A detector clicks. A graph slopes down. A small difference between two masses becomes an energy. These observations belong to nuclear physics, but none explains itself. The useful question is how a measurement lets us reason about something as small and inaccessible as an atomic nucleus.
The gold standard of nuclear physics is an explanation that preserves the identity of the nucleus, accounts for the allowed transformation and makes a prediction that an instrument can test. This guide develops that standard from proton and neutron counting to original worked cases in decay, background subtraction and reaction rates. The numerical cases are teaching models, not measurements of a particular isotope or instructions for handling radioactive material.
This subject root connects the existing nuclear mechanism guide with neighbouring science and applications. It follows the educational principle in The Gold Standard of Education: learning that transfers into life: understand a relationship, practise it and check whether it remains useful when the situation changes. Begin wherever your question is sharpest; every chapter stays open.
Find your way through nuclear physics
Build a clear account of nuclear identity, energy, decay and the detector evidence used to investigate them.
Full chapter index · Practice and explained answers · How X Works
Chapter contents
Identify nuclei and transformations · Chapters 1–3
Follow decay and detector evidence · Chapters 4–7
Test reaction models · Chapters 8–11
Connect science with practice · Chapters 12–14
CHAPTER 1 OF 18 · Identify nuclei and transformations
1. Identify the nucleus before interpreting the event
An element is identified by its proton number, Z. A particular nuclide also has a neutron number, N; its mass number is A = Z + N. Isotopes share Z and differ in N. Changing the electrical charge of an atom by removing an electron does not, by itself, change its nuclide. That distinction prevents a large family of mistakes before any calculation begins.
Suppose an invented nucleus has 12 protons and 14 neutrons. Its mass number is 26. An ion made from the same nucleus may have 10 electrons, but its proton and neutron counts remain 12 and 14. The ion has changed electrical state, not nuclear identity. If a nuclear process later changes one neutron into a proton, Z changes and the element changes.
This language is useful because a detector generally does not show a little labelled nucleus. It records signals produced by interactions. We infer the underlying process by combining identity, conservation, energies and probabilities. The label on a sample is part of the evidence, but it cannot replace the measurement.
The Department of Energy’s nuclei overview provides a short introduction to protons, neutrons, nuclear structure and decay. Use it to establish the objects of the subject. Keep atomic electrons, nuclei and the instrument’s electronic signal as three different things in your explanation.
CHAPTER 2 OF 18 · Identify nuclei and transformations
2. Binding energy means comparing complete states
A bound nucleus has a lower rest energy than the same constituent protons and neutrons separated and at rest. The difference is its binding energy. The familiar expression E = Δmc² connects that energy difference with a mass difference. It does not mean matter disappears from the accounting or that any chosen subtraction of masses is meaningful.
Consider an explicitly hypothetical mass difference of 0.0050 atomic mass units. Using the rounded conversion 1 u × c² ≈ 931.5 MeV gives 4.66 MeV. This is a total energy difference for the stated comparison. If the nucleus contains five nucleons, the corresponding average is approximately 0.932 MeV per nucleon. Total binding and binding per nucleon answer different questions.
Before calculating, specify whether masses are nuclear masses or neutral atomic masses. Atomic masses include electrons. In some reactions those contributions cancel when the complete equation is written consistently; in others they require attention. A shortcut learned for one reaction can fail when transferred without its conditions.
Binding energy also does not describe one detachable packet belonging to each nucleon. Removing a particular neutron involves a separation energy: the difference between the initial nucleus and a specified final nucleus plus a free neutron. The average over all nucleons need not equal that removal energy.
A useful check is to write the initial and final systems in words before touching the numbers. Ask which particles, excited states and electrons are included. An accurate subtraction begins with an accurate boundary.
CHAPTER 3 OF 18 · Identify nuclei and transformations
3. Decay changes particular properties
Different decay modes change different parts of the nuclear account. Alpha decay emits a helium nucleus and lowers the parent’s proton number by two and mass number by four. Beta-minus decay changes a neutron into a proton while emitting an electron and an antineutrino. The mass number stays the same; the proton number increases by one. Gamma emission can lower a nucleus’s excitation energy without changing its proton or neutron number.
These are descriptions of processes, not interchangeable kinds of light. A beta electron is not simply an atomic electron knocked loose from a shell. A gamma photon carries energy and momentum. In a complete beta account, the neutrino or antineutrino is essential to conservation; ignoring it produces an incomplete energy story.
Take the invented nucleus with Z = 12 and A = 26. After one beta-minus decay, the daughter has Z = 13 and A = 26. Its neutron number is 13. After an alpha decay from the original parent, the daughter would instead have Z = 10 and A = 22. Counting nucleons exposes the difference immediately.
Real decay schemes can have several branches, excited daughters and subsequent emissions. A single arrow in an introductory diagram is a deliberately simplified path. The National Nuclear Data Center's NuDat provides evaluated structure and decay information, including levels and gamma transitions. When using an actual nuclide, inspect its evaluated record and uncertainties rather than borrowing a half-life from a similarly named isotope.
CHAPTER 4 OF 18 · Follow decay and detector evidence
4. Half-life is a population rule, not a personal timer
For a simple population of identical, independently decaying nuclei with constant decay probability per unit time, the expected number remaining is N(t) = N₀ exp(−λt). The half-life T satisfies λT = ln 2. After each interval T, half the expected population remains. The rule does not assign each individual nucleus a scheduled departure time.
In an original teaching case, 16,000 nuclei begin in the parent state and the half-life is three hours. After six hours, two half-lives have passed, so the expected remaining parent count is 16,000/4 = 4,000. After nine hours it is 2,000. The expected number transformed during the interval from six to nine hours is 2,000, not half of the original 16,000.
Now change the context. A delivery is delayed by 90 minutes, half a half-life. The remaining fraction is 2^(−0.5), approximately 0.707. It is not 0.75: repeated halving is exponential, not a straight line between whole half-lives. Working from the exponent protects the relationship when the timing changes.
Activity is the expected number of decays per second, A = λN, measured in becquerels. An inventory count and an activity are therefore different quantities. Two different nuclides with the same number of atoms can have very different activities because their decay constants differ.
These equations assume the parent population is not being replenished. A daughter produced by another nuclide can grow even while its own nuclei decay. Ask whether you are following a disappearing parent, an accumulating daughter or a mixture before fitting a single exponential.
CHAPTER 5 OF 18 · Follow decay and detector evidence
5. A count rate includes the instrument
A detector count is an observed event registered by an instrument. It is not automatically one decay in the sample. Emission probability, detection efficiency, geometry, attenuation, selection windows and instrument dead time all affect what becomes a recorded count. Background adds events that may not come from the sample of interest.
Imagine a sample measurement recording 720 counts in 600 seconds. A separate background measurement records 240 counts in 1,200 seconds. The observed sample-plus-background rate is 1.20 counts per second. The background rate is 0.20 counts per second. The estimated net rate is therefore 1.00 count per second. Subtracting 240 directly from 720 would ignore the different observation times.
For independent Poisson counts in this simplified case, the variance of the net rate is 720/600² + 240/1,200². Its square root is about 0.0465 counts per second. This statistical uncertainty describes the counting model. It does not include a mistaken background location, efficiency calibration or an unnoticed change in geometry.
If the relevant emission occurs in half of decays and the detection efficiency for it is 0.10, the simple activity estimate is 1.00/(0.50 × 0.10) = 20 Bq. That inference is conditional on both factors applying to this setup. An efficiency measured for a different energy or source shape may be inappropriate.
Notice the sequence: record events, correct a rate, state a detection model, then infer activity. Reversing that sequence makes it easy to report a precise-looking number whose meaning has quietly changed.
Check a decay interpretation with a ratio
A useful independent check compares equal-duration measurements made at different times under unchanged geometry and instrument response. In a deliberately simple parent-only example, suppose the first net rate is 8.0 counts per second and the rate four hours later is 2.0. The ratio is one quarter, corresponding to two half-lives under a single-exponential model. The inferred half-life is therefore two hours.
Change the geometry during the second measurement so that only half as many emissions can be registered. The same observed ratio could now combine a factor-of-two activity reduction with a factor-of-two efficiency reduction. Inferring two half-lives from the uncorrected ratio would misattribute an instrumental change to a nuclear one.
| Account | Physical activity ratio | Efficiency ratio | Observed rate ratio |
|---|---|---|---|
| Unchanged setup and two half-lives | 1/4 | 1 | 1/4 |
| Changed setup and one half-life | 1/2 | 1/2 | 1/4 |
The table contains two explanations for the same observed ratio. It is not evidence that both occurred in a real experiment. It shows why a time series needs an instrument-history account. Independent geometry records or a suitable reference measurement can help distinguish the possibilities.
There is also a statistical question. Rates inferred from small counts fluctuate, so a fitted half-life needs uncertainty rather than a single exact-looking number. A second physical limitation is background: if the true signal is becoming small while background stays approximately constant, the uncorrected total rate no longer follows the parent’s exponential in the same way.
These are different reasons for a curve to depart from a simple model. More data points help only if the analysis preserves time, background and setup information. The diagnosis should ask which relationship became unstable before replacing the whole explanation.
CHAPTER 6 OF 18 · Follow decay and detector evidence
6. Spectra become evidence through calibration
An energy spectrum groups registered events according to an energy estimate. A peak can indicate a characteristic transition, but the horizontal axis begins as an instrument response. Calibration connects that response with energy. Resolution, background, overlapping features and incomplete energy deposition influence the pattern.
Suppose a teaching detector assigns channel 100 to 200 keV and channel 300 to 600 keV. Under an assumed linear response, the calibration is 2 keV per channel with zero intercept. A peak at channel 250 is assigned 500 keV. If a third calibration point disagrees with that line, the appropriate response is to investigate the model, not suppress the inconvenient point.
A peak’s centre and its width answer different questions. The centre estimates a transition energy through the calibration. The width reflects the detector response and potentially other broadening effects. Two nearby physical lines can merge into one observed feature. A narrow bin size cannot recover information that the detector did not resolve.
Peak area and peak height are also different. A broadened peak can have lower height while retaining similar integrated counts. Comparing heights without considering widths can create an apparent change in emission intensity.
A disciplined interpretation records the calibration data, energy range, fitting method and background treatment. Then ask whether an alternative set of transitions could explain the same features. Agreement with one familiar peak is a lead, not the end of identification. The observation becomes stronger when several independent features fit the same account.
CHAPTER 7 OF 18 · Follow decay and detector evidence
7. Conservation comes before a reaction story
For a proposed nuclear reaction, begin by checking charge, nucleon accounting and complete energy and momentum conservation. An attractive verbal story does not excuse an unbalanced equation. Conversely, a balanced equation only establishes necessary conditions; it does not establish that the reaction occurs readily.
Consider a hypothetical neutron capture: a target with Z = 8 and A = 16 absorbs one neutron and forms a nucleus with Z = 8 and A = 17, possibly with photon emission as it relaxes. Proton number is unchanged and one nucleon has been added. If someone instead writes a daughter with A = 18, the missing nucleon must be explained or the account rejected.
The reaction Q-value compares initial and final rest energies. A positive Q indicates energy is released in the specified transformation. Yet a repulsive electrical barrier, selection rules, competing channels or a small interaction probability can still suppress the reaction rate. Energetically allowed and frequently observed are different statements.
Momentum matters even when introductory diagrams omit it. An emitted photon or particle recoils against the remaining system. A complete energy allocation can include daughter recoil, excitation and kinetic energy, rather than assigning the entire release to the most visible product.
For students, the learning order is helpful: identify participants, balance the account, compare energies, then discuss probability and measurement. When a later example becomes unfamiliar, this sequence remains available even if the final numerical method changes.
CHAPTER 8 OF 18 · Test reaction models
8. Cross-section links a beam to a probability
A reaction cross-section describes the probability scale for a particular interaction under specified conditions. It is not generally the geometric silhouette of a nucleus. It can vary strongly with incident particle energy and reaction channel. A quoted value without those conditions may answer a different experiment.
For a thin target, a simplified reaction rate is R = ΦNₜσ. Here Φ is particle flux per area per time, Nₜ is the number of target nuclei exposed and σ is the relevant cross-section. The approximation assumes that the beam is not substantially depleted or changed through the target and that the target and beam overlap as described.
Use an invented case: flux 1.0 × 10⁶ particles per square centimetre per second, 1.0 × 10¹⁸ target nuclei and cross-section 2.0 × 10⁻²⁴ square centimetres. Multiplication gives 2.0 reactions per second. A detector that registers one quarter of the relevant products would produce an expected signal rate of 0.50 per second before background.
Now double the target thickness. The thin-target calculation predicts twice the rate only while its assumptions remain reasonable. A thick target can attenuate or slow the beam, changing both flux and energy-dependent cross-section through its depth. The elementary proportionality has not become wrong; its domain has been exceeded.
This distinction is useful far beyond nuclear physics. When a simple model predicts a linear improvement, ask whether changing the input also changes the conditions that justified linearity. State the boundary before extrapolating the trend.
CHAPTER 9 OF 18 · Test reaction models
9. Models expose different parts of nuclear structure
A nucleus is a quantum many-body system. There is no single classroom picture that explains every feature equally well. A liquid-drop description emphasises collective trends; shell descriptions organise single-particle states in an effective environment. Other methods address correlations, deformation and interactions at different levels of detail.
The models are tools with explanatory jobs. A smooth trend in binding energies and a local change associated with a closed shell need not demand the same approximation. A model can capture the broad background while missing a specific structural feature. That failure can become informative if the prediction and residual are preserved.
Imagine three hypothetical nuclei with measured excitation energies of 1.0, 1.1 and 2.0 MeV. A deliberately simple model predicts 1.0 MeV for all three. Its average error is not the only issue: the isolated discrepancy may identify a systematic feature absent from the model. Refitting a free constant until one value matches can conceal that clue.
A good comparison states which observations were used to determine model parameters and which were predicted afterwards. Otherwise, fitting and testing can become indistinguishable. It also reports a sensible uncertainty or expected accuracy, rather than presenting every calculated digit as equally meaningful.
The transferable habit is to ask what a model retains and what it discards. A helpful simplification removes detail while preserving the relationship needed for the question. An unhelpful simplification removes precisely the mechanism that the observation is testing.
CHAPTER 10 OF 18 · Test reaction models
10. Fission and fusion connect physics with larger systems
Fission divides a nucleus into fragments; fusion combines nuclei into a different nuclear system. Whether energy is released depends on the complete initial and final states. The broad binding-energy pattern helps explain why some heavy-nucleus fission and light-nucleus fusion reactions release energy, but it does not replace reaction-specific accounting.
A nuclear energy system adds engineering questions: maintaining suitable conditions, transferring heat, controlling processes, protecting people and managing material throughout its life. Demonstrating an energy-releasing reaction is therefore one step in a larger chain. The existing How Nuclear Energy Works guide is the appropriate route for that system-level connection.
An original arithmetic comparison makes the distinction visible. Suppose a hypothetical reaction releases 5 MeV per event and an experiment produces 10⁹ events per second. Using 1 eV ≈ 1.602 × 10⁻¹⁹ J gives an energy-release rate of approximately 0.000801 W. A large event count can correspond to a modest macroscopic power. Event energy and event rate must both be included.
That value is released reaction energy, not delivered electrical power. A complete system account would include input requirements, conversion, losses and the chosen reporting boundary. The example is deliberately not a reactor design or a prediction of a facility’s performance.
When reading a claim about a breakthrough, separate the measured physical result from the proposed application. Ask which quantity was demonstrated and which additional steps remain. This lets curiosity and careful judgement coexist.
CHAPTER 11 OF 18 · Test reaction models
11. Cosmic questions connect different scales
Nuclear reactions help explain the formation and transformation of elements in stars and energetic astrophysical environments. The relevant observations can include spectra, abundances and other astronomical signals. Connecting those observations with nuclear mechanisms requires several fields to work together.
The connection is not a permission to use any laboratory reaction rate unchanged in any star. A stellar environment has distributions of particle energies, densities, temperatures and possible reaction pathways. A laboratory measurement may cover only part of the relevant range. Extrapolation requires an explicit model and an account of uncertainty.
Imagine two candidate pathways that produce the same final species but respond differently to temperature. Observing the final abundance alone may not distinguish them. Additional abundance ratios, environmental information or reaction data may be needed. The logic resembles identifying a sample from several spectral features rather than one peak.
The existing stellar nucleosynthesis and astrophysics guides in the directory below develop this bridge. Their shared learning task is moving between scales without pretending that one scale contains the whole explanation. Nuclear structure influences reaction possibilities; astrophysical conditions influence which possibilities matter.
A useful student exercise is to label every arrow in a proposed explanation as observation, measured input, calculation or inference. The resulting chain reveals where the confidence comes from. It also shows which missing measurement would improve the account most, rather than merely adding another dramatic image.
CHAPTER 12 OF 18 · Connect science with practice
12. Applications require more than subject vocabulary
Nuclear physics supports imaging, materials analysis, dating methods and many research instruments. These applications use different observables and serve different decisions. An image reconstructed from emitted radiation, an age estimate and an elemental analysis are not interchangeable ways of reading a count rate.
For example, a tracer application requires an account of where a labelled substance goes, how its physical decay changes the signal and how the detector observes it. A measured decline can reflect both decay and transport. Attributing all change to one process can misrepresent the application even when the nuclear half-life is correct.
The linked PET and cardiac-perfusion explainers offer routes into the measurement and reconstruction questions. This educational root does not establish clinical interpretation or recommend an investigation. In professional practice, qualified teams use validated methods, applicable controls and the patient’s or project’s actual requirements.
Dating introduces another useful distinction. A decay equation can relate isotope amounts and time only within an appropriate model of initial conditions, exchange and system closure. Solving the equation accurately cannot compensate for an unsuitable sample history. The mathematical result inherits the assumptions used to obtain it.
The general capability is to ask what is being inferred from the nuclear signal. Specify the target quantity, the physical link and the independent evidence that the link applies. That question helps a reader understand why a professional result requires more than a working detector.
CHAPTER 13 OF 18 · Connect science with practice
13. Learn by building the account in layers
Begin with proton and neutron bookkeeping. Then learn exponential change, units and uncertainty before interpreting complicated decay schemes. Build energy and momentum conservation alongside simple quantum ideas. This sequence gives later techniques something stable to connect with.
A learner who can recite a half-life but confuses counts with becquerels needs a different repair from one who understands the instrument but cannot manipulate exponentials. Ask for the reasoning behind an answer. A wrong conversion may reveal a unit problem; a wrong daughter identity may reveal a nuclear-language problem.
Teaching can progress from a complete worked example to a partially completed account and then to an independent changed-context problem. In the count-rate case, first keep observation times equal. Next change the times. Finally introduce an efficiency factor and ask what evidence would justify using it. Each added difficulty has a visible purpose.
For more advanced study, computing is valuable when the model remains inspectable. A short simulation can generate independent decay events or Poisson counts. Compare its distribution with the analytical expectation. Preserve random seeds or repeat enough runs to show that one attractive simulated curve is not the distribution itself.
Professional work adds experimental preparation, instrumentation, data stewardship, collaboration and supervised training. Enjoying this article is a starting interest, not a professional qualification. A useful next step is a small calculation or data interpretation that you can explain clearly and then check independently.
CHAPTER 14 OF 18 · Connect science with practice
14. Build an evidence trail another person can inspect
A strong result preserves more than the final graph. It records sample identity, configuration, observation time, calibration, selection rules in the analysis and relevant environmental conditions. These details let another person understand why the reported quantity follows from the observations.
Separate raw events from derived products. A background-subtracted spectrum is useful, but retain the source and background spectra and their exposure times. If a later analyst changes the energy window, they should be able to reconstruct what changed rather than inherit an unexplained number.
Track uncertainty by its cause. Counting variability differs from calibration uncertainty, and both differ from a model choice. Repeating the same measurement longer may reduce statistical uncertainty while leaving a biased calibration unchanged. The appropriate next check depends on which limitation dominates.
In a fictional school analysis, two teams obtain different activities from the same counts. One used the correct energy-dependent efficiency; the other used a single reference value from a different geometry. More significant figures do not reconcile the disagreement. Reconstructing the efficiency assumption does.
The gold standard is an account in which the next person can follow the path from event to conclusion. It should be possible to say what would change your mind: a revised calibration, an alternative background, a missing branch or a measurement outside the model’s claimed accuracy. That openness strengthens an explanation because its limits are visible.
Is all radiation nuclear?
No. Electromagnetic radiation can arise from several processes. Nuclear gamma transitions and electron transitions are distinguished by origin, even though both produce photons. To explain a signal, identify the physical process rather than relying on the general word radiation.
Does a longer half-life always mean less concern?
Half-life alone does not answer a practical exposure or handling question. Activity depends on the number of nuclei as well as the decay constant; emitted radiation, amount, pathway and context also matter. This guide teaches the quantities rather than reducing professional decisions to one number.
Is a positive reaction energy enough to prove feasibility?
No. An energetically allowed channel may have a small rate. The relevant cross-section, incident energy, barriers and competing processes must enter the explanation. A conservation check is necessary but not a complete reaction model.
Does a background-subtracted zero prove nothing happened?
A net estimate near zero can arise when signal and background are comparable. Interpret it with the measurement uncertainty and detection method. It does not establish that the true process rate is exactly zero.
Why use more than one model?
Different models preserve different relationships. Compare their predictions with the observations relevant to the question. The goal is not to collect impressive model names; it is to explain which simplification is justified and what evidence could reveal its failure.
These distinctions make revision practical. When an answer is wrong, locate the category of error before repeating a calculation. A unit correction, a changed boundary and a better physical model repair different things.
CHAPTER 16 OF 18 · Check and explore
16. Practice, changed-context checks and explained answers
First, a hypothetical parent has half-life four hours. An expected population of 24,000 is observed initially. How many remain after ten hours? There are 2.5 half-lives, so N = 24,000 × 2^(−2.5) ≈ 4,243. A linear subtraction or two whole halvings would miss the fractional interval. The answer is an expectation under the simple decay model.
Second, a measurement records 900 counts in 300 seconds, while background records 600 counts in 600 seconds. The net rate is 3 − 1 = 2 counts per second. If the background had instead been measured for 300 seconds with 600 counts, the net rate would be 1 count per second. The same background count total does not mean the same background rate.
Third, a hypothetical target has Z = 20 and A = 44. After beta-minus decay, Z = 21 and A = 44; N = 23. After gamma emission from an excited state, Z and A would stay unchanged. Describe the process before applying a memorised change.
Fourth, the thin-target reaction rate is predicted to be 10 per second and total registration efficiency is 0.20. Expected registered signal is 2 per second. If only one quarter of reactions produces the chosen emission and that factor was not already included, the expectation becomes 0.50 per second. Always check whether a stated efficiency already includes branching.
Finally, explain why doubling collection time does not fix a wrong calibration. It changes the number of observed events, not the physical mapping from instrument output to the target quantity. A good answer names the limitation and proposes a check aimed at that limitation.
CHAPTER 17 OF 18 · Check and explore
17. Connected reading by the question you want to answer
Foundations and the central mechanism
Nuclear physics mechanism guide develops nuclear structure, decay, reactions and measurements. The Gold Standard of Physics connects these ideas with the broader discipline. The Gold Standard of Atomic Physics separates electronic states and precision spectroscopy from nuclear identity. The Gold Standard of Quantum Mechanics provides the language of states, probabilities and measurement.
Energy and the universe
How Nuclear Energy Works extends reaction physics into energy systems. The Gold Standard of Nuclear Engineering adds the distinct engineering responsibilities from design through decommissioning. How stars build chemical elements connects reaction pathways with element formation. Astrophysics develops the environment in which those pathways operate. Plasma Physics explains collective ionised matter without treating a plasma as a synonym for a nuclear reaction.
Signals, images and specialised evidence
Cloud chambers and track evidence introduces how interactions leave observable traces. Mössbauer spectroscopy and nuclear resonance develops a specialised nuclear probe. PET coincidence detection and reconstruction connects emitted particles with imaging geometry. Nuclear cardiac perfusion imaging follows radiotracers and reconstruction in a particular application. Hahn spin echo explores coherence and its recovery in magnetic-resonance experiments; it is a distinct measurement route from radioactive decay.
Models, learning and further exploration
Why scientists use different theories at different scales helps assess approximations. The Gold Standard of Geochemistry connects isotope measurements with Earth questions. The Gold Standard of Education returns to explanation, practice and transfer. How X Works offers the wider subject network. Choose the article that addresses your next question rather than treating every branch as a compulsory reading list.
CHAPTER 18 OF 18 · Check and explore
18. Sources, boundaries and a useful next step
The primary institutional introductions checked on 21 October 2026 include the Department of Energy’s Nuclear Physics programme and nuclei overview. NNDC NuDat is a route to evaluated nuclide structure and decay data; its ground-state, level and decay datasets serve different queries. These references support the subject framing and the distinction between an introductory model and evaluated nuclear data.
For additional conceptual study, the existing nuclear mechanism guide links its educational references alongside its own worked examples. The rounded physical constants can be checked against the NIST 2022 CODATA complete listing, including the Faraday constant, elementary charge and atomic-mass energy equivalent as relevant to the calculation.
The numerical examples here were constructed for this article and use rounded constants. They do not report a new measurement, specify a real isotope’s half-life or certify an experimental method. Their purpose is to make the reasoning inspectable.
Try one final explanation without looking back: describe how a recorded detector count becomes an estimate of sample activity. Include observation time, background, emission probability and registration efficiency, then name a condition that could invalidate the estimate. If one step remains unclear, that is the next useful learning task.
The subject becomes powerful when invisible events can support a transparent explanation. A nucleus is small, but the intellectual job is recognisable: identify the system, preserve the account, make a conditional prediction and compare it with evidence. Carry that sequence into the next unfamiliar example.