Infer radioactive half-life from corrected count evidence

Try: the detector does not count only the source

The original reference packet supplies expected counts of 920, 520 and 320 in equal ten-minute windows beginning at 0, 6 and 12 hours. Begin with proposed background zero. The gross decline looks exponential only approximately. Before changing the background, predict whether subtracting a constant positive contribution makes the source decline faster or slower in fractional terms.

A separate matched blank gives 120 counts per ten minutes. Set background to 12 counts/min. The corrected rates are 80, 40 and 20 counts/min: each six-hour separation halves the expected source contribution. These are original calculated examples, not a measured named isotope.

Connect nuclei, activity and detector counts

An isolated radioactive parent has expected population N(t) = N(0) exp(−λt), where λ is the decay constant per hour here. Activity is the number of nuclear transformations per unit time; with consistent time units its expectation is λN. A detector records only some events and also registers background. Detector counts/min are not automatically transformations/second or becquerels. Unknown efficiency prevents inferring absolute activity from these packets.

Individual events are random. A half-life describes survival probability and the large-population expectation; it does not guarantee exactly half of a small finite group decays at a scheduled instant. A single isolated parent with no feeding, no counted radioactive daughter and unchanged response is a deliberately limited inference model.

Count windows are finite

Subtract the blank rate from each gross count divided by its counting duration. Equal durations matter: the integral of an exponential over a fixed-width window contains the same multiplicative window factor at every start time. That factor cancels in a ratio. No instantaneous-window approximation is required for these equal windows. Different durations would require their proper integration factors, not an assumption that raw totals remain comparable.

For a corrected endpoint ratio r between zero and one over twelve hours, λ = −ln(r)/12 and half-life = 12 ln(2)/[−ln(r)]. At r = 1/4, two half-lives fit into twelve hours, giving six hours. The middle observation is not used to fit this endpoint result: compare it separately with the geometric mean of the corrected endpoints.

Experiment with bounds and refusal

The bounded packet assigns ±20 counts to each gross reading and ±2 counts/min to the common blank. These are stipulated absolute ranges, not Poisson standard deviations or confidence intervals. The lowest endpoint ratio uses the low final reading, high initial reading and the same high blank in both: (30−14)/(94−14) = 0.2. The highest uses (34−10)/(90−10) = 0.3. Transform the endpoints through the increasing half-life function. Do not subtract a different favorable background from each observation.

The changed-response packet withholds inference because detector changes can mimic source changes. The background-level packet allows a nonpositive corrected final signal, so a positive logarithmic ratio is not supported across its bounds. Increasing readings conflict with isolated decay. Neither clipping negative signals to a tiny positive value nor returning a negative half-life repairs the missing evidence.

Apply and continue

For gross counts 550 and 300, equal ten-minute windows six hours apart and blank 50 counts per window, correct first: 500 and 250. Explain the six-hour conditional half-life, then name an observation that could challenge the stable-response assumption. An actual isotope identification would need further evidence; this public packet cannot supply it.

Reuse first-order half-life for exponential arithmetic, the geological isotope clock for daughter-retention limits, and time-course discrimination for alternative chemical rate laws. This activity adds detector evidence and conditional inference rather than another assigned exponential slider.

The EPA radioactive-decay explanation supplies the nuclear-decay and half-life context. NIST Radioactivity SRM Procedure 15, version 5.00 explains detector inefficiency, background and measurement-model requirements. No measurement values or uncertainty claims from those sources are assigned to these original packets. Authentic empirical kinetics, practical acceptance and independent explanation review remain separate requirements.