What Is Exponential Decay? (Math & Physics Guide)
Exponential decay describes a quantity that falls by the same fraction over equal time intervals. Its continuous model is (Q(t)=Q_0e^{-\lambda t}); its discrete model is (Q_n=Q_0(1-r)^n). The decay constant, percentage loss, and half-life are related but not interchangeable. Matching the model and time units is essential for sound calculations.
Understanding decay can help you avoid costly errors when reading measurements, comparing products, or planning long-term use. A model may estimate how a quantity changes, but only if its assumptions fit the situation. I’ll show how to choose a model, calculate with it, and check whether an answer makes sense.
Exponential Decay Overview: Definition, Models, and When to Use Them
Exponential decay is a pattern in which a quantity decreases by a fixed proportion over equal intervals, rather than by a fixed amount. The amount lost therefore gets smaller as the quantity itself gets smaller. This pattern can be described with a continuous-time equation or a step-by-step equation.
What exponential decay means
The word “exponential” refers to a variable in the exponent, such as (t) in (e^{-\lambda t}). The word “decay” means the quantity goes down over time. A constant fraction is lost over equal intervals, so the amount remaining keeps changing, even when the percentage loss stays the same.
For example, a quantity that loses 10% each year does not lose the same number of units each year. It loses 10% of what remains. If it starts at 100 units, it has 90 after one year, then 81 after two years.
Continuous and discrete models
The continuous model tracks change at every instant. The discrete model applies a fixed percentage loss at set intervals, such as once per day or year. Use the model that matches how the change is described or measured; converting between them requires care.
| Model | Equation | Meaning of the loss parameter | Example time measure |
|---|---|---|---|
| Continuous | (Q(t)=Q_0e^{-\lambda t}) | (\lambda) is a rate per unit of time | Hours |
| Discrete | (Q_n=Q_0(1-r)^n) | (r) is the fraction lost per step | Days or years |
In both equations, (Q_0) is the starting quantity. In the continuous model, (t) is elapsed time and (\lambda>0) is the decay constant. In the discrete model, (n) is the number of equal steps and (0<r<1). A loss of 10% means (r=0.10), so the fraction remaining per step is (1-r=0.90).
Core Equations, Parameters, and Worked Requirements
The equations connect a starting amount to the amount that remains. Before calculating, identify the starting value, the kind of decay model, and the units for time. Then substitute values, solve, and check that the result fits the situation.
Reading the continuous decay equation
In (Q(t)=Q_0e^{-\lambda t}), (Q(t)) is the amount at time (t), and (Q_0) is the amount at time zero. The symbol (e) is a mathematical constant, about 2.718. The decay constant (\lambda) must use inverse time, such as per hour when (t) is measured in hours.
The equation also follows from the rate law:
[ \frac{dQ}{dt}=-\lambda Q. ]
This says that the instantaneous rate of decrease is proportional to the amount present. The minus sign marks a decrease. If more of the quantity is present, its rate of loss is greater; as less remains, the rate slows.
Units matter because the exponent (-\lambda t) must be unit-free. If (\lambda=0.2) per hour, use (t) in hours. If the time is in minutes, convert either the time or the rate before calculating.
Using half-life
Half-life is the time required for a quantity to fall to half its starting amount. For continuous decay, it is linked to the decay constant by (t_{1/2}=\ln(2)/\lambda). This gives a useful way to model decay when a half-life is known instead of a decay constant.
The equivalent equation is:
[ Q(t)=Q_0\left(\frac12\right)^{t/t_{1/2}}. ]
Suppose a quantity starts at 80 units and has a half-life of 6 hours. After 12 hours, two half-lives have passed:
[ Q(12)=80\left(\frac12\right)^{12/6}=20. ]
After 18 hours, three half-lives have passed, leaving 10 units. The model predicts a smaller amount after each half-life, but it does not predict exactly zero at any finite time.
To find the decay constant in this example, calculate (\lambda=\ln(2)/6), or about (0.1155) per hour. Keep the units attached to the answer.
Finding the time or the percentage lost
Logarithms help solve for elapsed time when you know the starting amount and the remaining amount. Rearranging the continuous model gives:
[ t=\frac{\ln(Q_0/Q)}{\lambda}. ]
For example, if (Q_0=100), (Q=25), and (\lambda=0.2) per day, then (t=\ln(4)/0.2), or about 6.93 days. The result is in days because the decay constant is per day.
For a discrete model, the remaining amount after (n) steps is multiplied by (1-r) each step. If 10% is lost per step, then (Q_n=Q_0(0.90)^n). Do not use (r) as though it were the continuous decay constant.
Learning the Topic: Prerequisite Skills, Graphs, and Common Checks
You need a few basic tools to work with decay: exponent rules, logarithms, proportional change, and unit conversion. You do not need to memorize every rearrangement. Knowing what each symbol means makes it easier to choose an equation and judge whether your answer is reasonable.
Reading an exponential decay graph
A graph of (Q(t)=Q_0e^{-\lambda t}) starts at (Q_0), decreases smoothly, and bends toward the horizontal axis. It falls quickly at first, then more slowly. In the ideal model, the curve approaches zero as time grows but never reaches zero at a finite time.
The starting slope is (-\lambda Q_0). A larger (\lambda) means faster decay, when the time unit is the same. A larger half-life means slower decay. These comparisons only work when the units match.
Checking your calculation
Use this short check after solving:
- At (t=0), does the equation give (Q_0)?
- Is the predicted amount positive and no greater than (Q_0) for (t\ge0)?
- Does the quantity decrease as time increases?
- Do the time units match (\lambda) or the half-life?
- Does the answer fit the stated model and its assumptions?
A frequent error is treating exponential decay as a fixed amount subtracted over time. Losing 10 units per day is linear change. Losing 10% of the current amount per day is discrete exponential decay. Another common error is confusing a percentage per step with a continuous rate. These values describe different models.
| Given information | Use | Check |
|---|---|---|
| A fraction lost in equal steps | (Q_n=Q_0(1-r)^n) | (r) is a decimal fraction |
| A continuous decay constant | (Q(t)=Q_0e^{-\lambda t}) | (\lambda) uses inverse-time units |
| Time to halve | (Q(t)=Q_0(1/2)^{t/t_{1/2}}) | (t) and (t_{1/2}) use the same unit |
Applications and Pathways: Physics, Chemistry, Engineering, and Data Analysis
Exponential decay is useful when evidence or a model supports a constant proportional rate of loss. It appears in several fields, but the equation alone does not prove that a real process follows this pattern. Check the conditions and measurements before relying on a prediction.
Worked examples across fields
These are illustrative calculations, not reports of specific experiments. They show how the same math can describe different quantities. The model is useful only when its assumptions suit the process.
| Context | Example model input | Prediction |
|---|---|---|
| Radioactive decay | 40 mg, half-life 5 years | 10 mg after 10 years |
| Drug elimination approximation | 200 mg, half-life 4 hours | 50 mg after 8 hours |
| Depreciation model | 1,000 units of value, 20% loss per year | 640 units after 2 years |
In the first two examples, the calculation uses half-lives: two half-lives leave one-quarter of the starting amount. In the depreciation example, the discrete model gives (1{,}000(0.80)^2=640). A real medicine’s concentration in the body can depend on absorption, dosing, and other factors; this simple example does not replace medical guidance. Depreciation methods also vary, so the example is not a valuation rule.
Limits of the model
A decay model describes a pattern; it does not explain its cause. Real measurements may not follow a single exponential curve over all times. Conditions can change, data can be noisy, and a process may include more than one rate. Compare predictions with available measurements and revise the model when the evidence calls for it.
For an introductory calculation, state the model, parameters, units, and time range. For decisions involving health, safety, or money, use the relevant professional or official guidance rather than relying on a simplified equation alone.
Conclusion and FAQ
Exponential decay means a fixed proportion is lost over equal intervals. Choose the continuous or discrete equation based on the process, keep units consistent, and check the result against the starting amount and expected trend. Half-life offers a clear way to compare rates, but it does not mean a quantity reaches zero after a set number of intervals.
What is exponential decay in simple terms?
Exponential decay is a steady percentage drop over equal time intervals. The amount lost gets smaller as less remains. For instance, losing 10% of the current amount is exponential decay, while subtracting 10 units each time is linear change.
What is the formula for exponential decay?
For continuous decay, use (Q(t)=Q_0e^{-\lambda t}). Here, (Q_0) is the starting amount, (t) is elapsed time, and (\lambda) is the decay constant. For a fixed percentage loss each step, use (Q_n=Q_0(1-r)^n).
What does the decay constant mean?
The decay constant (\lambda) measures the proportional rate of continuous decay. A larger value means faster decay for the same time unit. It has units such as per second or per year, not units of the quantity itself.
How do you calculate half-life from the decay constant?
Use (t_{1/2}=\ln(2)/\lambda). If (\lambda) is measured per hour, the half-life will be in hours. Make sure the rate and time units are compatible before using the result.
How do you find the amount left after several half-lives?
Divide the elapsed time by the half-life to find the number of half-lives. Then use (Q=Q_0(1/2)^{t/t_{1/2}}). Two half-lives leave one-quarter of the starting amount; three leave one-eighth.
Does exponential decay ever reach zero?
In the ideal continuous model, the quantity stays positive at every finite time and approaches zero as time increases. A real measurement may appear as zero if the amount is too small to detect, but that is a measurement limit.
What is the difference between exponential and linear decay?
Exponential decay removes a fixed fraction of the amount remaining during each equal interval. Linear decay subtracts a fixed amount during each interval. The two patterns produce different results, so identify which type the problem describes.
How do you choose between continuous and discrete decay?
Use a discrete model when the loss is stated per fixed step, such as a percentage each year. Use a continuous model when the process is represented as changing at every instant or when a continuous decay constant is supplied. Match the equation to the given assumptions.
(This article was written by one of our staff writers, Nathan Sterling. Visit our Meet the Team page.)
