📝 Carbon dating using differential equations (36 MCQs)

📖 Explanation: In radioactive decay models, the rate of change must be negative relative to the amount present. The differential equation should be $\frac{dy}{dt} = -ky$ where $k > 0$, or $\frac{dy}{dt} = ky$ where $k < 0$. Using a positive proportionality constant in the growth form leads to exponential increase rather than decay, resulting in physically impossible negative time values when solving for age. This tests understanding of sign conventions in modeling.

📖 Explanation: Starting from $y(t) = y_0 e^{-kt}$, we set $\frac{y(t)}{y_0} = 0.75$. Taking natural logs gives $\ln(0.75) = -kt$. Solving for $t$ yields $t = -\frac{\ln(0.75)}{k}$. Since $\ln(0.75)$ is negative and $k$ is positive, the result is positive. Option A would yield negative time, C uses incorrect ratio inversion without sign adjustment, and D has dimensional inconsistency. This requires careful algebraic manipulation of logarithmic equations.

📖 Explanation: Since decay is exponential, equal percentage drops do not correspond to equal time intervals. Going from 60% to 30% represents losing half of the remaining amount, which takes exactly one half-life regardless of starting point. However, reaching 60% from 100% takes less than one half-life (since 50% remains after one half-life). Thus, the time to go from 100%→60% plus one half-life exceeds twice the time for 100%→60%. This tests deep understanding of non-linear decay dynamics.

📖 Explanation: After ~10 half-lives, only about 0.1% of original C-14 remains. While mathematically defined, the absolute quantity becomes so small that instrumental noise and contamination overwhelm the signal. The rate of decay $\frac{dy}{dt} = -ky$ also becomes extremely small, meaning tiny measurement uncertainties in remaining atoms translate to enormous age uncertainties. This connects calculus concepts (derivatives, sensitivity) to practical scientific limits rather than pure mathematical breakdown.

📖 Explanation: Taking natural log of $y = y_0 e^{-kt}$ gives $\ln(y) = \ln(y_0) - kt$. This is linear in $t$ with slope $-k$. Therefore, $k$ equals the negative of the slope. The y-intercept gives $\ln(y_0)$, not $k$. This tests ability to transform exponential relationships into linear forms for parameter extraction, a key skill in data analysis involving differential equations.

📖 Explanation: If actual remaining C-14 is higher than expected for a true age, the simple model interprets this excess as less decay having occurred, implying less time has passed. Thus, the calculated radiocarbon age underestimates the true calendar age. This requires understanding that the standard model assumes constant initial concentration, and deviations create systematic biases. It integrates historical context with mathematical modeling assumptions.

📖 Explanation: While mathematically correct to include absolute values during indefinite integration, in physical contexts like radioactive decay, $y(t)$ denotes quantity of substance, which cannot be negative. Thus $|y| = y$ for all relevant $t$. This distinguishes mathematical rigor from applied simplification. Students often mechanically retain absolute values or drop them without justification; this question probes understanding of domain restrictions in modeling.

📖 Explanation: Because $t = -\frac{1}{k}\ln(f)$ where $f$ is fraction remaining, and $\ln(f)$ is nonlinear, equal absolute errors in $f$ produce unequal errors in $t$. At $f = 0.45$, decreasing $f$ to 0.43 increases $t$ more than increasing $f$ to 0.47 decreases $t$, since $|\frac{d}{df}\ln(f)| = \frac{1}{f}$ grows as $f$ shrinks. Thus, age confidence intervals are skewed toward older ages. This applies calculus (derivative sensitivity) to error propagation.

📖 Explanation: Contamination adds C-14 independently of existing amount, so it appears as a positive source term $+c$. The loss term remains proportional to current amount $-ky$. Option B incorrectly makes contamination proportional to current mass. Option C subtracts contamination (removal). Option D shifts the decay baseline. This tests ability to construct differential equations from verbal descriptions of competing processes, a core modeling skill beyond rote formula application.

📖 Explanation: The decay model $y(t) = y_0 e^{-kt}$ implies $\frac{y(t)}{y_0} = e^{-kt}$, which depends only on time and $k$, not absolute mass. Age determination relies solely on the fraction remaining, assuming known initial ratio. Mass affects measurement precision but not the theoretical age calculation. This reinforces that exponential decay is scale-invariant, distinguishing intensive from extensive properties in modeling.

📖 Explanation: If the student uses $y = y_0 e^{kt}$ with positive $k$, then solving gives $t = \frac{1}{k}\ln(y/y_0)$. Since $y < y_0$, $\ln(y/y_0) < 0$, yielding negative $t$. But if they ignore the sign and take absolute value or misapply logs, they might get the right magnitude. However, strictly speaking, the expression $\frac{1}{k}\ln(y/y_0)$ with positive $k$ and $y<y_0$ is negative. The distractor highlights common sign-handling mistakes where students compensate errors unconsciously.

📖 Explanation: Setting $y(T) = y_0 / 2$ gives $y_0 / 2 = y_0 e^{-kT}$. Dividing both sides by $y_0$ (assuming $y_0 \neq 0$) yields $1/2 = e^{-kT}$, so $T = \ln(2)/k$, independent of $y_0$. This demonstrates a key property of exponential decay: characteristic timescales depend only on the rate constant, not initial conditions. This conceptual understanding prevents misconceptions that larger samples decay slower.

📖 Explanation: Each halving corresponds to one half-life (~5730 yr). From 25% to 12.5% is one half-life. 22% is between 25% and 12.5%, closer to 25%. Ratio 25/22 ≈ 1.136. Time difference Δt = (1/k) ln(25/22). Since k = ln2 / 5730, Δt = 5730 * ln(25/22)/ln2 ≈ 5730 * 0.127/0.693 ≈ 5730 * 0.183 ≈ 1050 years. Closest to 1140 (which is 0.2*5730). Option B is reasonable approximation. Tests mental estimation using half-life scaling without calculator.

📖 Explanation: Using $\int_{y_0}^{y} \frac{1}{u} du = \int_{0}^{t} -k ds$ directly yields $\ln(y) - \ln(y_0) = -kt$, embedding $y(0)=y_0$ inherently. Indefinite integration requires finding C via substitution, adding a step prone to algebraic error. Definite integration streamlines IVP solutions and reinforces the connection between accumulation and boundary conditions. This promotes efficient, less error-prone methodology in applied calculus.

📖 Explanation: Correct model: $y = y_0 e^{-kt}$ with $k = \ln(2)/T$. Incorrect model uses exponent $-t/T$, implying decay constant $k_{wrong} = 1/T$. Since actual $k = \ln(2)/T ≈ 0.693/T$, the student’s model decays faster by factor $1/\ln(2) ≈ 1.44$. This tests recognition of parameter roles in exponential functions and consequences of misidentifying constants.

📖 Explanation: Marine organisms incorporate carbon from dissolved bicarbonate, which has different C-14/C-12 ratio due to ocean reservoir effects and slower exchange with atmosphere. Volcanic CO2 is depleted in C-14, but charcoal from plants fixing atmospheric CO2 is valid. Terrestrial biosphere generally equilibrates with atmosphere. Marine samples require reservoir correction. This tests awareness of model limitations tied to ecological pathways, not just mathematical form.

📖 Explanation: Differentiating gives $dy/dt = -k y_0 e^{-kt} = -k y(t)$. Thus rate is directly proportional to current amount, defining exponential decay. This is the core property distinguishing exponential from other decay models (e.g., linear or quadratic). Recognizing this proportionality is essential for validating whether a process follows first-order kinetics.

📖 Explanation: On semi-log plot, pure exponential decay is straight line. Upward deviation at old ages (low y) means measured y is higher than predicted, i.e., excess C-14. Most likely cause is contamination with newer carbon, adding modern C-14 to ancient sample. Instrumental drift could cause it but contamination is primary concern. Decreased decay rate would affect all ages uniformly. Higher initial concentration would shift entire line vertically, not cause curvature. This integrates graphical analysis with forensic interpretation of anomalies.

📖 Explanation: Euler’s global error is O(h), but roundoff accumulates as O(1/h). Optimal h balances truncation and roundoff. Adaptive methods adjust h where |dy/dt| changes rapidly (early decay) vs. slowly (late decay). Fixed small h wastes effort when derivative is tiny. Matching to half-life ignores local behavior. Largest h maximizes truncation error. Thus adaptive stepping optimally allocates computation. This advances beyond basic Euler application to numerical analysis considerations.

📖 Explanation: Solving $dy/dt = -k y^2$: $\int y^{-2} dy = -k \int dt \Rightarrow -1/y = -kt + C$. With $y(0)=y_0$, $C = -1/y_0$. So $1/y = kt + 1/y_0$. Half-life when $y=y_0/2$: $2/y_0 = kT + 1/y_0 \Rightarrow kT = 1/y_0 \Rightarrow T = 1/(k y_0)$. Thus inversely proportional to $y_0$. Contrasts sharply with exponential case. Tests ability to analyze alternative models and recognize uniqueness of exponential decay properties.

📖 Explanation: $y_0/y = 100/120 < 1$, so $\ln(y_0/y) < 0$. With $k>0$, $t<0$. Negative time implies sample formed in future, impossible. Problem arises because measured C-14 exceeds assumed initial amount, violating conservation. Causes include contamination, incorrect $y_0$ assumption, or measurement error. This tests sanity checking of results against physical constraints.

📖 Explanation: While any log base works with adjusted constants, natural log arises naturally from integrating $1/y$, simplifies derivative/integral calculus, and aligns with $e$-based exponential solutions. Conversion factor $\ln(10)$ would clutter formulas. All listed reasons contribute to convention. This acknowledges multiple valid perspectives while reinforcing mathematical coherence.

📖 Explanation: Since $t \propto 1/k$, relative error in $t$ equals relative error in $k$ (to first order). If $k$ increases 1%, $t$ decreases 1%. Fraction $f$ doesn’t affect relative error magnitude here because it’s multiplicative constant in numerator. This illustrates direct inverse proportionality error propagation, distinct from cases where variable appears inside transcendental functions.

📖 Explanation: All radioactive decay follows $dN/dt = -\lambda N$, regardless of isotope. Differences lie in $\lambda$ values and daughter product handling, not ODE order. Branching ratios modify effective $\lambda$ but preserve first-order form. Atmospheric diffusion affects initial conditions, not decay law itself. This generalizes calculus principles across nuclear chronometers, preventing siloed thinking.

📖 Explanation: Given $f = e^{-kt} = e^{-2}$, so $kt = 2$. Since $k = \ln(2)/T$, substitute: $(\ln(2)/T) t = 2 \Rightarrow t = 2T / \ln(2)$. Numerically ~2.885 T. Tests symbolic manipulation connecting exponential parameters to half-life without numerical evaluation.

📖 Explanation: C-14 half-life ~5730 yr; after ~50,000 yr (<10 half-lives), <0.1% remains, below detection. Dinosaurs extinct 65 Myr ago, far beyond limit. C-14 constantly produced, so existed then, but decayed completely. Bones may contain secondary carbon, but primary issue is temporal range. This links mathematical decay limits to geological timescales.

📖 Explanation: General solution is $\ln y = -kt + C$. At $t=0$, $C = \ln y_0$. If student omits C, they imply $C=0$, so $\ln y_0 = 0 \Rightarrow y_0 = 1$. Only then is solution correct. Otherwise, missing constant shifts entire curve. This highlights critical role of initial conditions and common pitfall in separation of variables.

📖 Explanation: Slope >1 means radiocarbon age increases faster than calendar age. For same calendar interval, more radiocarbon years pass, implying less actual decay occurred → higher remaining C-14 → interpreted as younger radiocarbon age. Wait: if radiocarbon age > calendar age, sample appears older than true age, meaning less C-14 than expected → lower past production. Yes: low production → lower initial C-14 → faster apparent decay → older radiocarbon age for given calendar age. Thus slope >1 indicates reduced atmospheric C-14 production. Requires chaining multiple conceptual links.

📖 Explanation: Adaptive RK methods adjust step size based on local error estimates, taking small steps where solution changes rapidly (near t=0) and larger steps where it’s smooth (large t). Fixed methods waste computation or sacrifice accuracy. Analytical solution exists but question specifies numerical context. This evaluates understanding of algorithm selection based on solution behavior.

📖 Explanation: Optimal combination weights by precision (inverse variance). Variances: 100²=10000, 150²=22500. Weights: w1=1/10000, w2=1/22500. Combined = (w1*5000 + w2*5200)/(w1+w2) = (0.0001*5000 + 0.0000444*5200)/(0.0001444) ≈ (0.5 + 0.231)/0.0001444 ≈ 5063. Close to 5056 (rounding). Simple average ignores reliability differences. This applies statistical reasoning to reconcile conflicting measurements in dating.

📖 Explanation: Nuclear decay rates depend only on internal nuclear structure, not external conditions like temperature, pressure, or chemical state (verified experimentally to high precision). This constancy allows universal $k$. Chemical/environmental effects are negligible for beta decay like C-14. This connects calculus model validity to fundamental physics, distinguishing nuclear from chemical kinetics.

📖 Explanation: $k$ has units [time]⁻¹. $\ln(y/y_0)$ is dimensionless. So RHS has units [time]⁻¹, LHS [time]. Mismatch confirms error. Correct form has $1/k$ with units [time]. Dimensional analysis catches structural flaws even without knowing correct formula. This emphasizes units as validation tool in modeling.

📖 Explanation: Mixing model: $dC/dt = -(F/V)C$, so $k = F/V$. Analogous to radioactive $k$. Volume alone isn’t rate; molecular weight irrelevant; inflow concentration affects steady state, not decay rate. This cross-domain analogy reinforces universality of first-order kinetics and helps transfer intuition between contexts.

📖 Explanation: Higher past production → higher initial C-14 in organism → more remaining today than expected for true age → interpreted as less decay → younger radiocarbon age. Uncalibrated date underestimates true age. Calibration curves correct such offsets. Requires simulating perturbation in boundary condition of decay model.

📖 Explanation: For $dy/dt = -ky$, slope magnitude proportional to y. As y→0, slopes flatten, characteristic of exponential approach to zero. Linear decay $dy/dt = -c$ has constant slope. Horizontal constancy would indicate autonomous system dependent only on t, not y. Vertical parallelism indicates dependence only on y, which is true, but doesn’t distinguish exponential from other y-dependent decays. Decreasing slope magnitude with y is signature of proportionality to y.

📖 Explanation: Option B assumes form without proof. Option C shows y’/y is periodic, not necessarily constant; needs additional smoothness/monotonicity to conclude constancy. Option A explicitly solves functional equation and identifies periodic component, then invokes physical regularity to eliminate it, providing complete characterization. This tests deep understanding of functional equations and physical constraints in deriving fundamental laws.