Question 2 of 7: Self-Powered Neutron Detectors — Materials, Mechanisms, and Placement
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. 98-Phys-B1 Radiation Physics, National Examination
December 2017 — a three-hour open-book examination in which any
non-communicating calculator is permitted (the candidate must record the calculator's make and
model on the first sheet). The cover page states the exam has 7 questions worth a total
of 74 points, of which only 60 points' worth need be answered for full marks; every
question and sub-part is nonetheless answered in full below so the paper remains a complete
study resource. The cover page also invites the candidate to submit a written statement of any
assumptions made where a question is open to interpretation — this licence is used below in
Question 1(a)–(b) (the historic DOE report's "roentgens per hour" reading is converted to
absorbed dose using the standard air-kerma factor since no calibration medium is stated) and
1(e) (the Canadian nuclear-energy-worker annual effective-dose limit, 50 mSv/yr, is used to
size the inspection-crew rotation since the source states no dose constraint of its own), and in
Question 6(a) (counting-statistics uncertainty is taken as Poisson, $\sigma(C)=\sqrt{C}$, on the
one-minute count reported in each row, since the source gives no separate counting-time
datum). Question 6 also carries a genuine internal inconsistency between the table header's
definition of $g(t)$ and the definition restated in part (c) — both readings and the
resolution adopted are flagged where they occur.
Reference texts. K. S. Krane, Introductory Nuclear Physics (nuclear
reaction kinematics, pair production, fission energetics); F. H. Attix, Introduction to
Radiological Physics and Radiation Dosimetry (exposure–dose conversion, photon
interactions, non-ionizing radiation); J. R. Cember and T. E. Johnson, Introduction to Health
Physics, 5th ed. (radiation weighting factors, ALARA dose planning, decay-counting
statistics); J. E. Turner, Atoms, Radiation, and Radiation Protection, 3rd ed. (neutron
detectors, radioactive decay/in-growth kinetics, radiation protection principles).
Given. Six candidate emitter materials for self-powered neutron detectors
(SPNDs), each with its capture reaction type, thermal cross-section, and resonance behaviour, as
tabulated above.
Find. (a) the physical meaning of "γ-photon reaction"; (b)
(i)–(v) the operating principle of SPNDs and why each signal type tracks reactor power;
(c) the best fixed in-core emitter choice; (d) the best movable in-core emitter choice.
Approach. Work from the physical signal chain of a self-powered detector
(neutron capture → charged-particle emission → net current across an insulating gap,
with no external bias) to explain each interaction type, then trade off sensitivity against
long-term burnup stability to choose emitters for the two deployment scenarios.
Part (a) — "reaction" vs. "interaction". "Interaction" is the general
term for any coupling between the incident particle and matter, including processes that leave
the interacting species fundamentally unchanged (e.g. elastic or Compton scattering, where the
photon or neutron survives, merely deflected and reduced in energy). "Reaction" is reserved for a
process that converts the incoming quantum into a genuinely different final state —
here, a capture-gamma photon is completely absorbed by an atomic electron (photoelectric
absorption) or gives up essentially all its usable energy in producing a free photoelectron, so
the photon ceases to exist as such and a new species (a free conduction electron plus an ionized
atom) appears in its place. That one-way, "reactant → product" conversion is exactly the
sense in which a chemical reaction consumes reactants to yield products, which is why the
question's hint points toward chemistry — it is the completeness of the conversion, not
merely the fact that a photon struck an electron, that earns the label "reaction" rather than
"interaction".
Part (b)(i) — self-powered. "Self-powered" means the detector needs no
external high-voltage bias supply to generate its signal, unlike an ion chamber, proportional
counter or GM tube, which all require an applied field to collect charge. In an SPND the emitter
material itself, upon neutron capture, directly emits a charged particle (a beta particle, or a
photoelectron/Compton electron produced by a capture gamma) that crosses a thin insulating gap
to a surrounding collector; the resulting net charge transfer is the signal current, driven
entirely by the nuclear/atomic process itself.
Part (b)(ii) — why (n,β) tracks reactor power. The rate at which
the emitter captures neutrons via (n,β) is directly proportional to the local thermal
neutron flux (rate $=n\sigma\phi$, cross-section $\sigma$ fixed for a given material, flux
$\phi$ proportional to local reactor power). Each capture is followed by a beta decay whose
emitted electron has enough range to cross the emitter–collector gap, so the net current is
proportional to the capture rate, and therefore to the local neutron flux/reactor power.
Part (b)(iii) — why (n,γ) still tracks reactor power despite delayed
fission-product γ's. The reactor core does carry a large, slowly-varying
background γ field from delayed fission-product decay, which is not instantaneously
proportional to the local neutron flux. However, the (n,γ)-type SPND signal is dominated by
the prompt capture gamma emitted the instant the emitter itself absorbs a neutron
(essentially co-located and coincident with the local flux), which for a well-chosen emitter
(large $\sigma_{n,\gamma}$, e.g. hafnium) is much larger than the ambient background gamma
contribution reaching that same small volume of emitter material. The signal therefore still
tracks local flux/power, with the delayed-fission-product background acting only as a
slowly-varying offset that a calibrated detector design accounts for.
Part (b)(iv) — how the γ-photon reaction produces a signal. The
prompt capture gamma from the (n,γ) event travels a short distance within the emitter and
photoelectrically ejects (or Compton-scatters) an atomic electron from the emitter material
itself. If that electron has enough energy and originates close enough to the emitter's outer
surface, it crosses the insulating gap to the collector, exactly as a directly-emitted beta
particle would — producing a net current (the "Compton current" or "gamma-current" mode)
that is again proportional to the local capture-gamma flux, and hence to the local neutron
flux/reactor power.
Part (b)(v) — low-power preference. At low power (e.g. reactor
startup), the (n,γ)/gamma-current type detector is preferred, because its signal is
essentially prompt (the capture-gamma-driven photoelectron current follows the
instantaneous flux with negligible time lag), which is exactly what is needed to track a rapidly
changing, low-magnitude flux during startup. The (n,β) emitters (Rh, V, Ag) all have a
build-up/decay time constant set by their beta-decay half-life (tens of seconds for Rh, for
example), so their signal lags the true instantaneous flux — a lag that matters far less at
steady, high power (where their larger cross-section gives better sensitivity) than it does
during a fast, low-power transient.
Part (c) — best FIXED in-core emitter. Vanadium is the best choice for
a permanently fixed, whole-lifetime in-core detector. Its cross-section follows a smooth $1/v$
law with no resonances, so its response stays proportional to flux without the
burnup-driven sensitivity drift that a resonance-dominated absorber suffers as its resonance
peak is progressively depleted over years of irradiation. Rhodium has the largest cross-section
(highest sensitivity) but its strong 1.25 eV resonance burns out over the reactor's
lifetime, steadily reducing sensitivity and forcing frequent recalibration — unacceptable
for a detector that must remain accurate for the life of the plant.
Part (d) — best MOVABLE in-core emitter. Rhodium is the best choice for
a movable (travelling in-core probe, TIP-type) detector. Because a movable probe is inserted only
briefly and periodically for flux mapping rather than left in place for years, long-term burnup
depletion of its resonance is not a limiting concern; what matters instead is getting the
strongest possible, well-characterized signal during each short traversal, and Rh's large
145 b cross-section (plus its well-known, fast beta decay) gives the best sensitivity of any
of the candidates for that short-duration use.
Question 2 — results
Part
Result
(a)
γ-photon reaction = complete photon→photoelectron conversion (reactant→product), unlike a mere scattering "interaction"
(b)(i)
No external bias – signal comes directly from the emitter's own charged-particle emission
(b)(ii)
(n,β) capture rate ∝ local flux; emitted β crosses the gap → current ∝ power