Exponentials, logarithms and modelling
引入| English | 释义 | 拼音 |
|---|---|---|
| half-life/hɑːf laɪf/ | 半衰期 | bàn shuāi qī |
A man from the ice, September 1991
- On 19 September 1991, two hikers in the Ötztal Alps on the Austrian–Italian border found a body in melting ice. Radiocarbon dating showed "Ötzi" had died about 5,300 years ago. Carbon-14 decays exponentially with a half-life 半衰期 of about 5,730 years, so measuring how much is left in a sample gives its age through a logarithm.
- Exponential models $N = N_0e^{-kt}$ describe decay; logarithms solve them for time.
The exponential model
- $N = N_0e^{kt}$: growth if $k > 0$, decay if $k < 0$. $N_0$ is the value at $t = 0$.
- The rate of change is proportional to the amount: $\dfrac{dN}{dt} = kN$.
支持 $N = 250e^{0.04t}$, find $N$ when $t = 0$.
难度 2/5 · 今日首次完成可获 140 金币及少量额外奖励
$e^0 = 1$.
Half-life
- After one half-life $T$, $N = \tfrac{1}{2}N_0$: $e^{-kT} = \tfrac{1}{2}$, so $k = \dfrac{\ln 2}{T}$.
- For carbon-14: $k = \dfrac{0.693}{5730} = 1.21 \times 10^{-4}$ per year.
Exponentials, logarithms and modelling
For y=Ae^(kt), k is a proportional rate
Explain why the half-life depends on 0.2 and not on the initial amount 80.
A substance has a half-life of 8 days. Find $k$ 入 $N = N_0e^{-kt}$, to 3 s.f.
难度 2/5 · 今日首次完成可获 140 金币及少量额外奖励
$\tfrac{\ln 2}{8}$.
What fraction of a sample remains after three half-lives?
难度 2/5 · 今日首次完成可获 140 金币及少量额外奖励
$\left(\tfrac{1}{2}\right)^3$.
Worked example: dating a sample
A sample of Ötzi's tissue has 53% of the carbon-14 of living tissue. Estimate its age.
- Model: $\tfrac{N}{N_0} = e^{-kt}$ with $k = \dfrac{\ln 2}{5730}$.
- Logarithm: $0.53 = e^{-kt}$, so $t = \dfrac{-\ln 0.53}{k} = \dfrac{0.635 \times 5730}{0.693}$.
- Answer: $t \approx 5250$ years — consistent with the 5,300 years from the full analysis.
A wooden tool has 70% of the carbon-14 of living wood. Estimate its age to the nearest 100 years (half-life 5730 years).
难度 2/5 · 今日首次完成可获 140 金币及少量额外奖励
$\tfrac{-\ln 0.7 \times 5730}{\ln 2} = 2948$.
Fitting a model from two points
- A population is 800 at $t = 0$ and 1200 at $t = 5$: $1200 = 800e^{5k}$, so $k = \tfrac{1}{5}\ln 1.5 = 0.0811$.
- Predict at $t = 10$: $800e^{0.811} = 1800$.
A culture grows from 500 to 2000 cells in 6 hours. Find $k$ 入 $N = 500e^{kt}$, to 3 s.f.
难度 2/5 · 今日首次完成可获 140 金币及少量额外奖励
$\tfrac{\ln 4}{6}$.
Limits of exponential models
- Growth cannot continue for ever: food and space run out, so real populations level off.
- Decay models work well for radioactivity, but predictions far beyond the data are less reliable for living systems.
$N_0$ is the value at $t = 0$, and $k$ is in the exponent. In $N = 80e^{-0.2t}$, 80 is the starting amount and $-0.2$ is the rate constant; the amount does not fall by 0.2 each unit of time.
In $N = 80e^{-0.2t}$, the amount falls by 0.2 each unit of time.
难度 1/5 · 今日首次完成可获 100 金币及少量额外奖励
It falls by a fixed proportion, not a fixed amount.
$N = N_0e^{kt}$, with $N_0$ the initial value and $k$ the rate constant. Half-life $T$ gives $k = \tfrac{\ln 2}{T}$. Take logarithms to find $k$ from two points or $t$ from a ratio, and judge whether the model still holds over the time asked about.
A cup of coffee contains 95 mg of caffeine, which the body removes with a half-life of 5 hours. How long until less than 20 mg remains? Comment on the model.
Half-life in context.