructures, exhibiting overall low microscopic cross sections, are those solids that incorporate highmicroscopic-cross-section constituents at dilute concentrations. In some cases, constituent incorporation into the
mature bulk semiconductor is a straightforward isoelectronic
replacement, as in p-doped silicon with isotopically enriched
10
B [344], while in other cases the constituent can be much
Figure 9. Neutron capture cross section for the (n, α) (– – –) and
(n, p) (——) reactions for (a) 28 Si (black); 29 Si (blue); 30 Si (red) and
(b) 70 Ge (black); 72 Ge (blue); 73 Ge (red); 74 Ge (green) and 76 Ge
(violet), where the isotopic mass increase is proportional to the
energy threshold for both Si and Ge. Cross-section data from [568].
more (e.g. complexed multivalent uranium [345]) or less (e.g.
quiescent helium [346–348]) chemically active than a conventional electronic dopant. Of the handful of reported work
on such semiconductors, pulse height signal-to-noise ratios of
only 2:1 for moderate-energy neutron capture products have
been demonstrated, despite doping concentrations as high as
1021 cm−3 ; lithium-drifted silicon [349] and germanium [16]
solid-state neutron detectors also fit in this category. While
doping mature semiconductors with high-microscopic-crosssection isotopes as minority constituents is convenient, this
strategy has no straightforward fundamental advantage over
the other direct-conversion detector subclasses. If, however,
the detectors are specifically intended to be monetarily
inexpensive, utilizing dopants at natural isotopic abundance
with standard semiconductor processes, an advantage for niche
dosimetric applications with very high flux (e.g. beam-position
monitors) at thermal neutron energies may be feasible.
The third subclass comprises heterostructures in which
the active semiconductor stoichiometry is built from at
least one isotope with a high neutron scattering or capture
microscopic cross section (i.e. high-macroscopic-cross-section
heterostructures) that also generates moderate-energy charged
particle products. Of the isotopes that meet these criteria, only
hydrogen, lithium, boron, gadolinium and uranium have been
proven to be reasonable majority constituents in compound
semiconductors, and of these compound semiconductors,
only a handful have shown the most basic electrical,
electronic, structural and chemical properties, combined with
the processability needed to fabricate a direct-conversion
neutron detector heterostructure.
13
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
In the case of hydrogen (and, to some extent, materials
constituted from 2 H, 12 C and 16 O), the hydrogen-rich organicbased conducting polymer polyacetylene (C2 H2 )n has been
postulated as being useful to detect neutrons in the Schottky
heterostructure [266, 350, 351], although this has not been
reduced in practice [352]. Given the tremendous recent
advances in organic light-emitting diodes [353] and organic
photovoltaics [354], especially towards enhancing electrical
carrier mobility (e.g.
from hundredths to hundreds of
cm2 (V s)−1 ) for very thick conducting polymer films [355],
and since this work was reported in 1983–85, the use of
hydrogen-rich conducting polymers as active layers for directconversion neutron detection heterostructures could be an
emerging near-term area of study (however, see the neutron
mean free path comparison in figure 8 before venturing in
this direction). Further, hybrid hydrogen-rich semiconductors
that incorporate high-loading-ratio gadolinium-chalcogenide
or uranium-based nanoparticles (borrowing from that learned
in scintillation [356]), or boron- or lithium-fluoride counterions
(for the oxidatively doped conducting polymers) could also
be of interest for increasing not only the macroscopic cross
section, but also for extending the useful energy range
for high-probability neutron scattering or capture. One
could also consider developing bulk hydrogen-rich organicbased semiconducting heterostructures coupled to the more
standard inorganic direct-conversion or indirect-conversion
heterostructures that would act as active moderators which
reduce neutron energy and extract e–h pairs rather than
the passive moderators that only act to reduce neutron
energy and flux (i.e. attenuate) such as polyethylene or
analogous moderators that have canonically been used. Such
a configuration could be used to increase the total efficiency
for fast neutrons and provide more information for incident
neutron energy analysis. It should be noted that even at
atomic concentrations of up to 20% for hydrogen in amorphous
hydrogenated silicon (a-Si:H) films of tens of microns in
thickness, no appreciable or resolvable signal from proton
recoil was found [201, 215, 245], which emphasizes that
very thick film-based heterostructures are needed to effectively
utilize proton recoil in hydrogen-rich organic or inorganic
semiconducting materials; this is further supported by the
energy-dependent mean free path for (C2 H2 )n in figure 8.
In the case of lithium, the author is not aware of any
reported p–n-junction- or Schottky-based direct-conversion
neutron detection heterostructure work. However, advances
in blue-light spontaneous- and stimulated-emission diodes,
progress in Li-ion battery anodes and the study of simple
superconductors (e.g. MgB2 analogs) have motivated the
prediction and subsequent fabrication of a number of lithiumrich solids, many of which are p-type semiconductors that
may be useful towards fabricating active thin films in directconversion neutron detection heterostructures. Li12 Si7 has an
energy bandgap (E g ) of 0.6 eV and an electrical resistivity
(ρ) of 106 cm [357–359]; the isostructural germanium
analog (Li12 Ge7 ) as well as LiSi ( E g = 0.06 eV) likely
have too small a bandgap to be considered useful at room
temperature [360]. A number of intermetallic complexes,
including LiInSe2 [361], LiInS2 [362], LiZnAs [363, 364],
LiCdP [363], LiZnP [363], Li8 MgSi6 [365] and variants
thereof [366], yield bandgaps that range from 0.7 to 2.1 eV
and resistivities from 10−1 to 1010 cm, with one report of
hole mobilities up to 30 cm2 V−1 s−1 at room temperature.
Despite initial predictions of an LiB semiconductor [367], both
α -LiB and β -LiB have been shown to have a strong density
of states at the Fermi edge [368, 369], while Li2 B6 [370]
and Li3 B14 [371] exhibit bandgaps of ∼2 eV, with a host
of other stoichiometries and structures that have not been
explored electronically [372, 373]. Some predicted challenges
in using most lithium semiconductors as active layers for
neutron detection include hydrolyzation from direct exposure
to atmosphere, electromigration of Li ions under applied bias in
the detector geometry, high density of recombination centers,
relatively poor carrier mobility and very low work function.
For the intermetallic complexes, inclusion of indium (113 In,
115
In) or cadmium (113 Cd) may cause unwanted background
in a neutron field and should be used cautiously. Further, any
ferroelectric, piezoelectric or comparably electrically polarized
lithium complex (e.g. LiNbO3 , Li2 B4 O7 ), despite having
bandgaps >3 eV, should be carefully considered, as an
enhanced capacitance will limit e–h pair sweepout. One major
attractive feature of Li is its very low cross section for gamma
interaction; however, with a natural abundance of 8% for 6 Li
and no established thin-film processing reported, it is likely
that the lithium-rich semiconducting complexes will remain
dormant in direct-conversion solid-state neutron detection.
Boron-rich solids are, by far, the most studied, applied,
controversial, varied and least understood semiconductor
materials in the context of solid-state direct-conversion neutron
detector heterostructures. Starting from 1952, patents filed
by Welker et al [374–376] (Siemens Corp.) outline the use
of bulk boron-pnictide crystals, and neutron subreactions with
group V isotopes, in the formation of p–n junctions for the
detection of thermal neutrons. It was immediately realized that
bulk elemental boron crystals, although excellent for stopping
neutrons, could not be grown or post-processed thin enough
to efficiently extract a charge pulse in the heterostructure
geometry, due in part to both poor electrical carrier mobility
and extremely high resistivity. Welker found that the addition
of nitrogen, phosphorus, arsenic or antimony allowed for
control over the bandgap (with the gap decreasing in that
order) and an enhancement in the carrier mobility; the addition
of group II-B dopants (Zn, Cd or Hg) was said to shift the
majority carrier to an acceptor while addition of group VIA dopants (S, Se or Te) was postulated to shift the balance
to a donor-type boron-rich solid. By 1963, Hill [133]
(Monsanto Chem. Co.) claimed to have refined Welker’s work,
finding that hexaboron phosphide (B6 P) exhibited superior
electronic properties to the other boron-pnictides and within
the boron-to-phosphorus stoichiometry for direct-conversion
neutron detection devices, as well as controlled growth of
thinner 500 μm thick crystals, ohmic contacts, and dopants for
surface-diffusion-based p–n junctions; however, actual neutron
detection spectra had yet to be presented. Hill did posit [133],
and later demonstrated [377], that group VI-B dopants (Cr,
Mo and W) would drive the boron-rich phosphides to a bulk
n-type material, a tremendous insight that would only begin
14
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
with contacts that exhibit relatively low work functions (Ti,
Cr, Al) [139–141, 143, 150, 151, 424–434]. For multilayer
contacts, where a low-work-function contact is capped by a
high-work-function contact to prevent oxidation, such as in
Cr/Au, one must ensure that the high-work-function metal
does not diffuse through a thin (one to tens of nanometers)
low-work-function contact. Further, the resistivity of the ntype silicon base substrate onto which the p-B5 C is deposited
(or p-type silicon in the case of doped (e.g. nickel) B5 C)
is also an important aspect to consider in understanding
e–h pair separation in the context of the pulse height
results [276, 390, 391, 419, 422, 423, 435], especially with less
than 50 V reverse bias applied. In returning to the sum peaks,
an important outcome of looking at the time evolution of the
e–h pair sweepout, generated by the B5 C/n -Si heterostructure
neutron capture primary reaction products, was that the charge
pulse rise time through the boron carbide semiconductor (i.e.
hundreds of nanoseconds to one microsecond) was found to
be very slow compared with that from silicon (i.e. one
to tens of nanoseconds). Thus, for the all-boron-carbide
homojunctions, one time constant setting is then sufficient to
sweepout and generate sum peaks in the pulse height spectra,
as were observed, whereas in the B5 C/n -Si heterojunction,
the two very different charge-producing pulse time windows
may only be observed from a long charge- or voltagetime (i.e. tens of microseconds) acquisition (e.g. by an
oscilloscope) as shown by Day [419, 422], Harken [420]
and more recently Hong [435]. From a charge-to-voltage
conversion point of view, the direct-conversion heterojunction
device is then not as effective in generating charge (i.e.
the voltage pulse is not as high) as the direct-conversion
homojunction when reducing the pulse collection time window
to match only one of the semiconductor layer sweepout
properties, but it is still more effective than indirect-conversion
heterostructures with reduced charge generated due to selfabsorption of both moderate-energy ion primary reaction
products. Total thermal neutron efficiencies normalized to
the film thickness for the B5 C/n -Si heterostructure have been
shown to be greater than 80%, but such normalization is
misleading as the electrical transport and junction properties
associated with e–h pair sweepout may not scale linearly
with B5 C film thickness, especially as one tries to enhance
the total neutrons stopped by increasing the film thickness
to, for example, 3 or 10 μm or greater. That is, the
total un-normalized empirically determined thermal neutron
efficiencies from the boron-rich class of direct-conversion
heterojunction and homojunction heterostructures have yet
to compete with the values regularly met with the indirectconversion heterostructures of >10% (vide supra), most
probably because of the high concentration of both holes and
electrons, as well as trap and/or recombination centers, leading
to low mobilities [414, 436–440] as found for similar boronrich solids. A recent study by Adenwalla et al [435] supports
this point with 1.8 μm thick B5 C on n -Si heterostructures
with <1% thermal neutron detection efficiency, compared with
the 0.3% thermal detection efficiency for a 232 nm thick
B5 C four years earlier [419, 422]. Ultimately, such low
total detection efficiencies for the semiconducting boron-based
to be understood many years later [378]. In 1963 and 1965,
portions of patents by Henderson [379] (Monsanto Chem.
Co.) and Bloom [380] (Martin Marietta Corp.) outlined the
use of thermal-based chemical vapor deposition (CVD) from
alkyl boron or boron trihalides (e.g. (CH3 CH2 )3 B or BBr3 )
to deposit boron-rich thin films as a means of overcoming the
bulk crystal electrical transport limitations. Although all of
the tools and know-how were available to fabricate a thinfilm-based p–n junction, only a neutron-sensitive thermistor
was reported [381] by Bloom during this period. Although
thermal-based CVD of boron-rich films continued to be
refined [382–385], especially towards mechanical applications,
it was not until the late 1980s that novel low-substratetemperature thin-film-deposition techniques, including pulsedlaser deposition, radio-frequency magnetron sputtering, laserassisted CVD, hot-wire CVD and plasma-enhanced CVD
(PECVD), were shown to provide the control needed (at
∼300 ◦ C) to fabricate and tune electronic-grade materials with
abrupt interfaces. Such deposition techniques are in strong
contrast to the impurity diffusion and segregation (especially
of carbon) problems (even for crucible-free single crystals)
incurred with thermal-based CVD films grown at >800 ◦ C
(see [386–389] and references therein for a complete history).
From 2002 to 2005, Robertson, Adenwalla and coworkers [276, 390–393], building on the PECVD and
magnetron sputtering deposition route for electronic-grade
thin-film boron-rich solids, developed since 1988 by Dowben
and co-workers [394–416], reported the first direct-conversion
pulse height spectra from a 276 nm B5 C film deposited
by PECVD from orthocarborane (C2 B10 H12 ) onto n-type
silicon (30 cm) with Cr (10 nm)/Au (40 nm) contacts.
Although it was clear that the pulse height character was
derived from the individual moderate-energy ion primary
reaction products from boron neutron capture, the rectification
character of the reported junction and expected reaction
product superposition energies were nontraditional or missing,
and therefore the validity of the B5 C layer being an active
semiconductor (i.e.
that the pulse height spectra were
derived from a genuine direct-conversion heterostructure) was
immediately challenged and debated [168–170]. Specifically,
McGregor and others pointed out the potential for mistaken
interfacial boron doping of the underlying n-type silicon
and questioned the absence of the sum peaks (at 2.3 and
2.8 MeV) that should be observable in direct-conversion
heterostructures where oppositely recoiling primary reaction
products create e–h pairs in both the n- and p-type layers
of a direct-conversion p–n junction. Although Dowben
had shown that rectifying junctions with boron-only-based
heterostructures (i.e. homojunctions) were possible [407, 417],
hence demonstrating that B5 C is an active semiconductor, it
was not until 2006 that the primary reaction product sum
peaks from neutron capture were first weakly observed and
rigorously modeled in both the all-boron-carbide homojunction
and B5 C/n-Si heterojunction heterostructures [171, 418–423].
It is important to note, however, that the interfacial metal
work function to p-type boron-rich solids must be considered
carefully, as Schottky barrier formation is known to occur with
high-work-function contacts (Au, Pd, Pt) and ohmic behavior
15
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
(GdB4 , GdB6 ), Gd silicides (GdSi2 ), Gd chalcogenides
(Gd2 O3 , Gd2 S3 ), Gd monopnictides (GdN, GdP, GdAs, GdSb,
GdBi) and other complexes (Gd5 Si2 B6 ) have been fabricated
and studied extensively. Unfortunately for solid-state neutron
detection, a nonzero density of states at or near the Fermi
edge and one to hundreds of μ cm resistivity at 300 K
have been observed for all of the above complexes [457–462]
except for Gd2 O3 and special strained versions of Gd
monopnictides [463]. Gd2 O3 has a bandgap greater than 5.5 eV
and a resistivity greater than 106 cm [464–467] at 300 K,
and thin-film deposition of this material by electron-beam
evaporation is now commonplace [456, 464–466]. However,
little is known regarding the electrical carrier mobility [468]
of Gd2 O3 or how to dope and control the majority carrier;
transition-metal doping to create Gd2 O3 suboxides has been
reported [469] and, despite slightly reducing the bandgap
and resistivity, may be required to produce the mobility
needed for efficient e–h pair separation. From a very
different point of view, Gd hydrides (GdH3−x for 0 <
x < 3) have recently become popular for spintronic and
optical switching applications with demonstrated bandgaps
in excess of 2.5 eV [470–472] and enhanced bandgaps with
Mg doping [473]. These hydrogen-rich gadolinium hydride
semiconducting phases, if grown as thick films, would create
an interesting total efficiency profile over both the fast and
thermal neutron energies.
With an average total fission fragment energy of
>165 MeV from the neutron capture of 235 U or 238 U [474],
coupled with a high fast-neutron fission cross section (at
least one order higher for 235 U than for 10 B for >600 keV
neutrons—see figure 6) and uranium (233 U, 235 U and, to
a lesser extent, 238 U) compound semiconductors should be
seriously considered for active semiconductor layers in directconversion heterostructures as well as uranium-based films
for indirect-conversion heterostructure devices. Specifically,
with a two or four order-of-magnitude increase in e–h
pair production from 235 U primary reaction products over
10
B or 157 Gd primary reaction products, the lower level
discriminator threshold can be set much higher, significantly
changing the landscape for background suppression; this
is a paradigm-shifting mechanism!
Further, because
the spontaneous fission (α, n) and/or delayed neutron
production from uranium radioisotopes per kg per second
are still much less than the cosmic-ray-induced spallation
background, even for uranium oxides (which can be further
improved by isotopically enriching with 16 O), single-neutron
detection of external impinging neutrons is still possible.
Reported semiconductors containing uranium include the
ternary intermetallics U3 T3 Sb4 , where T = Ni, Pd and
Pt [475, 476], the uranium–antimony oxide phases USb3 O10
and USbO5 [477], and the uranium sulfides and oxides
US2 , UO2 , UO3 , U4 O9 and U3 O8 [478–507]. The ternary
intermetallics have bandgaps of up to 0.2 eV and resistivities
of less than 0.15 cm at 300 K, while the uranium–antimony
oxides exhibit larger gaps ranging from 0.5 to 1.9 eV, with
substantially higher room temperature resistivities of 107 –
109 cm. The uranium oxides cover the middle ground
with E g ∼ 1.3 eV and ρ ∼ 103 cm. Given the large
heterostructures are temporary, rooted in the large gap in our
understanding of the physical, electronic, electrical transport
and chemical properties of the thin films produced, as well
as the differences reported between groups and their handling
procedures, especially considering the extreme oxygen and
moisture sensitivity of the films [441]. Recent work toward
systematically rectifying deficiencies in the understanding
of these basic properties of these boron-rich carbides is
occurring [442–444], but the determination and reproduction
of more basic properties among the many groups, including
the carrier concentration and mobility, carrier diffusion length,
work function and lifetime as a function of stoichiometry, and
pre- and post-growth treatment are needed if the normalized
>80% total neutron detection efficiencies are to scale with
>5 μm thick B5 C or ultimately >13 μm B5 C homojunctions.
Other boron-rich solids that may be promising semiconducting candidates for direct-conversion neutron detection
heterostructures, as per initial results, include electron-beamdeposited aluminum dodecaboride (AlB12 ) [445] and pulsedlaser deposited or sputtered AlMgB14 [446].
Uses for
these materials include homojunction, heterojunction and
Schottky barrier formation. Further, it would appear to be
a straightforward embodiment to apply developed recipes
for Schottky barrier formation to the boron-phosphides or
analogous boron-rich solids to these aluminum borides to
fabricate structures with low leakage current and capacitance
to measure direct-conversion neutron capture pulses.
Although a number of gadolinium-based semiconductors
can also be used as active layers for direct-conversion
neutron detector heterostructures, caveats regarding energy
conversion of the primary reaction products, in light of
potential applications as well as the romanticized 157 Gd cross
section, should first be reviewed. Gd neutron capture can
result in a number of gamma-ray and moderate-energy electron
primary reaction products [447], including those of interest
for creating e–h pairs with internal conversion electrons at
energies of 79 and 182 keV [101, 448]. Assuming a cost of
4 eV per e–h pair [25] and that the 79 keV conversion electron
is doing most of the work, a maximum of 3 femtocoulombs
(fC) of separable e–h pairs will be generated per conversion
electron (four times less charge will be generated for undoped
Gd2 O3 at a cost of 15.5 eV/e–h pair). In the single-neutron
detection limit, 3 fC approaches the intrinsic noise level of
many commercially available charge-to-voltage preamplifiers
and is almost two orders less than the charge available from the
primary reaction product transduction from neutron capture of
3
He, 6 Li or 10 B, despite their much smaller microscopic cross
section relative to 157 Gd (figure 6). However, if the incident
neutron flux is high enough and the preamplifier gain can be
optimized for these low charge production levels, gadoliniumbased semiconductors are worth serious consideration as active
layers for direct-conversion neutron detection heterostructures.
As a result of electronic and magnetic application
requirements from technology roadmap consortia [449, 450],
solids with enhanced refractory properties, deposited at
low temperatures, such as the rare-earth borides, silicides,
chalcogenides and pnictides, have been explored over the
last 40 years [451–456]. From these studies, Gd borides
16
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
the highly energetic and massive daughter fission fragments
is also expected to be significant, but a worthy trade-off
for many applications, especially as the critical elements in
detectors are more easily replaced physically and monetarily.
Lastly, although possible, the cross section for neutron
multiplication (n, x n) from 235 U (where x = 2 on average
and the secondary neutrons have a reduced energy for effective
moderation and total efficiency increase in a heterostructure
device) is quite small until significant neutron fluxes above
14 MeV are used, such that this property may only fulfill a
special niche. Thus, uranium-based compound semiconductors
are best considered as active layers in solid-state directconversion neutron detection heterostructures where either
high noise/background exists, high-temperature operation or
radiation hardness to high flux moderate-energy backgrounds
is required, or the neutron energy is in excess of 600 keV and
moderator real estate is limited.
Figure 10. Mass distribution of the neutron-induced 235 U fission
fragments (see table 1). The sum of the kinetic energy of the X1 and
X2 fission fragments is greater than 165 MeV. Fission fragment data
from [568].
3. Future solid-state neutron detector devices and
applications
number of e–h pairs generated per neutron-induced fission
event from uranium, the ternary intermetallics should not be
immediately discounted for their high thermally induced noise
level as with the low-bandgap and low-resistivity lithiumbased intermetallics. However, a reduced depletion region
width due to low resistivity may significantly limit the uranium
intermetallic film thickness despite their relatively high carrier
Fermi velocity or mobility compared to the boron-rich solids;
in a different light, enhanced pulse rise times with large
charge bursts may open a new direction for time resolution.
Meek and co-workers report both p–n junctions and p–n–p
transistors fabricated from UO3 and UO2 [483, 485]. The
UO2 , in the work reported by Meek, was processed into thin
films using the sol–gel process, although use of thermal-based
CVD and atomic-layer-like deposition from organometallic
uranium precursors [508, 509], as well as standard evaporation
and RF sputtering methods [510–512], with proper heavy
metal scrubbers would be a natural and safe evolution in
developing this area. Further, the refractory nature of the
uranium complexes (especially the oxides) may prove useful to
applications requiring radiation hardness and high-temperature
operation [483–485]. Neptunium (237 Np) and plutonium
(239 Pu) both exhibit energy-dependent cross sections for
neutron-induced fission as well as average fission fragment
energies that are very similar to those of the uranium isotopes
described above. However, given uranium’s abundance,
reduced relative toxicity and much more well-behaved physical
properties, it is unlikely that any other actinoid complex will
be preferentially considered as a main constituent in an active
semiconductor in a direct-conversion device. Further, although
the average fission fragment energy from uranium neutron
capture is quite high (see figure 10 and table 1), mean free paths
of the order of one micron for the primary intermediate energy
fragments and high self-absorption will still hinder the use
of uranium converters for most indirect-conversion devices; a
study of optimal uranium compound layer thicknesses, similar
to that completed by Shultis and McGregor for boron- and
lithium-rich solids [6], is needed. Radiation damage from
Future solid-state neutron detectors will not require cumbersome or costly power supplies; they will be able to resolve the
incident energy of a neutron to less than 10 keV without the
many meters of real estate needed for time-of-flight detection,
and will be able to detect neutrons in a high gamma-ray
background environment through the combination of clever
physics with current materials and device know-how. Below
are initial results and seedlings of ideas that may lead to desired
attributes for future neutron detectors to be used in research or
commercial applications.
3.1. Zero external bias
Electron–hole pair separation and sweepout in a solid, or
in solids joined to form a heterostructure, depend primarily
on the strength and extent of the internal electric field;
without a sufficiently large electric field, carriers may become
immobilized by traps or recombine. The internal electric field
may be generated, in magnitude and width, intrinsically by
a space charge region or through a superposition of a space
charge region and externally applied voltage. In applications
that require quantitative analysis of the linear energy transfer
of primary reaction products (see section 3.3), an electric field
strong enough to sweep out e–h pairs at or near unity is
required. However, in neutron counting applications where
it is only necessary to obtain voltage pulses with intensity
discernibly above the noise, use of the intrinsic electric field
from a p–n junction or Schottky device, operating in the voltaic
mode (i.e. with zero applied bias) is possible, as has recently
been demonstrated for both direct- [276, 419, 420, 422] and
indirect-conversion [117, 223, 280, 286, 304, 513] solidstate neutron detection heterostructures. Eliminating the
power conditioning and supply electronics required to establish
high DC voltages for reverse bias is a major improvement,
especially for dosimetric applications where a thin form
factor is advantageous. Further improvements to this area
for indirect-conversion devices will require reduced surface
17
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
Figure 11. (a) Proton recoil scattering; (b) scattering geometry used in proton recoil spectrometry to resolve E n after Gotoh [533] and (c)
schematic representation of a spatially resolved solid-state proton recoil spectrometer after Meijer [232].
emphasis on new ideas for future detectors that take advantage
of advances in spatial, temporal and energy resolution of
moderate-energy ions as well as unique moderator–detector
scattering geometries.
Conservation of energy may be used in one or any
combination of four methods to determine incident neutron
energy: (1) through the determination of the angular
distribution of neutron-induced recoil nuclei; (2) by the
predictable reduction of neutron energy through elastic
scattering; (3) from the determination of the difference
in energies of neutron-induced capture reaction or fission
fragment nuclei relative to the reaction energy Q or (4) by
the broadening of spectroscopically resolved peaks caused
by inelastic scattering. The physics associated with the first
method was initially described in 1941 by Amaldi [526, 527]
and subsequently optimized with solid-state detectors for
the proton recoil telescope geometry from 1948 to the
present [179, 232, 248, 252, 257, 330, 332, 528–539]. In this
geometry, the telescope diameter and distance from the proton
radiator form a solid angle by which neutron energy incident to
the radiator may be determined according to E p = E n cos2 ϕ ,
where E p and E n are the proton and neutron energies and
ϕ is the angle between the incident neutron and scattered
proton (figures 11(a) and (b)). One or more transmission
detector(s) (a.k.a. d E/dx detectors) and a stopping detector
are used in coincidence to remove background or protons from
stray neutrons. The thickness of the proton radiator, width
of the solid angle, as well as thickness and number of the
transmission detectors form the basis for the detector efficiency
and energy resolution. Energy resolutions of 500 keV
for 1–5 MeV incident neutrons [530], 70 keV for thermal
neutrons [330], 150 keV for 2–4 MeV neutrons [332], 270 keV
for 6 MeV neutrons [179] and 100 keV for 100 keV–25 MeV
neutrons have been reported with typical total efficiencies of
much less than 1%. Suggested improvements to the radiator
and heterostructure materials for enhancing energy resolution
include using liquid hydrogen in place of a polyethylene
radiator to remove the carbon contaminant and using
germanium rather than silicon transmission detectors to reduce
total detector thickness and extraneous (n, p) reactions [248].
Increasing spatial or x y sensitivity using crossed strip detectors
(figure 11(c)) has been posited [232] to provide a detector
capable of determining all scattered angles simultaneously,
recombination, especially for the etched structures, while
for direct-conversion devices, the trap concentration of the
materials and surface recombination of the heterostructures, in
general, must be decreased.
3.2. Resolving incident neutron kinetic energy
Knowledge of free neutron kinetic energy is necessary for
applications where the source or scattering of neutrons
is unknown or unpredictable. An application of current
relevance is in the solid-state detection of neutrons from
special nuclear materials by active or passive detection
(to serve as an alternative to gaseous 3 He [64, 514]).
A number of other applications, including absorbed-dose
determination and inelastic neutron scattering, also require
energy analysis, whose details may be expanded from the
following discussion [515]. In the detection of illicitly
transported or stored special nuclear materials, especially for
neutrons from the unmoderated spontaneous fission (or (α ,
n)) neutron production from plutonium (or plutonium oxides)
whose emission energy reaches an apex at 800 keV–1.1 MeV
(800 keV–2 MeV), it is advantageous to be able to separate
out unmoderated background cosmic-ray-induced spallation
neutrons whose spectral weight overlaps up to the top end of
the evaporation spectrum [516–525] in the fast-neutron range.
This means that a detector with an energy resolution of at least
100 keV (from 100 keV to 15 MeV), with a high total efficiency
for both thermal and fast neutrons, and which overcomes
the deficiencies of time-of-flight (>3 m throw distance) and
scintillator-based devices in the context of energy resolution is
desired.
Whether through a direct- or indirect-conversion solidstate heterostructure geometry, conservation of energy and
momentum forms the basis from which incident neutron
kinetic energy may be determined; one may additionally make
use of isotopes with abrupt changes in their microscopic cross
section as a function of incident neutron energy to transmit
(absorb) neutrons of a wanted (unwanted) energy (i.e. yielding
a positive (negative) filter effect). While Knoll [2] has provided
an excellent survey of most neutron spectroscopy detector
types in the context of the available transduction mechanisms,
the intention here is to briefly summarize reported solidstate heterostructure-based neutron spectrometers with an
18
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
Figure 12. (a) Cylindrically and (b) spherically symmetric solid-state Bonner spectrometers.
rather than having to sweep a traditional telescope through
Empirical demonstration of this
all desired ϕ angles.
proposed x y device [232, 536] has not been reported and is
a significant improvement on the traditional telescope and an
excellent direction for theoretical [532, 533], computational
and experimental development.
The second method was first reported by Bonner [540]
in 1960 with a scintillator-based detector and subsequently
adapted with solid-state detectors in a handful of work
reported only in the last six years [419, 420, 541–545]. The
geometry associated with this second method (figure 12)
relies on the statistically probable energy loss of neutrons
as they are slowed through elastic collisions (usually with
hydrogen). Because the neutron energy loss from elastic
collisions is statistically predictable, one can measure the
neutron flux as a function of elastic scattering medium
thickness, then work backwards, reconstruct or unfold the
spectra to determine the incident neutron energy [546–552].
The solid-state Bonner detectors are significantly advantageous
over their gaseous or scintillator-based counterparts because
they provide three-dimensional scattering information through
the detection of both recoiled protons as well as captured
neutrons to yield the highest-energy resolution possible from
unfolding calculations.
There is also the potential for
determining neutron energy without unfolding calculations by
simply determining the neutron capture intensity as a function
of calibrated moderator/detector depth for parallel incident
neutrons, using the cylindrically symmetric geometry shown
in figure 12(a), or radial dependence for omnidirectional
neutrons, using the spherically symmetric geometry shown
in figure 12(b). The three dimensionality comes about
from having pixilated detectors in the x y plane that are
subsequently stacked in the z direction (hence, giving a polar
[r, θ ] dependence in the cylindrically symmetric device and
spherical [r, θ, ϕ] dependence in the spherically symmetric
system). Engineering efforts to be completed for both designs
include the multiplexing, readout electronics and pulse analysis
required for hundreds or even thousands of neutron active
elements.
The third method for spectroscopically resolving neutrons,
based on conservation of energy, was first described
and demonstrated by Love and Murray [180, 553] in
1961 for a thin layer of 6 LiF deposited between two
solid-state surface-barrier detectors (figure 7(d)); a flurry
of development using other conversion layer isotopes in
the sandwich geometry continued thereafter into the late
1970s [182, 214, 249–251, 253, 254, 256, 330, 332–335,
554–556]. In this method, the energy of all reaction products
is determined, whose sum less the known reaction energy (Q)
yields the incident neutron energy. The challenge then lies
in precisely determining the energy of the reaction products
in excess of Q , especially in the limit where the incident
neutron energy is much less than Q . Values of Q for
some semiconductor-based isotopes relevant to neutron capture
primary reaction products are listed in table 1. Overcoming the
challenge of detecting neutron energies much less than Q can
be met by either utilizing direct-conversion heterostructures
where the primary reaction product energy is not lost in a
conversion layer and/or by utilizing neutron capture reactions
that result in the lowest possible Q (e.g. triton analysis
from 3 He—see table 1) up to the limit that the detector
can accurately detect the primary reaction product itself.
Reported resolutions for this method are 15 keV for 17 MeV
neutrons [335], 201 keV for 14.7 MeV neutrons [182], 90 keV
for 125 keV neutrons [250], 1 keV for hundreds of keV
neutrons [251] 1–40 keV for 10–400 keV neutrons [253] and
200 keV for 500 keV neutrons [256]. Thresholds for neutron
capture with limited or discrete reaction products in certain
energy ranges, most pronounced for the isotopes of Si and Ge
in figure 9 (e.g. 30 Si(n, α)27 Mg is the only reaction possible
below E n = 10 MeV), were posthumously exploited by
Bonner [334] and others in enhancing resolution. If pushed
further for the desired neutron energies (e.g. very high-energy
resolution for a diffraction study), the filtering effect from
thresholds or microscopic cross-section resonances could be
used to provide very high-energy resolution, limited only by
the width of the resonance. Given the advances made in the
energy resolution for charged particle detectors since the last
report above in 1978, even in the indirect-conversion geometry,
this third method, coupled with low Q reactions is worth
stringent consideration to empirically determine whether subkeV energy resolution is possible.
The fourth method described here to resolve incident
neutron energy is based on unfolding the neutron-energydependent broadening or response of spectroscopically
19
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
or scattered whose primary reaction products produce e–
h pairs that are subsequently separated, the resulting spinpolarized electrons (and holes) now, on average, have a 50%
spin polarization in each quantum mechanically allowed spin
direction. This finite carrier spin polarization in the allowed
direction of the spin-polarized metals, satisfying the carrier
spin-conservation rule, allows for a finite current as a result
of the non-infinite resistance state. Thus, a very high signalto-noise solid-state neutron detector that exploits the quantum
mechanical spin selection rule for tunneling is theoretically
possible. In practice, the neutron-sensitive tunnel barrier
is so thin that the probability for neutron interaction, even
at thermal energies, would be very poor, although a boroncarbide-based magnetic tunnel junction has been successfully
fabricated [565]. An adaptation that would help rectify this
deficiency, but still falls under the spintronics umbrella, would
be to use a giant magnetoresistive (GMR) geometry [566]
and replace the standard metallic spacer with a thick nonspin-polarized layer that is neutron sensitive and metallic.
However, the spin selectivity in the GMR geometry is not
rigorously restrictive as in tunneling transport, which may lead
to leakage current levels that are too high (when compared
to the charge generated by the primary reaction products
of a single neutron capture). Another adaptation would be
to create a conversion-layer-based magnetic tunnel junction
array, in which the primary reaction products are within
range of the tunnel barrier, such as a lithium-or boron-based
film deposited onto commercially available magnetic randomaccess memory [567]. Finally, other out-of-the-box ideas
for spintronics-based neutron detectors that utilize the spin
polarization injection materials rather than just the neutronsensitive insulators, such as uranium intermetallics [476] or
gadolinium-pnictides [459], may be possible.
resolved peaks caused by inelastic scattering. Very little work
has been reported [81, 557] using this technique and it has
been restricted to the 692 keV conversion electron produced
from neutron scattering with 72 Ge, although there has been
sufficient experimental detail of the 692 keV peak [14–16, 558]
to support that the unfolding technique is valid. Further
development of this method using isotopes that produce
conversion electrons, but with much larger cross sections
than that for the 72 Ge(n, n e− ) event, would be advantageous.
Ultimately, this method is interesting theoretically, but may
fundamentally yield too low a resolution relative to the third
method discussed above and will likely remain limited to
detectors whose primary purpose is to detect particles other
than neutrons.
Overall, improving incident neutron energy resolution to
less than 100 keV for neutron energies above 500 keV is of
technological relevance [256] and to less than 1 keV may be
considered a grand challenge, possibly requiring very different
transduction mechanisms than presented above.
3.3. Concept for a spintronic neutron detector for enhanced
signal-to-noise
Inherent in charge-to-voltage conversion is noise. The noise
depends on the input capacitance of the device, preamplifier
voltage and type of noise source. Shaping amplifiers, lockin techniques, coincidence subtraction and proper signal
shielding/guarding/grounding provide a significant reduction
in leakage current, signal attenuation and signal mixing.
However, single-neutron detection—which is sometimes like
looking for a needle in a haystack—in the solid state requires
the ultimate in signal-to-noise, especially for zero or very low
(<10 V) applied bias (vide ante). One very forward-thinking
multidisciplinary concept, suggested by Bandyopadhyay and
Tepper [559], couples the low noise attributes of a magnetic
tunnel junction with the secondary electrical carriers generated
from neutron capture to provide a very high signal-to-noise
spintronic-based solid-state neutron detector.
The magnetic tunnel junction [560, 561], postulated
by Tedrow and Meservey [562] in 1971, first realized
by Jullière [563] in 1975 and shown to operate at room
temperature by Moodera [564] in 1995, is a two-state resistive
device based on the conservation of electron (or hole) spin
momentum during quantum mechanical tunneling between two
spin-polarized metals across a thin electrically insulating gap
(i.e. a tunnel barrier). Ideally, when a small voltage is applied
and when the spin polarization of the two metals (assuming
100% spin polarization) is antiparallel, electrical carrier spin
conservation prevents tunneling of the spin-polarized carrier
between the metals and thus creates an infinite tunneling
resistance state, whereas when the spin polarization of the
two metals is parallel, the tunneling resistance is zero. If
one considers that the thin electrically insulating material
comprises isotopes that produce moderate-energy charged
particles upon neutron capture or scattering and that the spin
polarization of the metals is antiparallel, then when zero
neutrons are incident on the device, the tunneling resistance
state is infinite. However, when incident neutrons are captured
4. Summary
Despite the narrow focus of this review and tremendous
advances made over the last 57 years in charged-particle-based
neutron transduction using solid-state heterostructures, only
the surface of the basic and applied challenges—in light of
applications—has been scratched. More than ever before,
neutron-based applications in the search for water on distant
planets, in determining long range magnetic information of
a solid, in understanding the weak interaction with dark
matter, or in uncovering the presence of plutonium, require
neutron detectors with significantly enhanced attributes. The
neutron detector attributes necessary to meet the requirements
for these applications include combinations of very high
(>90%) total detection efficiency under flux conditions at or
near the cosmic-ray-induced spallation background over the
thermal- to fast-neutron energy range, and micron to submicron
spatial resolution, tens of nanosecond temporal resolution
and/or less than 10 keV incident energy resolution. To meet
these challenges, exploration of alternative neutron moderators
and reflectors, novel moderator mechanisms (e.g. neutron
multiplication) and new moderator geometries with little or
no attenuation is needed. Indirect-conversion neutron detector
heterostructure advances will likely come about from advanced
20
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
lithographies that can be used to construct device geometries
designed with computational packages whose accuracy for
reaction and transport of all particle types outmatches the
human ability to build such a device. Breakthroughs in directconversion-based neutron detector heterostructures will require
electronic, physical and chemical understanding of the active
semiconductors used at the level understood for silicon at least
30 years ago. Finally, I posit that the development of uraniumbased solids for direct- and indirect-conversion neutron
detector heterostructures, utilizing the neutron-induced fission
fragment reaction, is the most important and direct path
to realizing next-generation detectors that satisfy the above
application-specific requirements.
[15] Gete E et al 1997 Neutron-induced peaks in Ge detectors from
evaporation neutrons Nucl. Instrum. Methods Phys. Res. A
388 212–9
[16] Chasman C, Jones K W and Ristinen R A 1965 Fast neutron
bombardment of a lithium-drifted germanium gamma-ray
detector Nucl. Instrum. Methods 37 1–8
[17] Ejiri H, Matsuoka K and Choi E 1991 Low-background
measurement of fast neutrons by means of Ge detectors
Nucl. Instrum. Methods Phys. Res. A 302 482–8
[18] Chung C and Chen Y R 1991 Application of a germanium
detector as a low flux neutron monitor Nucl. Instrum.
Methods Phys. Res. A 301 328–36
[19] Jakubek J et al 2006 CdTe hybrid pixel detector for imaging
with thermal neutrons Nucl. Instrum. Methods Phys. Res. A
563 238–41
[20] Fasasi M et al 1988 Thermal neutron dosimetry with cadmium
telluride detectors Radiat. Prot. Dosim. 23 429–31
[21] McGregor D S, Lindsay J T and Olsen R W 1996 Thermal
neutron detection with cadmium1−x zincx telluride
semiconductor detectors Nucl. Instrum. Methods Phys. Res.
A 381 498–501
[22] Ashcroft N W and Mermin N D 1976 Solid State Physics
(New York: Brooks/Cole)
[23] Sze S M 1981 Physics of Semiconductor Devices (New York:
Wiley)
[24] Schroder D K 1998 Semiconductor Material and Device
Characterization (New York: Wiley–Interscience)
[25] Klein C A 1968 Bandgap dependence and related features of
radiation ionization energy in semiconductors J. Appl.
Phys. 39 2029–38
[26] Carron N J 2007 An Introduction to the Passage of Energetic
Particles Through Matter (Boca Raton, FL: CRC Press)
[27] Ogawa Y 2006 Measurements of depth doses by thermal
neutron and of neutron energy spectra using passive
neutron dosimeter with spherical polyethylene moderator
Radioisotopes 55 1–12
[28] Gozani T 2009 Fission signatures for nuclear material
detection IEEE Trans. Nucl. Sci. 56 736–41
[29] Gozani T 1981 Active nondestructive assay of nuclear
materials—principles and applications NUREG/CR-0602
[30] Ruddy F H et al 2006 Nonintrusive method for the detection
of concealed special nuclear material USPTO 7,151,815
[31] Locher G L 1936 Biological effects and therapeutic
possibilities of neutrons Am. J. Roentgenol Radium Ther.
Nucl. Med. 36 1–13
[32] Hawthorne M F, Shelly K and Wiersema R J 2001 Frontiers in
Neutron Capture Therapy (New York: Springer)
[33] Matsumoto T and Aizawa O 1988 Dose measuring system for
boron neutron capture therapy Nucl. Instrum. Methods
Phys. Res. A 271 662–70
[34] Enger S A et al 2006 Monte Carlo calculations of thermal
neutron capture in gadolinium: a comparison of GEANT4
and MCNP with measurements Med. Phys. 33 337–41
[35] Wollan E O and Shull C G 1948 The diffraction of neutrons
by crystalline powders Phys. Rev. 73 830–41
[36] Fischer J, Radeka V and Boie R A 1983 Position-Sensitive
Detection of Thermal Neutrons (New York: Academic)
[37] Evsukov A M et al 1991 Radiation and its Effects on Devices
and Systems (RADECS 91., 1st European Conf. on 1991)
pp 74–6
[38] Geandier G et al 2008 Benefits of two-dimensional detectors
for synchrotron x-ray diffraction studies of thin film
mechanical behavior J. Appl. Crystallogr. 41 1076–88
[39] Vohrer U et al 2005 Quantitative XPS imaging—new
possibilities with the delay-line detector Appl. Surf. Sci.
252 61–5
[40] Siegmund O H W et al 2007 A high spatial resolution even
counting neutron detector using microchannel plates and
cross delay line readout Nucl. Instrum. Methods Phys. Res.
A 579 188–91
Acknowledgments
This work was supported by the Office of Naval Research
under grant no. N00014-10-1-0419. Extreme gratitude is owed
to M M Paquette, M S Driver and S R Messenger for their
tremendous help in preparation of this review. The author
would also like to thank Z Bell, W K Pitts, P A Dowben,
R C Block, D S McGregor, N Balasaro and W H Miller for
their useful discussions, rigor and steadfast integrity over the
last ten years that is represented in a majority of this work.
References
[1] Bell Z W et al 2005 Neutron detection with cryogenics and
semiconductors Phys. Status Solidi c 2 1592–605
[2] Knoll G F 2000 Radiation Detection and Measurement
(Hoboken: Wiley)
[3] Leroy C and Rancoita P-G 2009 Principles of Radiation
Interaction in Matter and Detection (Hackensack: World
Scientific)
[4] Lutz G 1999 Semiconductor Radiation Detectors (Berlin:
Springer)
[5] McGregor D S et al 2003 Design considerations for thin film
coated semiconductor thermal neutron detectors—I: basics
regarding alpha particle emitting neutron reactive films
Nucl. Instrum. Methods Phys. Res. A 500 272–308
[6] Shultis J K and McGregor D S 2009 Design and performance
considerations for perforated semiconductor
thermal-neutron detectors Nucl. Instrum. Methods Phys.
Res. A 606 608–36
[7] Petrillo C et al 1996 Solid state neutron detectors Nucl.
Instrum. Methods Phys. Res. A 378 541–51
[8] Peurrung A J 2000 Recent developments in neutron detection
Nucl. Instrum. Methods Phys. Res. A 443 400–15
[9] Milbrath B D et al 2008 Radiation detector materials: an
overview J. Mater. Res. 23 2561–81
[10] Spieler H 2005 Semiconductor Detector Systems (Oxford:
Oxford University Press)
[11] Takahashi K et al 2003 Design of neutron detector by using a
novel superconductor MgB2 Physica C 392–396 1501–3
[12] Emin D and Aselage T L 2005 A proposed
boron-carbide-based solid-state neutron detector J. Appl.
Phys. 97 013529
[13] Febrenbacher G, Meckbach R and Paretzke H G 1996 Fast
neutron detection with germanium detectors: computation
of response function for the 692 keV inelastic scattering
peak Nucl. Instrum. Methods Phys. Res. A 372 239–45
[14] Seifert A et al 2007 Fast neutron sensitivity with HPGe IEEE
Nucl. Sci. Symp. Conf. Rec. N24–62 1175
21
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[41] Sundelin R M et al 1968 Spectrum of Neutrons from muon
capture in silicon, sulfur, and calcium Phys. Rev. Lett.
20 1198–200
[42] Beaumont S P et al 1991 GaAs solid state detectors for
physics at the LHC IEEE Nucl. Sci. Symp. Conf. Rec. 40
134–9
[43] Coleman R F 1966 The application of neutron activation
analysis to forensic science J. Forensic Sci. Soc. 6 19–27
[44] Fetter S et al 1990 Detecting nuclear warheads Sci. Glob.
Security 1 225–302
[45] Ruddy F H et al 1999 Monitoring of neutron and gamma
radiation USPTO 5,969,359
[46] Kania D R et al 1988 High speed detection of thermonuclear
neutrons with solid state detectors IEEE Trans. Nucl. Sci.
35 387–8
[47] Youmans A H 1956 Conduction counter for radioactivity well
logging USPTO 2,760,078
[48] Feldman W C et al 2002 Global distribution of neutron from
mars: results from mars odyssey Science 297 75–8
[49] Hennings-Yeomans R and Akerib D S 2007 A neutron
multiplicity meter for deep underground muon-induced
high-energy neutron measurements Nucl. Instrum. Methods
Phys. Res. A 574 89–97
[50] Squires G L 1996 Introduction to the Theory of Thermal
Neutron Scattering (New York: Dover)
[51] Mohideen U and Roy A 1998 Precision Measurement of the
Casimir Force from 0.1 to 0.9 μm Phys. Rev. Lett.
81 4549–52
[52] Baker C A et al 2006 Improved experimental limit on the
electric dipole moment of the neutron Phys. Rev. Lett.
97 131801
[53] Cho G-C, Haba N and Honda M 2005 Neutron electric dipole
moment and flavor changing interactions in
supersymmetric theories Mod. Phys. Lett. A 20 2969–76
[54] Golub R and Huffman P R 2005 Search for a neutron electric
dipole moment J. Res. Natl Inst. Stand. Technol. 110
169–72
[55] Epsy M A et al 1999 SQUIDs as detectors in a new
experiment to measure the neutron electric dipole moment
IEEE Trans. Appl. Supercond. 9 3696–9
[56] Kessler K 1985 Polarized Electrons (Berlin: Springer)
[57] Oppenheimer J R and Schwinger J 1941 On the interaction of
mesotrons and nuclei Phys. Rev. 60 150–2
[58] Kirschner J and Feder R 1979 Spin polarization in double
diffraction of low-energy electrons from W(001):
experiment and theory Phys. Rev. Lett. 42 1008–11
[59] Rashba E I 1960 Properties of semiconductors with an
extremum loop Sov. Phys.—Solid State 2 1109–22
[60] Dresselhaus G and Dresselhaus M S 1965 Spin-orbit
interaction in graphite Phys. Rev. A 140 401–12
[61] Schwinger J 1948 On the polarization of fast neutrons Phys.
Rev. 73 407–9
[62] Schwinger J 1946 Polarization of neutrons by resonance
scattering in helium Phys. Rev. 69 681
[63] Mezei F 1972 Neutron spin echo: a new concept in polarized
thermal neutron techniques Z. Phys. 255 146–60
[64] Kouzes R T 2009 The 3 He supply problem PNNL-18388
[65] Moon R M, Riste T and Koehler W C 1969 Polarization
analysis of thermal-neutron scattering Phys. Rev.
181 920–31
[66] Gay T J and Dunning F B 1992 Mott electron polarimetry
Rev. Sci. Instrum. 63 1635–51
[67] Cowan C L and Reines F 1956 Detection of the free neutrino:
a confirmation Science 124 103–4
[68] Fischer G J 1957 Cross section for the (n, 2n) reaction in Be9
Phys. Rev. 108 99–103
[69] Daire A 2000 Precision Electromagnetic Measurements
Digest pp 373–4
[70] Klein L J and Williams C C 2001 Single electron tunneling
detected by electrostatic force Appl. Phys. Lett. 79 1828–30
[71] Poignant R V and Przybylowicz E P 1975 Elements and
process for recording direct image neutron radiographs
USPTO 3,887,807
[72] Burton M and Neubert T J 1956 Effect of fast neutron
bombardment on physical properties of graphite: a review
of early work at the metallurgical laboratory J. Appl. Phys.
27 557–67
[73] Studenkel D, Holloway J P and Knoll G F 1999 Neutron
spectrum unfolding using a modified truncated singular
value decomposition Nucl. Sci. Eng. 132 261–72
[74] Evans R D 1955 The Atomic Nucleus (New York:
McGraw-Hill)
[75] Compton A H 1923 A quantum theory of the scattering of
x-rays by light elements Phys. Rev. 21 483–502
[76] Hertz H R 1887 ber einen Einfluss des ultravioletten Lichtes
auf die electrische Entladung Ann. Phys. 31 983–1000
[77] Anderson C D 1932 Energies of cosmic-ray particles Phys.
Rev. 41 405–21
[78] Kossel W 1914 Bemerkung zur absorption homogener
Röntgenstrahlen Verzeichnis der Veröffentlichungen
Goldsteins 16 898–909
[79] Siller L et al 2009 Core and valence exciton formation in
x-ray absorption, x-ray emission and x-ray excited optical
luminescence from passivated Si nanocrystals at the Si L2,3
edge J. Phys.: Condens. Matter 21 095005
[80] Meitner L 1923 Das β -strahlenspektrum von UX1 und seine
deutung Z. Phys. A 17 54–66
[81] Siiskonen T and Toivonen H 2005 A model for fitting peaks
induced by fast neutrons in an HPGe detector Nucl.
Instrum. Methods Phys. Res. A 540 403–11
[82] Skoro G P et al 1992 Environmental neutrons as seen by
germanium gamma-ray spectrometer Nucl. Instrum.
Methods Phys. Res. A 316 333–6
[83] Chao J-H and Chung C 1992 Low-level neutron monitoring
by use of germanium detectors Nucl. Instrum. Methods
Phys. Res. A 321 535–8
[84] Bethe H and Ashkin J 1953 Experimental Nuclear Physics ed
E Segre (New York: Wiley)
[85] Billington D S and Crawford J H 1960 Radiation Damage in
Solids (Princeton, NJ: Princeton University Press)
[86] Corbett J W and Watkins G D 1965 Production of divacancies
and vacancies by electron irradiation of silicon Phys. Rev.
A 138 555–60
[87] Dale C J et al 1988 High energy electron induced
displacement damage in silicon IEEE Trans. Nucl. Sci.
35 1208–14
[88] Denney J M and Downing R G 1963 Charged particle
radiation damage in semiconductors, VIII: the electron
energy dependence of radiation damage in photovoltaic
devices NASA Contract No. NAS5-1851
[89] Seitz F and Koehler J S 1956 Solid State Physics (New York:
Academic Press) p 305
[90] Dienes G J and Vineyard G H 1957 Radiation Effects in Solids
(New York: Interscience)
[91] Kinchin G H and Pease R S 1955 The displacement of atoms
in solids by radiation Rep. Prog. Phys. 18 1–51
[92] Boersch H, Radeloff C and Sauerbrey G 1961 Experimental
detection of transition radiation Phys. Rev. Lett. 7 52–4
[93] Rutherford E 1919 Collisions of alpha particles with light
atoms. IV. An anomalous effect in nitrogen Lond.
Edinburgh Dublin Phil. Mag. J. Sci. 37 581–7
[94] Norbeck E and Littlejohn C S 1957 Experimental survey of
nuclear transformations caused by 2 MeV lithium ions
Phys. Rev. 108 754–9
[95] Murphy P G 1957 Energy level spectrum of C15 from the
nuclear reaction Be9 (Li7 , p)C15 Phys. Rev. 108 421–6
[96] Norbeck E 1957 New nuclear reactions induced by 2 MeV
lithium ions the masses of B13 and C15 Phys. Rev.
105 204–9
22
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[123] Iwanenko D and Pomeranchuk I 1944 On the maximal energy
attainable in a betatron Phys. Rev. 65 343
[124] Cherenkov P A 1934 Visible emission of clean liquids by
action of γ radiation Dokl. Akad. Nauk SSSR 2 451–4
[125] Oster L 1960 Effects of collisions on the cyclotron radiation
from relativistic particles Phys. Rev. 199 1444–56
[126] Baldwin G C and Klaiber G S 1947 Photo-fission in heavy
elements Phys. Rev. 71 3–10
[127] Bohr N and Wheeler J A 1939 The mechanism of nuclear
fission Phys. Rev. 56 426–50
[128] Veselago V G 1968 The electrodynamics of substances with
simultaneously negative values of ε and μ Sov. Phys.—Usp.
10 509–14
[129] Duan Z Y et al 2009 Research progress in reversed cherenkov
radiation in double-negative metamaterials Prog.
Electromagn. Res. 90 75–87
[130] Allison S K and Warshaw S D 1953 Passage of heavy particles
through matter Rev. Mod. Phys. 25 779–817
[131] Ehrenberg W, Lang C-S and West R 1951 The electron voltaic
effect Proc. R. Soc. 64 424
[132] MacMahon A M and Snell A H 1951 Measuring device and
method of measuring USPTO 2,564,626
[133] Hill D E 1963 Neutron detector of crystalline boron phosphide
USPTO 3,113,210
[134] Imai T, Yashiki T and Fujimori N 1992 Radiation detecting
element USPTO 5,079,425
[135] Jones B L, Nam T L and Keddy R J 1993 Diamond neutron
detector USPTO 5,216,249
[136] Siffert P M et al 1976 Nuclear detectors sensitive to alpha,
beta, and gamma rays and to thermal neutrons and to
methods of treatment of crystals of such detectors USPTO
3,999,071
[137] Hassard J F and Choi P 2000 Ionizing radiation detector
USPTO 6,072,181
[138] Doty F P, Zwieback I and Ruderman W 2002 Solid state
neutron detector and method for use USPTO 6,388,260
[139] Doty F P 2004 Boron nitride solid state neutron detector
USPTO 6,727,504
[140] Sane A Y, Leist J R and Moore W 2006 Neutron detector
employing doped pyrolytic boron nitride and method of
making thereof USPTO 7,034,307
[141] Leist J R, Moore W A and Sane A Y 2003 Semiconductor
detector for thermal neutrons based on pyrolytic boron
nitride USPTO 6,624,423
[142] McGregor D S, Unruh T C and McNeil W J 2008 Thermal
neutron detection with pyrolytic boron nitride Nucl.
Instrum. Methods Phys. Res. A 591 530–3
[143] Losee J R et al 2008 Solid-state thermal neutron detector
USPTO 7,372,009
[144] Schieber M M, Zuck A and Marom G 2009 Solid-state
neutron and alpha particles detector and methods for
manufacturing and use thereof USPTO US2009/0302226
[145] Butler M A, Ginley D S and Bryson J W 1986 Neutron
induced conductivity in trans-polyacetylene Appl. Phys.
Lett. 48 1297–9
[146] Butler M A and Ginley D S 1987 Organic metal neutron
detector USPTO 4,641,037
[147] LeVert F E 1997 Development of an improved direct neutron
sensor Rep. No. A796233 Defense Nuclear Agency
[148] Alekeev I V 2008 A neutron semiconductor detector based on
TlInSe2 Instrum. Exp. Techn. 51 331–5
[149] Bykovskii Y A et al 1993 Neutron fluence sensor based on
boron carbide Tech. Phys. Lett. 19 457–8
[150] Ananthanarayanan K P, Gielisse P J and Choudry A 1974
Boron compounds for thermal-neutron detection Nucl.
Instrum. Methods 118 45–8
[151] Kumashiro Y et al 1988 Thermal neutron irradiation
experiments on 10 BP single-crystal wafers J. Less Common
Met. 143 71–5
[97] Cohen B L, Newman E and Handley T H 1955
( p, pn) + ( p, 2n) and ( p, 2 p) cross sections in medium
weight elements Phys. Rev. 99 723–7
[98] Blatt J M and Weisskopf V F 1952 Theoretical Nuclear
Physics (New York: Wiley)
[99] Robertson J C and Rowland M S 1989 Pulsed neutron detector
USPTO 4,814,623
[100] Egerton R F 1996 Electron Energy-Loss Spectroscopy in the
Electron Microscope (New York: Plenum)
[101] Schultz D et al 2010 The K-shell auger electron spectrum of
gadolinium obtained using neutron capture in a solid state
device J. Phys. D: Appl. Phys. 43 75502–8
[102] Srour J R, Marshall C J and Marshall P W 2003 Review of
displacement damage effects in silicon devices IEEE Trans.
Nucl. Sci. 50 653–70
[103] Narendra S G and Chandrakasan A 2006 Leakage in
Nanometer CMOS Technologies (New York: Springer)
[104] Holmes-Siedle A 1974 The space-charge dosimeter: general
principles of a new method of radiation detection Nucl.
Instrum. Methods 121 169–79
[105] Loferski J J and Rappaport P 1958 Radiation damage in Ge
and Si detected by carrier lifetime changes: damage
thresholds Phys. Rev. 111 432–9
[106] Pease R L et al 1987 Comparison of proton and neutron
carrier removal rates IEEE Trans. Nucl. Sci. 34 1140–6
[107] Lund J C, Sinclair F and Entine G 1986 Neutron dosimeter
using a dynamic random access memory as a sensor IEEE
Trans. Nucl. Sci. 33 620–3
[108] Arita Y et al 2001 Experimental investigation of thermal
neutron-induced single event upset in static random access
memories Japan. J. Appl. Phys. 40 L151–3
[109] Phillips G W et al 2002 Feasibility of a neutron
detector-dosimeter based on single-event upsets in dynamic
random-access memories Radiat. Prot. Dosim. 101 129–32
[110] Hughes H L 2005 Semiconductor substrate incorporating a
neutron conversion layer USPTO 6,867,444
[111] August R A et al 2007 Neutron detection device and method
of manufacture USPTO 7,271,389
[112] Nagarkar V, Entine G and Stoppel P 1994 Flight prototype of
a solid state neutron dosimeter for space applications IEEE
Trans. Nucl. Sci. 41 1333–7
[113] Nagarkar V et al 1992 Solid state neutron dosimeter for space
applications IEEE Trans. Nucl. Sci. 39 966–70
[114] Swinehart P R and Swartz J M 1979 Sensitive silicon pin
diode fast neutron dosimeter USPTO 4,163,240
[115] Swartz J M and Thurston M O 1966 Analysis of the effect of
fast-neutron bombardment on the current–voltage
characteristic of a conductivity-modulated p–i–n diode
J. Appl. Phys. 37 745–55
[116] Darambara D G and Spyrou N M 1994 Neutron detection
using soft errors in dynamic random access memories Nucl.
Instrum. Methods Phys. Res. A 348 491–5
[117] Orlic M, Lazarevic V and Boreli F 1989 Microdosimetric
counter based on semiconductor detectors Radiat. Prot.
Dosim. 29 21–2
[118] Kronenberg S and Brucker G J 1995 The use of hydrogenous
material for sensitizing pMOS dosimeters to neutrons IEEE
Trans. Nucl. Sci. 42 20–6
[119] Blamires N G et al 1986 pMOS dosimeters: long-term
annealing and neutron response IEEE Trans. Nucl. Sci.
33 1310–5
[120] Kramer G 1967 Semiconductor fast-neutron dosimeter with
low threshold of sensitivity IEEE Trans. Nucl. Sci.
14 365–71
[121] Hossain T Z, Crawford F D and Tiffin D A 2000 Neutron
detecting semiconductor device USPTO 6,075,261
[122] Anderson L I 1994 Nikola Tesla, Lecture Before the New York
Academy of Sciences, April 6th, 1897 (Breckenridge:
Twenty First Century Books)
23
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[177] Feigl B and Rauch H 1968 Der Gd-neutronenzahler Nucl.
Instrum. Methods 61 349–56
[178] Rose A 1967 Sputtered boron films on silicon surface barrier
detectors Nucl. Instrum. Methods 52 166–70
[179] Birk M, Goldring G and Hillman P 1963 Fast neutron
spectroscopy with solid state detectors Nucl. Instrum.
Methods 21 197–201
[180] Love T A and Murray R B 1964 Fast neutron spectrometer
using spaced semiconductors for measuring total energy of
neutrons captured USPTO 3,129,329
[181] Sun K H 1965 Discriminating radiation detector for
determining a given radiation in the presence of other
radiation USPTO 3,201,590
[182] Wolfe R A and Stubbins W F 1968 A neutron spectrometer
employing charged-particle collimation to improve
resolution Nucl. Instrum. Methods 60 246–52
[183] Griffith R V et al 1979 Recent developments in personnel
neutron dosimeters—a review Health Phys. 36 235–60
[184] Gibson J A B 1985 Personal alarm neutron dosemeters Radiat.
Prot. Dosim. 10 197–205
[185] Griffith R V 1988 Review fo the state of the art in personnel
neutron monitoring with solid state detectors Radiat. Prot.
Dosim. 23 155–60
[186] Lindsay J T, Matsubayashi M and Nurul Islam M 1994 Water
diffusion profile measurements in epoxy using neutron
radiography Nucl. Instrum. Methods Phys. Res. A
353 149–51
[187] Barthe J et al 1993 Personal neutron diode dosemeter Radiat.
Prot. Dosim. 47 397–9
[188] Tanner R J et al 2007 Achievements in workplace neutron
dosimetry in the last decade: lessons learned from the
EVIDOS project Radiat. Prot. Dosim. 126 471–6
[189] Schroder O and Schmitz T 1995 The application of
commercial semiconductor chips for personal neutron
dosimetry Radiat. Prot. Dosim. 61 9–12
[190] Ruddy F H et al 2003 Semiconductor radiation source power
monitor USPTO 6,542,565
[191] De Lurgio P M et al 2003 A neutron detector to monitor the
intensity of transmitted neutrons for small-angle neutron
scattering instruments Nucl. Instrum. Methods Phys. Res. A
505 46–9
[192] Klann R T et al 2000 Development of semiconductor
detectors for fast neutron radiography Proc. 16th Intl. Conf.
Appl. Accl. Res. Ind. (Nov.) pp 1–4
[193] McGregor D S et al 2000 ICONE 8: 8th Intl. Conf. Nuclear
Engineering (Baltimore, MD: ASME) pp 1–8
[194] Sanders J D, Lindsay J T and McGregor D S 2001 Nuclear
Science Symposium Conference Record, IEEE2001
pp 2326–9
[195] Kronenberg S 1990 Dose and dose rate sensor for the pocket
radiac USPTO 4,893,017
[196] Riggan W C 1985 Pulse flux measuring device USPTO
4,516,028
[197] Barthe J, Bordy J M and Lahaye T 1997 Electronic neutron
dosemeters: history and state of the art Radiat. Prot.
Dosim. 70 59–66
[198] Baker C A et al 2002 Development of solid-state silicon
devices as ultra cold neutron detectors Nucl. Instrum.
Methods Phys. Res. A 487 511–20
[199] Monk S D et al 2008 A portable energy-sensitive cosmic
neutron detection instrument Rev. Sci. Instrum. 79 023301
[200] Kozlov S F 1974 Device with a diamond detector for neutron
detection USPTO 3,805,078
[201] Mireshghi A et al 1992 Amorphous silicon position sensitive
neutron detector IEEE Trans. Nucl. Sci. 39 635–40
[202] Lott R G, Ruddy F H and Dulloo A R 1998 Radiation
Resistant Solid State Neutron Detector USPTO 5,726,453
[203] Klann R T and McGregor D S 2002 Coated semiconductor
devices for neutron detection USPTO 6,479,826
[152] Alekseev A G et al 2004 Fast x-ray and neutron detectors
based on natural diamond Instrum. Exp. Tech. 47 153–6
[153] Seidl E and Schwertfuehrer W 1966 Lithium borate single
crystals as neutron detectors Atomkernenergie 11 155–62
[154] Schieber M et al 2007 Composite polycrystalline boron
nitride for alpha and neutron detectors J. Optoelectron. Adv.
Mater. 9 1746–9
[155] Uher J et al 2007 Efficiency of composite boron nitride
neutron detectors in comparison with helium-3 detectors
Appl. Phys. Lett. 90 124101
[156] Schieber M et al 2007 Composite polycrystalline
semiconductor neutron detectors Nucl. Instrum. Methods
Phys. Res. A 579 180–3
[157] Schieber M et al 2006 Polycrystalline BN and LiF based
semiconductor alpha particle and neutron detectors IEEE
Nucl. Sci. Symp. Conf. Rec. R08-1 3728–31
[158] Sangeeta et al 2007 Neutron flux measurements with a
Li2 B4 O7 crystal Nucl. Instrum. Methods Phys. Res. A
571 699–703
[159] Carron N J and Goldflam R 1995 Detector for thermal
neutrons utilizing alternating boron slabs and ccd arrays
USPTO 5,399,863
[160] Krishnamoorthy V and Eiting C J 2008 Boron thin films for
solid state neutron detectors USPTO 2008/0017804
[161] Chapuy S et al 2001 Real-time flat-panel pixel imaging
system and control for x-ray and neutron detection IEEE
Trans. Nucl. Sci. 48 2357–64
[162] Lehmann E and Vontobel P 2004 The use of amorphous
silicon flat panels as detector in neutron imaging Appl.
Radiat. Isotopes 61 567–71
[163] Bruckner G et al 1999 Position sensitive detection of thermal
neutrons with solid state detectors (Gd Si planar detectors)
Nucl. Instrum. Methods Phys. Res. A 424 183–9
[164] Eremin V et al 2004 Nuclear Science Symposium Conference
Record, IEEE2004 vol 3, pp 2003–6
[165] McGregor D S et al 2001 Thin-film-coated bulk GaAs
detectors for thermal and fast neutron measurements Nucl.
Instrum. Methods Phys. Res. A 466 126–41
[166] McGregor D S et al 1992 Development of bulk GaAs room
temperature radiation detectors IEEE Trans. Nucl. Sci.
39 1226–36
[167] Alietti M et al 1995 Performance of a new ohmic contact for
GaAs particle detectors Nucl. Instrum. Methods Phys. Res.
A 362 344–8
[168] McGregor D S and Shultis J K 2004 Spectral identification of
thin-film-coated and solid-form semiconductor neutron
detectors Nucl. Instrum. Methods Phys. Res. A 517 180–8
[169] Hallbeck S et al 2005 Comment on ‘spectral identification of
thin film coated and solid form semiconductor neutron
detectors’ by McGregor and Shultis Nucl. Instrum.
Methods Phys. Res. A 536 228–31
[170] McGregor D S and Shultis J K 2005 Response to the
‘Comment’ by S Hallbeck et al Nucl. Instrum. Methods
Phys. Res. A 536 232–4
[171] Lundstedt C et al 2006 Modeling solid-state boron carbide
low energy neutron detectors Nucl. Instrum. Methods Phys.
Res. A 562 380–8
[172] Gersch H K, McGregor D S and Simpson P A 2000 Nuclear
Science Symposium Conference Record, IEEE2000 vol 1,
pp 4/42–6
[173] Rosenfeld A B et al 1990 P–I–N diodes with a wide
measurement range of fast neutron doses Radiat. Prot.
Dosim. 33 175–8
[174] Moyer J W and Hurwitz H 1956 Neutron flux measuring
device USPTO 2,753,462
[175] Ross B 1966 Neutron detecting solid state device or the like
USPTO 3,227,876
[176] Rauch H, Grass F and Feigl B 1967 Ein neuartiger detektor
fur langsame neutronen Nucl. Instrum. Methods 46 153–6
24
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[204] Nakamura T et al 1989 A real time wide energy range
personal neutron dosimeter with two silicon detectors
Radiat. Prot. Dosim. 27 149–56
[205] Barelaud B et al 1992 Principles of an electronic neutron
dosemeter using a PIPS detector Radiat. Prot. Dosim.
44 363–6
[206] Alberts W G et al 1994 Response of an electronic personal
neutron dosemeter Radiat. Prot. Dosim. 51 207–10
[207] Luszik-Bhadra M, Wendt W and Weierganz M 2004 The
electronic neutron/photon dosemeter PTB DOS-2002
Radiat. Prot. Dosim. 110 291–5
[208] Nunomiya T et al 2007 Development of advanced-type
multi-functional electronic personal dosemeter Radiat.
Prot. Dosim. 126 284–7
[209] Aroua A and Hofert M 1996 Investigation of the PDM-303
neutron pocket dosemeter in various neutron fields Nucl.
Instrum. Methods Phys. Res. 372 318–21
[210] Luszik-Bhadra M et al 1997 Feasibility study of an individual
electronic neutron dosemeter Radiat. Prot. Dosim.
70 97–102
[211] Klann R T and McGregor D S 2000 ICONE 8: 8th Intl. Conf.
Nuclear Engineering (Baltimore, MD: ASME) pp 1–6
[212] Litovchenko P G et al 2004 Semiconductor detectors for
neutron flux measurements Nucl. Instrum. Methods Phys.
Res. A 518 423–5
[213] Filho T M et al 2001 Development of neutron detector using
the surface barrier sensor with polyethylene (n, p) and
10
B(n, α) converters Nucl. Instrum. Methods Phys. Res. A
458 441–7
[214] Jeter T R and Kennison M C 1967 Recent improvements in
helium-3 solid state neutron spectrometry IEEE Trans.
Nucl. Sci. 14 422–7
[215] Foulon F et al 1998 Neutron detector made from chemically
vapour deposited semiconductors Mater. Res. Soc. Symp.
Proc. 487 591–6
[216] Dulloo A R et al 1999 The neutron response of miniature
silicon carbide semiconductor detectors Nucl. Instrum.
Methods Phys. Res. A 422 47–8
[217] Schulte R L, Swanson F and Kesselman M 1994 The use of
large area silicon sensors for thermal neutron detection
Nucl. Instrum. Methods Phys. Res. A 353 123–7
[218] Lu G and Cann G L 1998 CVD diamond radiation detector
USPTO 5,773,830
[219] Kania D R et al 1993 Diamond radiation detectors Diamond
Relat. Mater. 2 1012–9
[220] Bergonzo P et al 1998 High collection efficiency CVD
diamond alpha detectors IEEE Trans. Nucl. Sci. 45 370–3
[221] Foulon F et al 1994 CVD diamond films for radiation
detection IEEE Trans. Nucl. Sci. 41 927–32
[222] Kimblin C et al 2007 STL-20: conducting polymers for
neutron detection OSTI ID:934438
[223] Ha J H et al 2008 A self-biased neutron detector based on an
SiC semiconductor for a harsh environment Appl. Radiat.
Isotopes 67 1204–7
[224] Aoyama T et al 1992 A neutron detector using silicon PIN
photodiodes for personal neutron dosimetry Nucl. Instrum.
Methods Phys. Res. A 314 590–4
[225] Bordy J M et al 1997 Single diode detector for individual
neutron dosimetry using a pulse shape analysis Radiat.
Prot. Dosim. 70 73–8
[226] Zamani M and Savvidis E 1996 A real-time personal neutron
dosemeter based on PIPS diodes and a double layer neutron
detector Radiat. Prot. Dosim. 63 299–303
[227] Fernadez F et al 1997 Separation of the neutron signal from
the gamma component in (n,g) fields using differential
pulse analysis techniques with a double silicon diode
Radiat. Prot. Dosim. 70 87–92
[228] Barthe J et al 1994 IRPA Regional Congr. on Radiological
ProtectionPortsmouth pp 97–100
[229] Paul D et al 1992 Gamma interference on an electronic
dosemeter response in a neutron field Radiat. Prot. Dosim.
44 371–4
[230] Eisen Y et al 1986 A small size neutron and gamma
dosemeter with a single silicon surface barrier detector
Radiat. Prot. Dosim. 15 15–30
[231] Jung M, Teissier C and Siffert P 1998 Theoretical current
response of silicon detectors to fast neutron beams in mixed
n, γ fields Nucl. Instrum. Methods Phys. Res. A
408 503–10
[232] Meijer A C 1971 Neutron spectrometer employing two
semiconductor counter in combination with a coincidence
circuit for eliminating background USPTO 3,581,091
[233] Ndoye A et al 1999 Neutron radiation of an electronic sensor
using coincidence method for an active detector Nucl.
Instrum. Methods Phys. Res. A 423 414–20
[234] Decossas J-L et al 1992 Individual measuring device for the
neutron dose equivalent USPTO 5,083,028
[235] Bluhm H 1974 A γ -discriminating 3 He-semiconductor
sandwich spectrometer Nucl. Instrum. Methods 115 325–37
[236] Pospisil S et al 1993 Si diode as a small detector of slow
neutrons Radiat. Prot. Dosim. 46 115–8
[237] Campbell M et al 1997 Silicon pixel devices as a slow neutron
precise position detector Nucl. Instrum. Methods Phys. Res.
A 395 457–8
[238] Schelten J et al 1997 Position-sensitive neutron detection with
6
LiF layers on silicon semiconductors Nucl. Instrum.
Methods Phys. Res. A 389 447–53
[239] Seidel J G et al 1999 Solid state neutron detector array
USPTO 5,940,460
[240] Petrillo C et al 1999 Space resolution of a Si/Gd microstrip as
linear position-sensitive detector for thermal neutrons Nucl.
Instrum. Methods Phys. Res. A 424 523–32
[241] Uher J et al 2005 Performance of a pixel detector suited for
slow neutrons Nucl. Instrum. Methods Phys. Res. A
542 283–7
[242] McGregor D S et al 1996 Semi-insulating bulk GaAs thermal
neutron imaging arrays IEEE Trans. Nucl. Sci. 43 395–9
[243] McGregor D S et al 1996 Semi-insulating bulk GaAs as a
semiconductor thermal-neutron imaging device Nucl.
Instrum. Methods Phys. Res. A 380 271–5
[244] Bruckner G and Rauch H 1996 A position sensitive Gd
detector for thermal neutrons J. Neutron Res. 4 141–7
[245] Mireshghi A et al 1994 High efficiency neutron sensitive
amorphous silicon pixel detectors IEEE Trans. Nucl. Sci.
41 915–21
[246] Eisen Y et al 1983 A combined real time wide energy range
neutron dosimeter and survey meter for high neutron dose
rates with Si surface barrier detectors Nucl. Instrum.
Methods Phys. Res. 211 171–8
[247] Albrecht P and Schneider S 1994 Advanced neutron detector
USPTO 5,281,822
[248] Flynn E R and Bryant H C 1966 High efficiency solid state
neutron telescope for 15–35 MeV neutrons Rev. Sci.
Instrum. 37 215–23
[249] Bishop G B 1968 A 6 Li sandwich semiconductor detector for
fast neutron spectrum measurements Nucl. Instrum.
Methods 62 247–52
[250] Silk M G 1968 The determination of fast-neutron spectra in
thermal reactors by means of a high-resolution
semiconductor spectrometer Nucl. Instrum. Methods
66 93–101
[251] Maroni C, Russo F and Verondini E 1969 A new type of
neutron spectrometer in the energy range 1–100 keV Nucl.
Instrum. Methods 74 256–60
[252] Cambiaghi M, Fossati F and Pinelli T 1970 Fast neutron
spectrometry with identification of recoil protons Nucl.
Instrum. Methods 82 106–8
25
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[278] Unruh T C et al 2009 Design and operation of a 2-D thin-film
semiconductor neutron detector array for use as a beamport
monitor Nucl. Instrum. Methods Phys. Res. A 604 150–3
[279] Solomon C J et al 2007 A hybrid method for coupled
neutron-ion transport calculations for 10 B and 6 LiF coated
and perforated detector efficiencies Nucl. Instrum. Methods
Phys. Res. A 580 326–30
[280] McGregor D S et al 2000 Self-biased boron-10 coated
high-purity epitaxial GaAs thermal neutron detectors IEEE
Trans. Nucl. Sci. 47 1364–70
[281] Bellinger S L et al 2007 Nuclear Science Symposium
Conference Record, NSS ’07. IEEE2007 pp 1904–7
[282] Bellinger S L, McNeil W J and McGregor D S 2009 Improved
fabrication technique for microstructured solid-state
neutron detectors Mater. Res. Soc. Symp. Proc. 1164 L06–1
[283] Conway A M et al 2009 Numerical simulations of pillar
structured solid state thermal neutron detector: Efficiency
and gamma discrimination IEEE Trans. Nucl. Sci.
56 2802–7
[284] Nikolic R J et al 2006 Nanotech Boston, MA, p 4
[285] Siegmund O H et al 2007 High spatial resolution neutron
sensing microchannel plate detectors Nucl. Instrum.
Methods Phys. Res. A 576 178–82
[286] LiCausi N et al 2008 Hard X-ray, Gamma-Ray, and Neutron
Detector Physics X (Proceedings of the SPIE) ed A Burger,
L A Franks and R B James (Bellingham, WA: SPIE Optical
Engineering Press) pp 707908-1–11
[287] Schelten J and Reinartz R 1999 Neutron detector USPTO
5,880,471
[288] McGregor D and Klann R 2003 Pocked surface neutron
detector USPTO 6,545,281
[289] McGregor D S and Klann R 2007 High-efficiency neutron
detectors and methods of making same USPTO 7,164,138
[290] Schelten J et al 1997 A new neutron detector development
based on silicon semiconductor and 6 LiF converter Physica
B 234–236 1084–6
[291] McNeil W J et al 2009 1-D array of perforated diode neutron
detectors Nucl. Instrum. Methods Phys. Res. A 604 127–9
[292] Nikolic R J et al 2005 Roadmap for high efficiency solid-state
neutron detectors Proc. SPIE 6013 601305
[293] McNeil W J et al 2007 Nuclear Science Symposium
Conference Record, NSS ’07. IEEE2007 pp 2340–2
[294] McGregor D S et al 2002 New surface morphology for low
stress thin-film-coated thermal neutron detectors IEEE
Trans. Nucl. Sci. 49 1999–2004
[295] Henderson C M et al 2010 Characterization of prototype
perforated semiconductor neutron detectors Radiat. Phys.
Chem. 79 144–50
[296] McGregor D S et al 2009 Microstructured semiconductor
neutron detectors Nucl. Instrum. Methods Phys. Res. A
608 125–31
[297] McGregor D S et al 2009 Perforated diode neutron detector
modules fabricated from high-purity silicon Radiat. Phys.
Chem. 78 874–81
[298] Jahan Q et al 2007 Neutron dosimeters employing
high-efficiency perforated semiconductor detectors Nucl.
Instrum. Methods Phys. Res. B 263 183–5
[299] McGregor D S et al 2009 Non-streaming high-efficiency
perforated semiconductor neutron detectors, methods of
making same and measuring wand and detector modules
utilizing same USPTO 2009/0302231
[300] McGregor D S et al 2002 X-Ray and Gamma Ray Detectors
and Applications IV ed R B James (Bellingham, WA: SPIE
Optical Engineering Press) pp 164–82
[301] Nikolic R J et al 2008 Int. Conf. Solid-State and
Integrated-Circuit Technology (Beijing, Ching)
[302] Nikolic R J et al 2008 6:1 aspect ratio silicon pillar based
thermal neutron detector filled with 10 B Appl. Phys. Lett.
93 133502
[303] Deo N et al 2008 Conformal filling of silicon micropillar
platform with 10boron J. Vac. Sci. Technol. B 26 1309–14
[253] Rickard I C 1973 Neutron spectroscopy in the energy range
10–500 keV using the 6 Li sandwich detector Nucl. Instrum.
Methods 113 169–74
[254] Clements P J 1975 The calibration of 6 Li semiconductor
sandwich spectrometers Nucl. Instrum. Methods 127 61–72
[255] Ryves T B 1976 A proton recoil telescope for 12–20 MeV
neutrons Nucl. Instrum. Methods 135 455–8
[256] Pinelli T et al 1978 A 6 Li semiconductor sandwich neutron
spectrometer with background discrimination by
identification of reaction products Nucl. Instrum. Methods
150 497–503
[257] Shiraishi F et al 1985 A new fast neutron spectrometer made
of epitaxial integrated dE-E Si detector IEEE Trans. Nucl.
Sci. 32 471–5
[258] Mao S S and Perry D L 2010 Layered semiconductor neutron
detectors WIPO WO/2010/011859
[259] Kitaguchi H, Shigeru I and Kaihara A 1994 Neutron
individual dose meter, neutron dose rate meter, neutron
detector and its method of manufacture USPTO 5,321,269
[260] Sasaki M et al 1998 Development and characterization of
real-time personal neutron dosemeter with two silicon
detectors Nucl. Instrum. Methods Phys. Res. A 418 465–75
[261] Philips B F et al 2007 Neutron detection using large area
silicon detectors Nucl. Instrum. Methods Phys. Res. A
579 173–6
[262] Vareille J C et al 1997 Advanced detectors for active neutron
dosemeters Radiat. Prot. Dosim. 70 79–82
[263] Dubarry B et al 1992 Electronic sensor response in neutron
beams Radiat. Prot. Dosim. 44 367–70
[264] Kitaguchi H et al 1996 Silicon semiconductor detectors for
various nuclear radiations IEEE Trans. Nucl. Sci.
43 1846–50
[265] Shiraishi F et al 1988 A new type of personnel neutron
dosimeter with thin Si detectors IEEE Trans. Nucl. Sci.
35 575–8
[266] Kress K A 1983 Solid state neutron detector USPTO
4,419,578
[267] Hendricks J S 1994 A Monte Carlo code for particle transport:
an algorithm for all seasons Los Alamos Sci. 22 30–43
[268] Giani S 1998 GEANT4: an object-oriented toolkit for
simulation in HEP CERN/LHCC 98–44 pp 1–6
[269] Tanner J E et al 1995 An assessment of the feasibility of using
Monte Carlo calculations to model a combined
neutron/gamma electronic personal dosemeter Radiat. Prot.
Dosim. 61 183–6
[270] Tremsin A S, Feller W B and Downing R G 2005 Efficiency
Optimization of microchannelplate (MCP) neutron imaging
detectors. I. Square channels with 10 B doping Nucl.
Instrum. Methods Phys. Res. A 539 278–311
[271] Shultis J K and McGregor D S 2006 Efficiencies of coated and
perforated semiconductor neutron detectors IEEE Trans.
Nucl. Sci. 53 1659–65
[272] Shultis J K et al 2007 Calculation of ion energy-deposition
spectra in silicon, lithium-fluoride, boron, and boron
carbide KSU Experimental Station Report 299
[273] Grupen C, Shwartz B A and Spieler H 2008 Particle Detectors
(Cambridge: Cambridge University Press)
[274] Rinard P 1991 Passive Nondestructive Assay of Nuclear
Materials ed D Reilly, N Ensslin and H Smith (Washington,
DC: Nuclear Regulatory Commission) pp 357–77
[275] McGregor D S et al 2001 Designs for thin-film-coated
semiconductor thermal neutron detectors IEEE Nucl. Sci.
Symp. Conf. Rec. 4 2454–8
[276] Harken A D et al 2005 Boron-rich semiconducting boron
carbide neutron detector Japan. J. Appl. Phys. 44 444–5
[277] Schulte R L and Kesselman M 1999 Development of a
portable directional thermal neutron detection system for
nuclear monitoring Nucl. Instrum. Methods Phys. Res. A
422 852–7
26
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[304] Nikolic R J et al 2007 Fabrication of pillar-structured thermal
neutron detectors IEEE Nucl. Sci. Symp. Conf. Rec.
N24-342 1577–80
[305] Ellis R and Wankling J L 1972 Radiation detector comprising
semi-conductor body incorporating a two-dimensional
array of p–i–n devices USPTO 3,691,389
[306] Muminov R A and Tsvang L D 1987 High-efficiency
semiconductor thermal-neutron detector At. Energy
62 316–9
[307] Uher J et al 2007 Characterization of 3D thermal neutron
semiconductor detectors Nucl. Instrum. Methods Phys. Res.
A 576 32–7
[308] Uher J et al 2007 Nuclear Physics Methods and Accelerators
in Biology and Medicine ed C Granja, C Leroy, and
I Stekl pp 101–4
[309] Suzuki A 1998 High-efficiency silicon space solar cells Sol.
Energy Mater. Sol. Cells 50 289–303
[310] Linhart V et al 2006 Measurements of fast-neutron-induced
signals in silicon pad detectors Nucl. Instrum. Methods
Phys. Res. A 563 263–7
[311] Jakubek J et al 2006 Neutron imaging with Medipix-2 chip
and coated sensor Nucl. Instrum. Methods Phys. Res. A
560 143–7
[312] Jakubek J et al 2005 Spatial resolution of Medipix-2 device as
neutron pixel detector Nucl. Instrum. Methods Phys. Res. A
546 164–9
[313] Jakubek J et al 2004 Properties of the single neutron pixel
detector based on the Medipix-1 device Nucl. Instrum.
Methods Phys. Res. A 531 276–84
[314] Greiffenberg D et al 2009 Detection efficiency of
ATLAS-MPX detectors with respect to neutrons Nucl.
Instrum. Methods Phys. Res. A 607 38–40
[315] Holy T et al 2004 Astroparticle, Particle and Space Physics,
Detectors and Medical Physics Applications ed
M Barone (Como: World Scientific) pp 374–80
[316] Uher J et al 2008 Detection of fast neutrons with the
Medipix-2 pixel detector Nucl. Instrum. Methods Phys.
Res. A 591 71–4
[317] Holy T et al 2008 Pattern recognition of tracks induced by
individual quanta of ionizing radiation in Medipix2 silicon
detector Nucl. Instrum. Methods Phys. Res. A 591 287–90
[318] Jakubek J et al 2008 Pixel detectors for imaging with heavy
charged particles Nucl. Instrum. Methods Phys. Res. A
591 155–8
[319] Da Via C et al 2008 Radiation hardness properties of full-3D
active edge silicon sensors Nucl. Instrum. Methods Phys.
Res. A 587 243–9
[320] Fraser G W 1995 Neutrons and their Applications
(Bellingham, WA: SPIE Optical Engineering Press)
pp 287–301
[321] Feller W B, Downing R G and White P L 2000 Hard X-Ray,
Gamma-Ray, and Neutron Detector Physics II
ed R B James and R C Schirato (Bellingham, WA: SPIE
Optical Engineering Press) pp 291–302
[322] Tremsin A S et al 2004 Very compact high performance
microchannel plate thermal neutron collimators IEEE
Trans. Nucl. Sci. 51 1020–4
[323] Ambrosi R M et al 2003 Large area microchannel plate
detector with amorphous silicon pixel array readout for fast
neutron radiography Nucl. Instrum. Methods Phys. Res. A
500 351–61
[324] Tremsin A S and Feller W B 2006 The theory of compact and
efficient circular-pore MCP neutron collimators Nucl.
Instrum. Methods Phys. Res. A 556 556–64
[325] Tremsin A S et al 2008 On the possibility to image thermal
and cold neutron with sub-15 μm spatial resolution Nucl.
Instrum. Methods Phys. Res. A 592 374–84
[326] Tremsin A S, Feller W B and Siegmund O H W 2008 A new
concept of thermal neutron counting with sub-microsecond
timing resolution IEEE Trans. Nucl. Sci. 53 1664–9
[327] Tremsin A S et al 2009 Detection efficiency, spatial and timing
resolution of thermal and cold neutron counting MCP
detectors Nucl. Instrum. Methods Phys. Res. A 604 140–3
[328] Fraser G W and Pearson J F 1990 The direct detection of
thermal neutrons by imaging microchannel-plate detectors
Nucl. Instrum. Methods Phys. Res. A 293 569–74
[329] Rougeot H 1978 Solid detector for ionizing radiation USPTO
4,119,852
[330] Dearnealy G and Whitehead A B 1961 The semiconductor
surface barrier for nuclear particle detection Nucl. Instrum.
Methods 12 205–26
[331] Deuchars W M and Lawrence G P 1961 Interaction of
14-MeV neutrons with a silicon semiconductor nuclear
particle detector Nature 191 995
[332] Dearnealy G, Ferguson A T G and Morrison G C 1962
Semiconductor fast neutron detector I. R. E. Nucl. Sci.
9 174–80
[333] Marcazzan M G, Merzari F and Tonolini F 1962 Application
of silicon detectors to measurements of monoenergetic
neutron beams Phys. Lett. 1 21–2
[334] Mainsbridge B, Bonner T W and Rabson T A 1963 The
disintegration of silicon by fast neutrons Nucl. Phys.
48 83–9
[335] Miller R G and Kavanagh R W 1967 Semiconductor detector
as a fast neutron spectrometer Nucl. Instrum. Methods
48 13–27
[336] Mingay D W, Sellschop J P F and Johnson P M 1971 Neutron
induced reactions in silicon semiconductor detectors Nucl.
Instrum. Methods 94 497–507
[337] McKenzie J M and Waugh J B S 1960 Silicon junctions as
particle spectrometers I. R. E. Nucl. Sci. 7 195–9
[338] Mayer W and Gossick B 1956 Use of Au–Ge broad area
barrier as alpha particle spectrometer Rev. Sci. Instrum.
27 407–8
[339] Blankenship J L and Borkowski C J 1961 Performance of
silicon surface barrier detectors with charge sensitive
amplifiers I. R. E. Nucl. Sci. 8 17–20
[340] Elevant T et al 1986 Silicon surface barrier detector for fusion
neutron spectroscopy Rev. Sci. Instrum. 57 1763–5
[341] Bosch H-S et al 1986 Calibration of a surface barrier detector
for 14-MeV neutron flux measurements on TFTR Rev. Sci.
Instrum. 59 1718–20
[342] Ruskov E et al 1995 Measurement of 14 MeV neutrons at
TFTR with Si-diode detectors Rev. Sci. Instrum. 66 910–2
[343] Zhou C, Zhao J and Xiao W 2005 PIN silicon diode fast
neutron detector Radiat. Prot. Dosim. 117 365–8
[344] Matsumoto T 1991 PIN diode for real time dosimetery in a
mixed field of neutrons and gamma rays Radiat. Prot.
Dosim. 35 193–7
[345] Friedland S S, Mayer J W and Wiggins J S 1962
Discriminating radiation detector USPTO 3,043,955
[346] Reinitz K 1998 Neutron detector based on semiconductor
materials USPTO 5,707,879
[347] Sato N, Suzuki T and Ishiwata O 1991 Semiconductor-based
radiation-detector element USPTO 5,019,886
[348] Feldman L C et al 1986 High resistivity group III–V
compounds by helium bombardment USPTO H147
[349] Mayer J W 1965 Method of measuring nuclear radiation
utilizing a semiconductor crystal having a lithium
compensated intrinsic region USPTO 3,225,198
[350] Brackenbush L W and Quam W 1985 Hydrogenous
semiconductor neutron detectors Radiat. Prot. Dosim. 10
303–5
[351] Selph W E 1984 Solid state fast-neutron
spectrometer/dosimeter and detector therefor USPTO
4,445,036
[352] Baum R 1987 Chem. Eng. News 65 24
[353] Shinar J 2004 Organic Light-Emitting Devices: A Survey
(New York: Springer)
27
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[382] Bourdeau R G 1967 Process of preparing boron carbide from
boron halide and a hydrocarbon USPTO 3,334,967
[383] Heestand R L and Leitten C F 1968 Boron carbide article and
method of making USPTO 3,367,826
[384] Krochmal J J, Shapiro I and Lynch C T 1968 Preparation of
films of boron carbide USPTO 3,398,013
[385] Reeves R B and Gebhardt J J 1970 Low temperature method
for producing amorphous boron-carbon deposits USPTO
3,537,877
[386] Thevenot F 1990 Boron carbide—a comprehensive review
J. Eur. Ceram. Soc. 6 205–25
[387] Emin D et al 1985 Boron-Rich Solids vol 140 (New York:
AIP)
[388] Emin D et al 1990 Boron-Rich Solids vol 231 (New York:
AIP)
[389] Sezer A O and Brand J I 2001 Chemical vapor deposition of
boron carbide Mater. Sci. Eng. B 79 191–202
[390] Robertson B W et al 2002 A class of boron-rich solid-state
neutron detectors Appl. Phys. Lett. 80 3644–6
[391] Robertson B W et al 2002 Advances in Neutron Scattering
Instrumentation (Proceedings of SPIE) ed I S Anderson
and B Guerard, (Bellingham, WA: SPIE Optical
Engineering Press) pp 226–33
[392] Adenwalla S et al 2003 Penetrating Radiation Systems and
Applications V (Proceedings of SPIE) ed F P Doty,
H B Barber and H Roehrig (Bellingham, WA: SPIE Optical
Engineering Press) pp 70–4
[393] Dowben P A et al 2004 Boron-carbide solid state neutron
detector and method of using the same USPTO 6,771,730
[394] Kim Y-G, Dowben P A and Spencer J T 1989 Chemical vapor
deposition of boron and boron nitride from decaborane (14)
J. Vac. Sci. Technol. A 7 2796–9
[395] Zhang Z et al 1989 Chemical Perspectives of Microelectronic
Materials ed M Gross, J M Jasinski and J Yates, Materials
Research Society, pp 407–12
[396] Mazurowski J et al 1991 Plasma Processing and Synthesis of
Material III ed D Apelian and J Szekely, Materials
Research Society, pp 101–6
[397] Mazurowski J et al 1992 Wide Bandgap Semiconductors
ed T D Moustakas, J I Pankove and Y Hamakawa Materials
Research Society, pp 637–42
[398] Lee S et al 1992 Characterization of boron carbide thin films
fabricated by PECVD from boranes J. Appl. Phys.
72 4925–33
[399] Lee S et al 1993 The structural homogeneity of boron carbide
thin films fabricated using plasma enhanced chemical vapor
deposition from B5 H9 + CH4 J. Appl. Phys. 74 6919–24
[400] Lee S and Dowben P A 1994 The properties of boron
carbide/silicon heterojunction diodes fabricated by
plasma-enhanced chemical vapor deposition Appl. Phys. A
58 223–7
[401] Byun D et al 1995 Comparison of different chemical vapor
deposition methodologies for the fabrication of
heterojunction boron-carbide diodes Nanostruct. Mater.
5 465–73
[402] Spencer J T, Dowben P A and Kim Y G 1990 Deposition of
boron-containing films from decaborane USPTO 4,957,773
[403] Dowben P A 1995 Forming B1−x Cx semiconductor devices by
chemical vapor deposition USPTO 5,468,978
[404] Hwang S-D et al 1996 Fabrication of boron carbide–boron
heterojunction devices Appl. Phys. Lett. 68 1495–7
[405] Hwang S-D et al 1996 Nickel doping of boron carbide grown
by plasma enhanced chemical vapor deposition J. Vac. Sci.
Technol. B 14 2957–60
[406] Ahmad A A et al 1996 Optical properties of boron carbide
(B5 C) think films fabricated by plasma enhanced chemical
vapor deposition J. Appl. Phys. 79 8643–7
[407] Hwang S-D et al 1997 Fabrication of n-type nickel doped
B5 C1+δ homojunction and heterojunction diodes Appl.
Phys. Lett. 70 1028–30
[354] Brabec C, Dyakonov V and Scherf U 2008 Organic
Photovoltaics (Weinheim: Wiley–VCH)
[355] Skotheim T A and Reynolds J 2007 Handbook of Conducting
Polymers (Boca Raton, FL: CRC Press)
[356] Ledoux G et al 2004 Synthesis and optical characterization of
Gd2 O3 :Eu3+ nanocrystals: surface states and VUV
excitation Radiat. Measur. 38 763–6
[357] von Schnering H G et al 1980 Li12 Si7 , a compound having a
trigonal planar Si4 cluster and planar Si5 rings Angew.
Chem. Int. Edn Engl. 19 1033–4
[358] Bohm M C et al 1984 Tight-binding approach to the
solid-state structure fo the complex Zintl-phase Li12 Si7
Phys. Rev. B 30 4870–3
[359] Kuriyama K, Maeda T and Mizuno A 1988 Anomalous
electrical resistivity and photovoltaic phenomenon in the
fast-mixed conductor lithium silicide Li12 Si7 Phys. Rev. B
34 13436–8
[360] Stearns L A et al 2003 Lithium monosilicide (LiSi), a
low-dimensional silicon-based material prepared by high
pressure synthesis: NMR and vibrational spectroscopy and
electrical properties characterization J. Solid State Chem.
173 251–8
[361] Smith C J and Lowe C W 1989 Stoichiometric effects on the
optical properties of LiInSe2 J. Appl. Phys. 66 5102–4
[362] Kamijoh T and Kuriyama K 1980 Blue-band emission in
LiInS2 crystals J. Appl. Phys. 51 1827–8
[363] Bacewicz R and Ciszek T F 1988 Preparation and
characterization of some AI BII CV type semiconductors
Appl. Phys. Lett. 52 1150–1
[364] Kuriyama K and Nakamura F 1987 Electrical transport
properties and crystal structure of LiZnAs Phys. Rev. B
36 4439–41
[365] Ramirez R et al 1985 The solid-state electronic structure and
the nature of the chemical bonds of the ternary Zintl-phases
Li8 MgSi6 Chem. Phys. 95 17–35
[366] Belin C H and Tillard M 2006 Inorganic Chemistry in Focus
III ed G Meyer, D Naumann and L Wesemann (Weinheim:
Wiley–VCH)
[367] Christensen N E 1985 Structural phase stability of B2 and B32
intermetallic compounds Phys. Rev. B 32 207–10
[368] Rosner H and Pickett W E 2003 Theoretical investigation of
stoichiometry lithium monoboride Phys. Rev. B 67 054104
[369] Kolmogorov A N and Curtarolo S 2006 Prediction of different
crystal structure phases in metal borides: a lithium
monoboride analog to MgB2 Phys. Rev. B 73 180501R–4R
[370] Mair G and Von Schnering H G 1999 Dilithium Hexaboride,
Li2 B6 Z. Anorg. Allg. 625 1207–11
[371] Mair G, Nesper R and von Schnering H G 1987 Trilithium
tetradecaboride Li3 B14 : synthesis, structure, and properties
J. Solid State Chem. 75 30–40
[372] Borgstedt H B 2003 The B–Li (boron–lithium) system J.
Phase Equilib. 24 572–4
[373] Okamoto H 1989 The B–Li (boron–lithium) system Bull.
Alloy Phase Diagr. 10 230–2
[374] Welker H 1957 Semiconductor devices and methods of their
manufacture USPTO 2,789,989
[375] Welker H and Gremmelmaier R 1959 Method and device for
the sensing of neutrons USPTO 2,867,727
[376] Welker H and Gremmeimaier R 1961 Method and device for
sensing neutrons USPTO 2,988,639
[377] Hill D E and Epstein A 1964 Thermoelectricity USPTO
3,138,486
[378] Werheit H 2009 16th Int. Symp. Boron, Borides and Related
Materials IOP pp 012019–29
[379] Henderson C M and Harris D M 1963 Thermoelectricity
USPTO 3,087,002
[380] Bloom J L 1965 Thermal neutron flux measuring device using
two boron thermistors USPTO 3,226,547
[381] Gaule G K, Ross R L and Bloom J L 1964 Boron
ed G K Gaule (New York: Plenum) p 317
28
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[408] Hwang S-D et al 1997 Nickel doping of boron carbide and
Fermi level shifts J. Vac. Sci. Technol. A 15 854–8
[409] Dowben P A 1997 Forming B1−x Cx semiconductor layers by
chemical vapor deposition USPTO 5,658,834
[410] McIlroy D N et al 1998 The incorporation of nickel and
phosphorous dopants into boron carbon alloy thin films
Appl. Phys. A 67 335–42
[411] Ahmad A A et al 1998 Sputter deposition of high resistivity
boron carbide Thin Solid Films 335 174–7
[412] Dowben P A 2000 Boron-carbide and boron rich
rhombohedral based transistors and tunnel diodes USPTO
6,025,611
[413] Adenwalla S et al 2001 Boron carbide/n-silicon carbide
heterojunction diodes Appl. Phys. Lett. 79 4357–9
[414] Abdul-Gader M M et al 2001 Low-field current transport
mechanisms in rf magnetron sputter deposited boron
carbide (B5 C/p-type crystalline silicon junctions in the
dark Int. J. Electron. 88 873–901
[415] Dowben P A 2002 Boron-carbide and boron rich
rhombohedral based transistors and tunnel diodes USPTO
6,440,786
[416] Dowben P A 2003 Boron-carbide and boron rich rhomboderal
based transistors and tunnel diodes USPTO 6,600,177
[417] Caruso A N et al 2004 The heteroisomeric diode J. Phys.:
Condens. Matter 16 L139–L46
[418] Caruso A N et al 2006 The all boron carbide diode neutron
detector: comparison with theory Mater. Sci. Eng. B
135 129–33
[419] Day E E 2006 Boron Carbide Devices for Neutron Detection
Applications Engineering (Lincoln: University of
Nebraska)
[420] Harken A A 2006 Modeling and Signal Analysis of
Semiconducting B5 C Neutron Detectors Engineering
(Lincoln: University of Nebraska)
[421] Osberg K et al 2006 A handheld neutron-detection sensor
system utilizing a new class of boron carbide diode IEEE
Sensors J. 6 1531–8
[422] Day E, Diaz M J and Adenwalla S 2006 Effect of bias on
neutron detection in thin semiconducting boron carbide
films J. Phys. D: Appl. Phys. 39 2920–4
[423] Harken A D and Robertson B W 2006 Comparative modelling
of performance limits of solid-state neutron detectors based
on planar B-rich capture layers J. Phys. D: Appl. Phys.
39 4961–8
[424] Stone B and Hill D 1960 Semiconducting properties of cubic
boron phosphide Phys. Rev. Lett. 4 282–4
[425] Mead C A 1966 Metal-semiconductor surface barriers
Solid-State Electron. 9 1023–33
[426] Takenaka T, Takigawa M and Shohno K 1976 Dielectric
constant and refractive index of boron monophosphide
Japan. J. Appl. Phys. 15 2021–2
[427] Kumashiro Y and Okada Y 1985 Schottky barrier diodes
using thick, well-characterized boron phosphide wafers
Appl. Phys. Lett. 47 64–6
[428] Lund J C et al 1990 Boron phosphide on silicon for radiation
detectors Mater. Res. Soc. Symp. Proc. 162 601–8
[429] Kumashiro Y, Hirabayashi M and Takagi S 1990 Boron
phosphide as a refractor semiconductor Mater. Res. Soc.
Symp. Proc. 162 585–91
[430] Kumashiro Y 1990 Thermophysical properties of boron
phosphide single crystalline wafers Rigaku J. 7 21–6
[431] Kumashiro Y 1990 Refractory semiconductor of boron
phosphide J. Mater. Res. 5 2933–47
[432] Goossens A and Schoonman J 1995 Electrochemical
investigations of silicon/boron phosphide heterojunction
photoelectrodes Electrochem. Acta 40 1339–44
[433] Wang S H et al 2005 Cr/Pt ohmic contacts to B12 As2 Appl.
Phys. Lett. 87 042103
[434] Dalui S and Pal A K 2008 Synthesis of n-type boron
phosphide films and formation of Schottky diode:
Aln-BP/Sb Appl. Surf. Sci. 254 3540–7
[435] Hong N et al 2010 Boron carbide based solid state neutron
detectors: the effects of bias and time constant on detection
efficiency J. Phys. D: Appl. Phys. 43 275101
[436] Mazurowski J, Lee S and Dowben P A 1992 Deposition and
characterization of boron carbide: carriers and mobility
ISHM Proc. 1847 112–6
[437] Aselage T et al 1994 Structure and electronic transport of
undoped and phosphorous-doped boron carbide Japan. J.
Appl. Phys. 10 58–61
[438] Werheit H et al 1991 Electronic transport properties of boron
carbide within the homogeneity range AIP Conf. Proc.
231 104–7
[439] Schemechel R and Werheit H 1998 Dynamical transport in
icosahedral boron-rich solids J. Mater. Proc. Manufact. Sci.
6 329–37
[440] Bandyopadhyay A K et al 1984 The role of free carbon in the
transport and magnetic properties of boron carbide J. Phys.
Chem. Solids 45 207–14
[441] Rybakov S Y, Sharapov V M and Gavrilov L E 1995 Oxygen
effect on the stability of PECVD boron-carbide films
J. Physique IV 5 921–6
[442] Hong N et al 2010 Ni doping of semiconducting boron
carbide J. Appl. Phys. 107 024513
[443] Schulz D L et al 2008 Characterization of a-B5C:H prepared
by PECVD of orthocarborane: results of preliminary FTIR
and nuclear reaction analysis studies J. Non-Cryst. Solids
354 2369–71
[444] Billa R B et al 2009 Annealing effects on the optical
properties of semiconducting boron carbide J. Appl. Phys.
106 033515
[445] Kocsis M and Major J 2002 Advances in Neutron Scattering
Instrumentation (Proceedings of SPIE) ed
I S Anderson and B Guerard, (Bellingham, WA: SPIE
Optical Engineering Press) pp 234–42
[446] Cook B A et al 2008 AlMgB14 and related icosahedral boride
semiconducting materials for neutron sensing applications
USPTO 7,375,343
[447] Ali M A et al 1994 Properties of the 158 Gd compound state
gamma-decay cascades J. Phys. G: Nucl. Part. Phys.
20 1943–53
[448] Sakurai Y and Kobayashi T 2002 Experimental verification fo
the nuclear data of gadolinium for neutron capture therapy
J. Nucl. Sci. Technol., Suppl. 2 1294–7
[449] Paolo G 2002 The 2002 international technology roadmap for
semiconductors Proc. Electrochem. Soc. 1 5–19
[450] Kang S H 2008 Recent advances in spintronics for emerging
memory devices J. Miner. Met. Mater. Soc. 60 28–33
[451] Itoh H et al 1987 Synthesis of cerium and gadolinium borides
using boron cage compounds as a boron source Mater. Res.
Soc. Bull. 22 1259–66
[452] Kher S S, Yexin T and Spencer J T 1996 Chemical vapor
deposition of metal borides, 4: the application of
polyhedral boron clusters to the chemical vapor deposition
formation of gadolinium boride thin-film materials Appl.
Organomet. Chem. 10 297–304
[453] Dowben P A, Zhong Z and Sellmyer D J 2000 Method for
depositing rare-earth boride onto a substrate USPTO
6,025,038
[454] Chung K B et al 2003 Phase selective synthesis of gadolinium
silicide films on Si(111) using an interfacial SiO2 layer
J. Appl. Phys. 94 212–5
[455] Babizhetskyy V et al 2004 Gd5 Si2 B8 : a ternary
rare-earth-metal silicide boride compound Solid-State
Chem. 116 2013–7
[456] Chang W H et al 2009 Nanometer-thick single crystal
hexagonal Gd2 O3 on GaN for advanced complementary
metal-oxide-semiconductor technology Adv. Mater.
21 4970–4
[457] Tanaka T et al 1980 The preparation and properties of CeB6 ,
SmB6 , and GdB6 J. Appl. Phys. 51 3877–83
29
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[482] Meek T T, von Roedern B G and Haire M J 2003 The effect of
Co-doping on the photoelectric properties of UO2 Trans.
Am. Nucl. Soc. 89 905–8
[483] Meek T T and Hu M Z 2004 Fabricating electronic devices
using actinide oxide semiconductor materials USPTO
6,730,538
[484] Bates J L, Hinman C A and Kawada T 2006 Electrical
conductivity of uranium dioxide J. Am. Ceram. Soc.
50 652–6
[485] Meek T T and von Roedern B G 2009 Semiconductor devices
fabricated from actinide oxides Vacuum 83 226–8
[486] Cox L E et al 1987 Valence-band photoemission in UO2 (111)
near the 5d resonant photon energy Phys. Rev. B 35 5761–5
[487] Kelly P J and Brooks M S S 1987 Electronic structure and
ground-state properties of the actinide dioxides J. Chem.
Soc., Faraday Trans. 2 1189–203
[488] Yun Y et al 2002 Electronic state of uranium dioxide
J. Korean Nucl. Soc. 34 202–10
[489] Roy L E et al 2008 Dispersion in the Mott insulator UO2 : a
comparison of photoemission spectroscopy and screen
hybrid density functional theory J. Comput. Chem.
29 2288–94
[490] Winter P W 1989 The electronic transport properties of UO2
J. Nucl. Mater. 161 38–43
[491] Killeen J C 1980 The effect of niobium oxide additions on the
electrical conductivity of UO2 J. Nucl. Mater. 88 185–92
[492] Casado J M, Harding J H and Hyland G J 1994 Small-polaron
hopping in Mott-insulating UO2 J. Phys.: Condens. Matter
6 4685–98
[493] Ruello P et al 2004 Thermal variation of the optical
absorption of UO2 : determination of the small polaron
self-energy J. Nucl. Mater. 328 46–54
[494] Myers H P, Jonsson T and Westin R 1964 Intrinsic
semiconductors in uranium dioxide Solid State Commun.
2 321–2
[495] Iida S 1966 Photocurrent and thermal release time studies of
uranium dioxide Japan. J. Appl. Phys. 5 752–5
[496] Young R A 1979 Model for the electronic contribution to the
thermal and transport properties of ThO2 , UO2 , and PuO2
in the solid and liquid phases J. Nucl. Mater. 87 283–96
[497] Schoenes J 1980 Electronic transitions, crystal field effects
and phonon in UO2 Phys. Rep. 63 301–36
[498] Killeen J C 1980 The measurement of the electron-to-hole
mobility ratio in UO2 and its effect on the thermal
conductivity J. Nucl. Mater. 92 136–40
[499] Sorriaux A 1970 Evaluation of forbidden band and the
acceptor energy levels in natural uranium dioxide C. R.
Seances Acad. Sci. B 270 77–9
[500] Sorriaux A and Djerassi H 1971 Electrical conductivity and
thermoelectric power of uranium dioxide UO2+x a high
temperatures C. R. Seances Acad. Sci. B 272 1373–6
[501] Cojocaru L N 1969 Influence of the nuclear radiation on the
electrical conductivity and thermo-electric power of the
UO2 –SiO2 system J. Nucl. Mater. 32 346–50
[502] Devreese J, de Coninck R and Pollak H 1966 On the
conduction mechanism in uranium dioxide Phys. Status
Solidi b 17 825–9
[503] de coninck R and Devreese J 1969 The thermoelectric power
and the conduction mechanism in nearly stoichiometric
uranium oxide single crystals Phys. Status Solidi b
32 823–9
[504] Nagels P, Denayer M and Devreese J 1963 Electrical
properties of single crystals of uranium dioxide Solid State
Commun. 1 35–40
[505] Ikeda S, Sakai H and Tateiwa N 2009 Possible existence of
magnetic polaron in nearly ferromagnetic semiconductor
β -US2 J. Phys. Soc. Japan 78 114704
[506] Pireaux J J et al 1977 Shake-up satellites in the x-ray
photoemission spectra of uranium oxides and fluorides: a
band-structure scheme for uranium dioxide, UO2 Chem.
Phys. 22 113–20
[458] Kimura S-I et al 1993 Optical study of electronic structure and
nonmetal–metal transition of Gd2 S3 J. Phys. Soc. Japan
62 4331–7
[459] Duan C-G et al 2007 Electronic, magnetic and transport
properties of rare-earth pnictides J. Phys.: Condens. Matter
19 315220
[460] Oomi G, Ohashi M and Cho B 2009 Effect of pressure on the
anomalous magnetoresistance and antiferromagnetism of
single crystal GdB4 J. Phys.: Conf. Ser. 176 012038
[461] Ohta M et al 2008 Thermoelectric properties of Th3 P4 -type
rare-earth sulfides Ln2 S3 (Ln = Gd, Tb) prepared by
reaction of their oxides with CS2 gas J. Alloys Compounds
451 627–31
[462] Guizzetti G et al 1990 Electrical and optical characterization
of GdSi2 and ErSi2 alloy thin films J. Appl. Phys.
67 3393–9
[463] Dowben P A 2010 University of Nebraska-Lincoln, private
communication
[464] Bhattacharyya D and Biswas A 2005 Spectroscopic
ellipsometric study on dispersion of optical constants of
Gd2 O3 films J. Appl. Phys. 97 053501
[465] Sahoo N K et al 2003 Reactive electron beam evaporation of
gadolinium oxide optical thin films for ultraviolet and deep
ultraviolet laser wavelengths Thin Solid Films 440 155–68
[466] Landheer D et al 2001 Characterization of Gd2 O3 films
deposited on Si(100) by electron-beam evaporation
J. Electrochem. Soc. 148 G29–35
[467] Bogoroditzkii N et al 1965 Sov. Phys.—Dokl. 10 85
[468] Vickery R C and Muir H M 1961 Thermoelectric properties of
rare earth chalcogenides Adv. Energy Convers. 1 179–86
[469] Hemeda O M, Said M Z and Barakat M M 2001 Spectral and
transport phenomena in Ni ferrite-substituted Gd2 O)3
J. Magn. Magn. Mater. 224 132–42
[470] Lee M W and Lin C H 2000 Determination fo the optical
constants of the γ -phase GdH3 thin films J. Appl. Phys.
87 7798–801
[471] Miniotas A et al 2002 Gadolinium trihydride—a new
magnetoresistive semiconductor Europhys. Lett. 58 442–7
[472] Nowicka E, Nowakowski R and Dus R 2008 Surface and bulk
phenomena in the process of GdHx (0.01 < x < 3)
formation in the thin Gd films: transition from metal to
transparent semiconductor Appl. Surf. Sci. 254 4346–51
[473] van der Sluis P, Ouwerkerk M and Duine P A 1997 Optical
switches based on magnesium lanthanide alloy hydrides
Appl. Phys. Lett. 70 3356–8
[474] Madland D G 2006 Total prompt energy release in the neutron
induced fission of 235 U, 238 U, and 239 Pu Nucl. Phys. A
772 113–37
[475] Takabatake T et al 1990 Heavy-fermion and semiconductor
properties of the ternary uranium compounds U3 T3 Sn4 and
U3 T3 Sb4 (T = Ni, Cu, Pd, Pt and Au) J. Phys. Soc. Japan
59 4412–8
[476] Palstra T T M et al 1988 Electrical and magnetic properties of
semiconducting ternary U compounds: UTSn and UTSb
J. Appl. Phys. 63 4279–81
[477] Herrmann J-M et al 1995 Semiconductive properties of some
uranium–antimony oxide phases used as catalysts in the
mild oxidation of but-1-ene to butadiene J. Chem. Soc.,
Faraday Trans. 91 2343–8
[478] Willardson R K, Moody J W and Goering H L 1958 The
electrical properties of uranium oxides J. Inorg. Nucl.
Chem. 6 19–33
[479] Price R R, Haire M J and Croff A G 2000 Waste Management
2001 Symp., Session 9.3: Advances in Depleted Uranium
Technology, Tuscon
[480] Meek T, Hu M and Haire M J 2000 Waste Management 2001
Symp. (Tuscon)
[481] Meek T T, von Roedern B G and Haire M J 2003 Some
electrical properties of UO2 —Part I Trans. Am. Nucl. Soc.
88 416–8
30
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[532] Gotoh H and Yagi H 1972 The detection efficiency and the
response function of semiconductor proton recoil counters
with axial symmetry to a point neutron source Nucl.
Instrum. Methods 101 395–6
[533] Gotoh H, Yagi H and Harayama Y 1973 Calculation of the
response function of semiconductor proton recoil counters
with a radiator of finite thickness Nucl. Instrum. Methods
109 349–53
[534] Ryves T B 1976 A proton recoil telescope for 12–20 MeV
neutrons Nucl. Instrum. Methods 135 455–8
[535] Kim C et al 1980 Epitaxial integrated E-dE silicon detector
with a buried low-resistive diffused layer IEEE Trans. Nucl.
Sci. 27 258–65
[536] Cinausero M et al 2006 A proton recoil telescope for neutron
spectroscopy J. Phys.: Conf. Ser. 41 219–24
[537] Agosteo A et al 2007 Neutron spectrometry with a monolithic
silicon telescope Radiat. Prot. Dosim. 126 210–7
[538] Donzella A et al 2010 A proton recoil telescope fo neutron
spectroscopy Nucl. Instrum. Methods Phys. Res. A
613 58–64
[539] Stephan A C 2008 Neutron detector USPTO 20080315109
[540] Bramblett R L, Ewing R J and Bonner T W 1960 A new type
of neutron spectrometer Nucl. Instrum. Methods 9 1–12
[541] Kronenberg S and Brucker G J 2003 Dodecahedron neutron
spectrometer with aluminum proton absorber for aircraft
USPTO 6,625,243
[542] Kronenberg S and Brucker G J 2004 Neutron spectrometer
USPTO 6,714,616
[543] Kronenberg S and Brucker G J 2004 Neutron spectrometer
with titanium proton absorber USPTO 6,765,978
[544] Kronenberg S and Brucker G J 2005 Dodecahedron neutron
spectrometer with tantalum proton absorber for aircraft
USPTO 6,928,130
[545] Schulte R L 1991 Variable threshold, directional neutron
sensor USPTO 5,029,262
[546] Esposito A and Nandy M 2004 Measurement and unfolding of
neutron spectra using Bonner spheres Radiat. Prot. Dosim.
110 555–8
[547] Reginatto M and Goldhagen P 1999 MAXED, a computer
code for maximum entropy deconvolution of multisphere
neutron spectrometer data Health Phys. 77 579–83
[548] Mares V and Schraube H 1994 Evaluation of the response
matrix of a Bonner sphere spectrometer with LiI detector
from thermal energy to 100 MeV Nucl. Instrum. Methods
Phys. Res. A 337 461–73
[549] Miller W H and Brugger R M 1985 Additional experimental
tests of the Bonner sphere neutron spectrometer Nucl.
Instrum. Methods Phys. Res. A 236 333–9
[550] Routti J T and Sandberg J V 1985 Unfolding activation and
multisphere detector data Radiat. Prot. Dosim. 10 103–10
[551] Johnson T L and Gorbics S G 1981 An iterative perturbation
method for unfolding neutron spectra from bonner sphere
data Health Phys. 41 859
[552] Lowry K A and Johnson T L 1984 Modifications to recursion
unfolding algorithms to find more appropriate neutron
spectra Health Phys. 47 587–93
[553] Love T A and Murray R B 1961 The use of surface-barrier
diodes for fast-neutron spectroscopy I.R.E. Nucl. Sci.
8 91–7
[554] Rydin R A 1967 Design considerations for a semiconductor
sandwich fast neutron spectrometer IEEE Trans. Nucl. Sci.
14 377–92
[555] Brown W K, Ellis A N and Peterson D D 1968 A 90◦ 3 He
neutron spectrometer IEEE Trans. Nucl. Sci. 15 404–10
[556] Rickard I C 1972 The use of the lithium-6 semiconductor
sandwich spectrometer for the measurement of fast-neutron
spectra Nucl. Instrum. Methods 105 397–411
[557] Fehrenbacher G, Mecbach R and Paretzke H G 1997 Fast
neutron detection with germanium detectors: unfolding the
692 keV peak response for fission neutron spectra Nucl.
Instrum. Methods Phys. Res. A 397 391–8
[507] Hanafi Z M et al 1990 The electrical conductivity of the
different uranium trioxide UO3 phases Radiochem. Acta 49
35–7
[508] Burrell A K et al 2007 Controlling oxidation states in uranium
oxides through epitaxial stabilization Adv. Mater.
19 3559–63
[509] McCleskey T M et al 2009 Precursors for the polymer-assisted
deposition of films WPO WO2009134698
[510] Idriss H 2010 Surface reactions of uranium oxide powder, thin
films and single crystals Surf. Sci. Rep. 65 67–109
[511] Cook G B and Hudswell F 1950 The preparation of thin layers
of uranium oxide J. Sci. Instrum. 27 230–1
[512] Myake C et al 1990 Characterization of uranium oxide thin
film prepared by vacuum deposition method J. Nucl. Sci.
Technol. 27 382–5
[513] Gersch H K, McGregor D S and Simpson P A 2002 The effect
of incremental gamma-ray doses and the incremental
neutron fluences upon the performance of self-biased
10B-coated high-purity epitaxial GaAs thermal neutron
detectors Nucl. Instrum. Methods Phys. Res. A 489 85–98
[514] van Ginhoven R M, Kouzes R T and Stephens D L 2009
Alternative neutron detector technologies for homeland
security PNNL-18471
[515] Copley J R D and Udovic T J 1993 Neutron time-of-flight
spectroscopy J. Res. Natl Inst. Stand. Technol. 98 71–87
[516] Hess W N et al 1959 Cosmic-ray neutron energy spectrum
Phys. Rev. 116 445–57
[517] O’Brien K et al 1978 Cosmic ray induced neutron background
sources and fluxes for geometries of air over water, ground,
iron and aluminum J. Geophys. Res. 83 114–20
[518] Goldhagen P et al 2002 Measurement of the energy spectrum
of cosmic-ray induced neutron aboard an ER-2
high-altitude airplane Nucl. Instrum. Methods Phys. Res. A
476 42–51
[519] Tommasino L et al 2003 Cosmic-ray neutron spectroscopy by
solid state detectors Radiat. Measur. 36 307–11
[520] Sheu R J and Jiang S H 2003 Cosmic-ray-induced neutron
spectra and effective dose rates near air/ground and
air/water interfaces in Taiwan Health Phys. 84 92–9
[521] Goldhagen P, Clem J M and Wilson J W 2004 The energy
spectrum of cosmic-ray induced neutrons measured on an
airplane over a wide range of altitude and latitude Radiat.
Prot. Dosim. 110 387–92
[522] Kouzes R T et al 2007 Passive neutron detection at borders
IEEE Nucl. Sci. Symp. Conf. Rec. N24-4 1115–9
[523] Kouzes R T et al 2008 Cosmic-ray-induced ship-effect
neutron measurements and implications for cargo scanning
at borders Nucl. Instrum. Methods Phys. Res. A 587
89–100
[524] Gromushkin D M et al 2009 31st Int. Cosmic Ray Conf.
(Lodz) pp 1–3
[525] Gromushkin D M et al 2009 Thermal neutron flux detection
near the earth’s surface Bull. Russ. Acad. Sci.: Phys.
73 407–9
[526] Amaldi E, Bocciarelli D and Trabacchi G C 1941 On
proton-neutron collision. I Ricerca Sci. 12 830–42
[527] Amaldi E et al 1944 On proton-neutron collision. II Ricerca
Sci. 13 502–31
[528] Kinsey B B, Cohen S G and Dainty J 1948 A coincidence
method of measuring a flux of fast neutrons Proc. Cambr.
Phil. Soc. 44 96–113
[529] Wilkinson D H 1948 The stopping power of polythene and
fast neutron flux measurements Proc. Cambr. Phil. Soc.
44 114–23
[530] Reid G C 1954 Energy levels in 10 B and 8 Be Proc. Phys. Soc.
A 67 466–8
[531] Furr A K and Runyon R S 1964 A fast neutron spectrometer
for reactor flux measurements Nucl. Instrum. Methods
27 292–8
31
J. Phys.: Condens. Matter 22 (2010) 443201
Topical Review
[564] Moodera J S et al 1995 Large magnetoresistance at room
temperature in ferromagnetic thin film tunnel junctions
Phys. Rev. Lett. 74 3273–6
[565] Bernard L et al 2003 Magnetoresistance in boron carbide
junctions Appl. Phys. Lett. 83 3743–5
[566] Baibich M N et al 1988 Giant magnetoresistance of
(001)Fe/(001)Cr magnetic superlattices Phys. Rev. Lett.
61 2472–5
[567] Engel B N et al 2005 A 4-Mb toggle MRAM based on a novel
bit and switching method IEEE Trans. Magn. 41 132–6
[568] Chadwick M B et al 2006 ENDF/B-VII.0: next generation
evaluated nuclear data library for nuclear science and
technology Nucl. Data Sheets 107 2931–3060
[569] Messenger S R 2010 unpublished ab initio calculations NRL
Code 6816
[558] Chao J-H and Niu H 1997 Measurement of neutron dose by a
moderating germanium detector Nucl. Instrum. Methods
Phys. Res. A 385 161–5
[559] Bandyopadhyay S and Tepper G 2009 private communication,
Virginia Commonwealth University
[560] Tsymbal E Y, Mryasov O N and LeClair P R 2003
Spin-dependent tunneling in magnetic tunnel junctions
J. Phys.: Condens. Matter 15 R109–42
[561] Bandyopadhyay S and Cahay M 2008 Introduction to
Spintronics (Boca Raton, FL: CRC Press)
[562] Tedrow P M and Meservey R 1971 Spin-dependent tunneling
into ferromagnetic nickel Phys. Rev. Lett. 26 192–5
[563] Julliere M 1975 Tunneling between ferromagnetic films Phys.
Lett. A 54 225–6
32