📚 Decay Randomness and Statistical Laws | 衰变的随机性与统计规律
Radioactive decay is one of the most striking examples of quantum indeterminacy in the physical world. When a single unstable nucleus sits in a sample, there is no way to predict exactly when it will decay – we can only assign a probability per unit time. Yet when we observe a large population of such nuclei, a remarkably smooth exponential law emerges. This article explores the random nature of individual decays, the statistical rules that govern large ensembles, and how these ideas are examined in CIE A-Level Physics.
放射性衰变是物理世界中最能体现量子不确定性的现象之一。当一个不稳定的原子核存在于样品中时，我们无法准确预测它将在何时衰变——只能给出单位时间内的衰变概率。然而，当我们观察大量这样的原子核时，一个非常平滑的指数规律便会显现出来。本文探讨单个衰变的随机性、支配大量原子核的统计规律，以及 CIE A-Level 物理中如何考查这些概念。
1. What Does “Random” Mean in Radioactive Decay? | 放射性衰变中的”随机”意味着什么？
In A-Level physics, saying that radioactive decay is random means three things. First, it is impossible to predict which particular nucleus will decay next. Second, it is impossible to predict when a given nucleus will decay, even if we know its entire history. Third, the decay of one nucleus does not influence the decay of another nucleus; each nucleus is independent.
在 A-Level 物理中，说放射性衰变是随机的，包含三方面含义。第一，无法预测哪一个原子核接下来会衰变。第二，即使知道某个原子核的全部历史，也无法预测它何时衰变。第三，一个原子核的衰变不会影响另一个原子核的衰变；每个原子核都是独立的。
This randomness is not due to our lack of measuring instruments or poor experimental technique. It is a fundamental property of nature described by quantum mechanics. Every unstable nucleus has a constant probability of decaying per unit time, regardless of its age, temperature, pressure, or chemical environment (with very rare exceptions).
这种随机性并非源于测量仪器不足或实验技术欠佳，而是量子力学所描述的自然界基本属性。每个不稳定的原子核在单位时间内都具有恒定的衰变概率，与其年龄、温度、压强或化学环境无关（极少数例外情况除外）。
One common misconception is that an “old” nucleus is more likely to decay than a “young” one. In reality, a nucleus that has existed for millions of years has exactly the same probability of decaying in the next second as an identical nucleus created just now. The process has no memory.
一个常见的误解是”老的”原子核比”年轻的”原子核更可能衰变。事实上，一个已经存在数百万年的原子核，在下一秒内衰变的概率与一个刚刚产生的相同原子核完全一样。这一过程没有记忆。
2. The Decay Constant λ | 衰变常数 λ
The decay constant λ is defined as the probability of decay per unit time for a single nucleus. If λ = 0.02 s⁻¹, then each nucleus has a 2% chance of decaying in any one-second interval. Crucially, λ is constant for a given isotope; it does not change with time or with the number of nuclei present.
衰变常数 λ 定义为单个原子核在单位时间内发生衰变的概率。如果 λ = 0.02 s⁻¹，那么每个原子核在任意一秒的时间间隔内都有 2% 的几率发生衰变。关键在于，对于给定的同位素，λ 是常数；它不随时间或原子核数量的变化而改变。
The units of λ are s⁻¹ (or min⁻¹, year⁻¹, etc.). Although these units look like those of a rate constant, λ is a probability, not a rate. The actual rate of decay (number of decays per second) depends on both λ and the number of undecayed nuclei N.
λ 的单位是 s⁻¹（或 min⁻¹、year⁻¹ 等）。虽然这些单位看起来像速率常数的单位，但 λ 是一个概率，而不是速率。实际的衰变速率（每秒的衰变次数）取决于 λ 和未衰变原子核数量 N 两者的乘积。
Activity A = λN
Here, A is the activity measured in becquerels (Bq), where 1 Bq = 1 decay per second. The equation shows that activity is proportional to the number of undecayed nuclei remaining, as long as λ remains constant.
其中，A 是活度，单位为贝克勒尔（Bq），1 Bq = 1 次衰变每秒。该方程表明，只要 λ 保持不变，活度就与剩余的未衰变原子核数目成正比。
3. The Exponential Decay Law | 指数衰变定律
Because each nucleus has the same constant probability of decaying per unit time, the rate of decrease of N is proportional to N itself. This leads to the exponential decay equation:
由于每个原子核在单位时间内具有相同的恒定衰变概率，N 的减少速率与 N 本身成正比。由此得到指数衰变方程：
N = N₀e⁻λᵗ
where N₀ is the initial number of undecayed nuclei, N is the number remaining after time t, and e is the base of natural logarithms. The same form applies to activity:
其中 N₀ 是初始未衰变原子核数目，N 是经过时间 t 后剩余的数目，e 是自然对数的底。同样的形式也适用于活度：
A = A₀e⁻λᵗ
This equation is called exponential decay because the exponent contains t directly. A graph of N against t shows a curve that falls steeply at first and then approaches zero asymptotically – it never quite reaches zero in finite time.
这个方程被称为指数衰变，因为指数中直接包含 t。N 对 t 的图形呈现一条先急剧下降、然后渐近趋近于零的曲线——在有限时间内永远不会真正达到零。
For a linear relationship, we take natural logarithms of both sides:
为了得到线性关系，我们对两边取自然对数：
ln N = ln N₀ − λt
A graph of ln N against t therefore gives a straight line with gradient −λ and intercept ln N₀. This is a common CIE practical analysis technique.
因此，ln N 对 t 的图形是一条斜率为 −λ、截距为 ln N₀ 的直线。这是 CIE 考试中常见的实验数据分析方法。
4. Half-Life and Mean Lifetime | 半衰期与平均寿命
The half-life T₁/₂ is defined as the time taken for half of the unstable nuclei in a sample to decay, or equivalently, the time taken for the activity to fall to half its initial value. It is related to λ by:
半衰期 T₁/₂ 定义为样品中一半不稳定原子核发生衰变所需的时间，或者等价地，活度下降到初始值一半所需的时间。它与 λ 的关系为：
T₁/₂ = ln 2 / λ ≈ 0.693 / λ
The half-life is constant for a given isotope and does not depend on the initial number of nuclei. After one half-life, half remain; after two, one quarter; after three, one eighth, and so on. After n half-lives, the fraction remaining is (1/2)ⁿ.
对于给定的同位素，半衰期是常数，不依赖于初始原子核数目。一个半衰期后剩一半；两个后剩四分之一；三个后剩八分之一，以此类推。经过 n 个半衰期后，剩余比例为 (1/2)ⁿ。
Another useful quantity is the mean lifetime τ, also called the average lifetime. Despite the random nature of decay, we can calculate the average time a nucleus survives. This turns out to be:
另一个有用的量是平均寿命 τ。尽管衰变是随机的，我们仍然可以计算原子核存活的平均时间。结果如下：
τ = 1 / λ
Note that τ is larger than T₁/₂. In fact, τ = T₁/₂ / ln 2 ≈ 1.44 T₁/₂. The reason is that a few nuclei survive for many half-lives, pulling the average upward.
注意 τ 大于 T₁/₂。实际上，τ = T₁/₂ / ln 2 ≈ 1.44 T₁/₂。原因是少数原子核会存活很多个半衰期，从而拉高了平均值。
5. Statistical Fluctuations | 统计涨落
If we measure the count rate from a radioactive source repeatedly over equal time intervals, we rarely obtain exactly the same number of counts each time. This variation is called statistical fluctuation or random fluctuation, and it is a direct consequence of the randomness of individual decays.
如果我们对同一放射源在相同时间间隔内反复测量计数率，每次得到的计数很少完全相同。这种变化称为统计涨落或随机涨落，是单个衰变随机性的直接后果。
For a fixed average count m, the actual counts in different intervals follow a Poisson distribution. The standard deviation of the counts is given by:
对于固定的平均计数 m，不同时间间隔内的实际计数服从泊松分布。计数的标准差为：
σ = √m
For example, if the average count in 10 seconds is 100, the standard deviation is √100 = 10. About 68% of measurements will lie between 90 and 110 counts, and about 95% will lie between 80 and 120 counts.
例如，若 10 秒内的平均计数为 100，则标准差为 √100 = 10。约 68% 的测量值将落在 90 到 110 之间，约 95% 将落在 80 到 120 之间。
This has practical implications. To reduce the percentage uncertainty in a count measurement, we must increase the total number of counts. Since σ/m = √m / m = 1/√m, the fractional uncertainty decreases as the count increases.
这具有实际意义。为了减小计数测量的百分比不确定度，我们必须增加总计数。由于 σ/m = √m / m = 1/√m，分数不确定度随计数的增加而减小。
6. Why Large Samples Give Smooth Curves | 为什么大样本给出平滑曲线
The exponential decay law is deterministic in the sense that it makes a very precise prediction for the number of undecayed nuclei remaining after time t. But this prediction is reliable only because the number of nuclei is huge. A typical radioactive sample contains billions or trillions of nuclei; individual random events average out.
指数衰变定律在某种意义上是确定性的：它对经过时间 t 后剩余未衰变原子核的数量做出了非常精确的预测。但这种预测之所以可靠，仅仅因为原子核的数量极其庞大。一个典型的放射性样品含有数十亿甚至数万亿个原子核；单个随机事件被平均掉了。
Consider an analog: flipping a coin. One flip is unpredictable. But if we flip a coin 10,000 times, we expect very close to 5,000 heads and 5,000 tails. Similarly, with billions of nuclei, the fraction decaying per second is extremely close to λ, even though each individual nucleus decays unpredictably.
考虑一个类比：抛硬币。一次抛掷无法预测。但如果抛 10,000 次，我们预计会非常接近 5,000 次正面和 5,000 次反面。类似地，对于数十亿个原子核，每秒衰变的比例极其接近 λ，尽管每个原子核的衰变是不可预测的。
However, as a sample decays and N becomes very small, fluctuations become relatively more important. In a sample with only 10 nuclei remaining, the actual decay time will deviate noticeably from the smooth exponential prediction. In A-Level experiments, we often use a “long-lived” source so that N remains effectively constant over the measurement period, reducing relative fluctuations.
然而，随着样品不断衰变且 N 变得非常小，涨落会变得相对更加重要。在一个只剩下 10 个原子核的样品中，实际的衰变时间会明显偏离平滑的指数预测。在 A-Level 实验中，我们通常使用”长寿命”放射源，使 N 在测量期间内基本保持不变，从而减小相对涨落。
7. The Poisson Distribution in Decay Counting | 衰变计数中的泊松分布
When counting random events such as radioactive decays, the number of counts n observed in a fixed time interval follows the Poisson distribution. The probability of observing exactly n counts when the mean is m is:
当对放射性衰变等随机事件进行计数时，在固定时间间隔内观察到的计数 n 服从泊松分布。当平均值为 m 时，观察到恰好 n 个计数的概率为：
P(n) = mⁿ e⁻ᵐ / n!
For large m, the Poisson distribution becomes approximately symmetric and approaches a Gaussian (normal) distribution. This is why for large counts, we can safely use the ±√m uncertainty rule.
当 m 较大时，泊松分布近似对称并趋近于高斯（正态）分布。这就是为什么对于较大的计数，我们可以放心使用 ±√m 的不确定度规则。
In a CIE practical paper, you may be asked to estimate the uncertainty in a count rate. If a background count of 40 is recorded in 5 minutes, the uncertainty is √40 ≈ 6.3 counts, giving a background count rate of 8 ± 1.3 counts per minute.
在 CIE 实验卷中，你可能会被要求估计计数率的不确定度。如果 5 分钟内记录了 40 个本底计数，则不确定度为 √40 ≈ 6.3 个计数，因此本底计数率为 8 ± 1.3 个每分钟。
8. Background Radiation and Corrected Count Rates | 本底辐射与修正计数率
No matter how well we shield a detector, it always records some counts that do not come from the source under study. This is called background radiation. Sources include cosmic rays, natural radioisotopes in rocks and building materials, and traces of radioactive gases in the air.
无论探测器屏蔽得多好，它总会记录到一些并非来自所研究放射源的计数。这称为本底辐射。其来源包括宇宙射线、岩石和建筑材料中的天然放射性同位素，以及空气中微量的放射性气体。
To obtain the corrected count rate from the source alone, we take the measured gross count rate and subtract the background count rate:
为了获得仅来自放射源的修正计数率，我们需要用测得的总计数率减去本底计数率：
Corrected rate = (Gross count / time) − (Background count / time)
Both the gross and background counts have statistical uncertainties. The uncertainty in the corrected rate is found by adding the uncertainties in quadrature:
总计数和本底计数都具有统计不确定度。修正计数率的不确定度通过将两者的不确定度按平方和合成：
σ_total = √(σ_gross² + σ_background²)
For example, if gross count = 500 in 50 s and background count = 100 in 50 s, then corrected rate = (500−100)/50 = 8 s⁻¹. The uncertainties are √500 ≈ 22.4 and √100 = 10, so σ_total = √(22.4² + 10²) ≈ 24.6 counts, giving an uncertainty in rate of 24.6/50 ≈ 0.49 s⁻¹.
例如，若总计数为 50 秒内 500 次，本底计数为 50 秒内 100 次，则修正计数率 = (500−100)/50 = 8 s⁻¹。不确定度分别为 √500 ≈ 22.4 和 √100 = 10，因此 σ_total = √(22.4² + 10²) ≈ 24.6 个计数，所以计数率的不确定度为 24.6/50 ≈ 0.49 s⁻¹。
9. Simulated Decay and the Importance of Large N | 模拟衰变与大 N 的重要性
A classic classroom demonstration of decay statistics uses dice. Throw 200 dice, remove those that show a “6”, count the remainder, and repeat. Each die has a fixed probability of 1/6 of being removed per throw, mimicking a decay constant of λ = 1/6 per throw.
一个经典的课堂演示使用骰子模拟衰变统计。抛掷 200 个骰子，移走显示”6″的骰子，数剩余数量，然后重复。每个骰子每轮被移走的概率固定为 1/6，模拟衰变常数 λ = 1/6 每轮。
With a large number of dice, the removal process closely follows an exponential decay curve with half-life T₁/₂ = ln2/(1/6) ≈ 4.16 throws. But if we repeat the experiment with only 20 dice, the results from each run will scatter noticeably around the ideal exponential curve.
当骰子数量很大时，移除过程非常接近半衰期 T₁/₂ = ln2/(1/6) ≈ 4.16 轮的指数衰变曲线。但如果仅用 20 个骰子重复实验，每次运行的结果会明显偏离理想的指数曲线。
This simulation illustrates a central idea: the exponential law is a statistical average. It describes what happens on average, not what happens in any particular small sample. The law becomes more accurate as N increases.
这个模拟说明了一个核心思想：指数定律是一种统计平均。它描述的是平均会发生什么，而不是任何特定小样本中会发生什么。随着 N 的增大，定律变得更加精确。
In the laboratory, we must also remember that a detector cannot record every decay. Some decays may occur when the detector is “dead” (unable to record), and some gamma or beta particles may be absorbed before reaching the detector. The count rate is therefore always lower than the true activity.
在实验室中，我们还必须记住探测器无法记录每一次衰变。有些衰变可能发生在探测器的”死时间”内，有些 β 或 γ 粒子可能在到达探测器之前被吸收。因此，计数率总是低于真实活度。
10. Applications in Carbon Dating and Medicine | 在碳定年和医学中的应用
The statistical nature of decay does not prevent its practical use; it simply requires careful accounting of uncertainty. In carbon dating, the ratio of carbon-14 to carbon-12 in an ancient artifact is measured. Because the decay of carbon-14 is random, the measured ratio has an associated uncertainty, which sets a limit on the precision of the dating.
衰变的统计性质并不妨碍其实际应用，只是要求仔细处理不确定度。在碳定年中，我们测量古代文物中碳-14 与碳-12 的比例。由于碳-14 的衰变是随机的，测得的比例带有相关的不确定度，这限制了定年的精度。
For an object around 5,000 years old, the fraction of carbon-14 remaining is about 55%. A one-standard-deviation uncertainty in the measurement might correspond to an uncertainty of ±100 years in age. For older objects, the fraction of carbon-14 is smaller and the relative uncertainty grows, so dating becomes less precise.
对于一个约 5,000 年历史的物体，剩余碳-14 的比例约为 55%。测量中一个标准差的不确定度可能对应年龄上 ±100 年的不确定度。对于更古老的物体，碳-14 的比例更小，相对不确定度增大，因此定年的精度变得更差。
In medical radiotherapy, the randomness of decay means that the dose delivered to a tumor over a given time has a tiny statistical fluctuation. However, because billions of decays occur each minute, the fractional fluctuation is negligible for typical therapeutic doses. The same statistics also govern the noise in PET scans and gamma cameras.
在医疗放射治疗中，衰变的随机性意味着在给定时间内输送到肿瘤的剂量存在微小的统计涨落。然而，由于每分钟发生数十亿次衰变，对于典型的治疗剂量而言，这种分数涨落可以忽略不计。同样的统计学也支配着 PET 扫描和伽马相机中的噪声。
Engineers and physicists designing radiation shielding must account for the Poisson statistics of background radiation to decide how long to count in order to achieve a desired precision. Doubling the counting time increases the total count, but only reduces the fractional uncertainty by a factor of √2.
设计辐射屏蔽的工程师和物理学家必须考虑本底辐射的泊松统计，以决定需要计数多长时间才能达到所需的精度。将计数时间加倍会增加总计数，但只能将分数不确定度降低 √2 倍。
11. Common Exam Calculations and Pitfalls | 常见考试计算与易错点
One common exam question asks you to use the activity equation A = λN. If you are given the half-life, first convert it to λ using λ = ln2/T₁/₂. Then substitute N and compute A. Keep careful track of units: if T₁/₂ is in seconds, λ is in s⁻¹.
一个常见的考试题目要求你使用活度方程 A = λN。如果已知半衰期，先用 λ = ln2/T₁/₂ 将其转换为 λ。然后代入 N 并计算 A。务必注意单位：若 T₁/₂ 以秒为单位，则 λ 的单位为 s⁻¹。
Another classic pitfall is confusing count rate with activity. Count rate is what the detector records; activity is the actual number of decays per second in the source. They are related by the detection efficiency, which is always less than 100%, plus a correction for background radiation.
另一个经典易错点是混淆计数率与活度。计数率是探测器记录的数值；活度是放射源中每秒实际发生的衰变次数。它们之间通过探测效率（通常小于 100%）以及本底辐射修正相关联。
A third common error: when calculating the number of half-lives, students mistakenly divide the total time by the half-life and round incorrectly. For example, after 3.5 half-lives, the fraction remaining is (1/2)³·⁵ ≈ 0.0884, not (1/2)³ = 0.125.
第三个常见错误：计算半衰期个数时，学生错误地将总时间除以半衰期并四舍五入。例如，经过 3.5 个半衰期后，剩余分数为 (1/2)³·⁵ ≈ 0.0884，而不是 (1/2)³ = 0.125。
Finally, when asked about the randomness of decay, avoid vague phrases like “it is unpredictable.” Be precise: each nucleus has a constant probability λ per unit time of decaying; the process has no memory; individual decay times cannot be predicted; and the exponential law only applies to large populations.
最后，当被问及衰变的随机性时，避免使用”不可预测”之类的模糊表述。要准确表述：每个原子核在单位时间内具有恒定的衰变概率 λ；这一过程没有记忆；单个衰变时间无法预测；指数定律只适用于大数量的原子核群体。
12. Summary of Key Equations | 关键公式总结
The following table summarises the most important equations from this topic. You should be able to derive them quickly and explain every symbol:
下表总结了本主题最重要的公式。你应该能够快速推导并解释每个符号的含义：
| Quantity | Equation | Notes |
| Activity | A = λN | N = number of undecayed nuclei |
| Decay law | N = N₀e⁻λᵗ | Also applies to A and count rate |
| Half-life relation | T₁/₂ = ln2 / λ | ln2 ≈ 0.693 |
| Mean lifetime | τ = 1 / λ | τ ≈ 1.44 T₁/₂ |
| Linearised decay | ln N = ln N₀ − λt | Gradient = −λ |
| Poisson uncertainty | σ = √m | m = mean count |
Mastering these equations and understanding the physical meaning behind each one will help you tackle both calculation and explanation questions in the CIE A-Level physics paper with confidence.
掌握这些公式并理解每一条公式背后的物理含义，将帮助你在 CIE A-Level 物理试卷中自信地应对计算题和解释题。
Published by TutorHao | Physics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导