How to Calculate Decay Using the Exponential Decay Formula

Calculating decay with the exponential decay formula comes down to one equation and three pieces of information: the starting amount, the decay constant, and how much time has passed. The formula is N(t) = N₀ × e^(−λt), where N₀ is what you start with, λ (lambda) is the decay constant, t is time, and e is the mathematical constant roughly equal to 2.718. Plug in those values and you get the amount remaining at any point in the future. The real skill is knowing where λ comes from and recognizing when the formula applies cleanly versus when reality gets more complicated.

What Each Piece of the Formula Means

The exponential decay formula works because it captures a specific pattern: the rate at which something disappears is proportional to how much is left. A large pile shrinks quickly; a small pile shrinks slowly. That proportional relationship produces the smooth, curving decline that shows up in radioactive decay, drug clearance, capacitor discharge, and dozens of other processes.

Here is what each variable represents in practical terms:

N(t): The amount remaining after time t. This is what you are solving for.
N₀: The initial amount at time zero. Could be milligrams of a drug, atoms of a radioactive isotope, volts across a capacitor, or grams of a pollutant in soil.
λ: The decay constant, sometimes written as k in chemistry and ecology. It tells you the fraction of the remaining quantity that decays per unit of time. A larger λ means faster decay.
t: Time elapsed, measured in whatever units λ uses. If λ is per year, t must be in years.
e: Euler’s number, approximately 2.71828. Your calculator has an “e^x” button; use it.

The formula is sometimes written as N(t) = N₀ × e^(−kt), where k replaces λ. The meaning is identical. Ecology and environmental science tend to use k; physics and pharmacology lean toward λ. Watch the sign: the exponent must be negative for decay. If it is positive, you are modeling growth, not decay.

Getting the Decay Constant from a Half-Life

Most people do not start with a decay constant handed to them. What you usually know is the half-life: the time it takes for half the material to disappear. Converting between the two is straightforward. The relationship is λ = ln(2) / t½, where ln(2) is the natural logarithm of 2, approximately 0.693. So if a substance has a half-life of 10 days, its decay constant is 0.693 / 10 = 0.0693 per day.

This relationship appears across fields. In environmental modeling, researchers studying pollutant degradation use the same conversion: for each environmental compartment, the overall degradation rate is the sum of individual rate constants for processes like photolysis and biodegradation, and the half-life equals ln(2) divided by that total rate constant.¹ You can run the conversion in reverse, too. If you know λ from experimental data, multiply it into the formula t½ = 0.693 / λ to get the half-life.

A Worked Example

Suppose you have 500 milligrams of a substance with a half-life of 4 hours, and you want to know how much remains after 10 hours. First, find the decay constant: λ = 0.693 / 4 = 0.173 per hour. Then plug everything in: N(10) = 500 × e^(−0.173 × 10) = 500 × e^(−1.73). Punch e^(−1.73) into a calculator and you get roughly 0.177. Multiply: 500 × 0.177 = about 88.5 milligrams remaining.

You can sanity-check this with the half-life directly. After 4 hours, half remains: 250 mg. After 8 hours, half again: 125 mg. At 10 hours, you are partway between the second and third half-lives, so something between 125 and 62.5 mg makes sense. The 88.5 mg answer sits right in that range.

If you prefer working directly with half-lives instead of decay constants, there is an equivalent formula: N(t) = N₀ × (1/2)^(t/t½). For the same example, N(10) = 500 × (0.5)^(10/4) = 500 × (0.5)^2.5 = 500 × 0.177 = 88.5 mg. Same result, no need to compute λ first. Use whichever version feels more natural.

Solving for Time or for the Decay Constant

Sometimes the question runs backward. You know the starting amount, the amount remaining, and either the decay constant or the time, and you need to solve for the missing piece. This requires taking the natural logarithm of both sides.

To find how long it takes for a quantity to drop to a certain level: rearrange to t = −ln(N(t)/N₀) / λ. If you started with 1,000 units, need to know when you will hit 200 units, and λ = 0.05 per day, then t = −ln(200/1000) / 0.05 = −ln(0.2) / 0.05 = 1.609 / 0.05 = about 32.2 days.

To find the decay constant when you have measurements at two different times: λ = −ln(N₂/N₁) / (t₂ − t₁). Measure 800 units at day 3 and 450 units at day 12, and you get λ = −ln(450/800) / (12 − 3) = −ln(0.5625) / 9 = 0.576 / 9 = 0.064 per day. From there the half-life is 0.693 / 0.064 = about 10.8 days. This two-point method is simple but sensitive to measurement error. Researchers often fit curves to many data points instead, but the two-point approach is perfectly fine for quick estimates.

Real-World Processes That Follow the Formula Well

The exponential decay model works best when each unit of the decaying substance has an independent, constant probability of disappearing in any given time interval. Radioactive decay is the textbook case: each atom has a fixed chance of decaying regardless of what its neighbors are doing, and no external condition changes that probability. This is why radioactive half-lives are so reliable and why carbon dating works.

Many chemical and biological processes also follow this pattern closely enough for practical use. Pharmaceutical compounds in the body typically clear following first-order kinetics at normal therapeutic doses, meaning the rate of elimination is proportional to the drug’s concentration in blood.² The same first-order model describes the degradation of many pollutants in soil. Studies of pharmaceuticals and personal care products in agricultural soils found degradation generally followed first-order exponential decay, with half-lives ranging from less than a day to about 20 days depending on the compound and soil composition.³ Similar patterns hold for pesticides, where degradation half-lives in non-sterile soil ranged from roughly 8 to 44 days.⁴

Ecologists use the same formula to model how dead wood decomposes on a forest floor. The equation is typically written as WD = a × e^(−kt), where WD is wood density at time t and k is the decay rate constant.⁵ These k values vary dramatically by tree species and environment. One study in the Qinling Mountains found that a pine species had a decay rate constant of about 0.04 per year, meaning it would take roughly 16 years for half its wood to decompose, while an oak species decayed about twice as fast, reaching the 50% mark in about 10 years.⁶ The same study estimated it would take 67 years for 95% of the pine to decompose, compared with 44 years for the oak. Air pollution can slow things down further, reducing the wood decomposition rate constant by anywhere from 16% to 60%.⁷

When the Simple Formula Breaks Down

The exponential decay formula assumes one clean process with one constant rate. Reality often layers multiple processes on top of each other, and when it does, the single-exponential model stops fitting the data well.

A striking example comes from virology. When antiviral therapy is given for hepatitis B, the decline in viral load is not a smooth single curve. Instead, researchers observe two distinct phases. The first, fast phase reflects free virus particles being cleared from the bloodstream with a half-life of about 1.1 days. The second, slower phase reflects infected cells gradually dying off, with a half-life of roughly 18 days.⁸ Each phase individually looks roughly exponential, but the overall curve is biphasic: fast decline followed by a slower tail. The same biphasic pattern appears in hepatitis C treatment, where viral RNA drops sharply in the first one to two days and then shifts to a slower decline as virus-producing cells are eliminated.⁹ If you tried to fit a single exponential to the full curve, you would get a decay constant that is too slow for the early phase and too fast for the late phase, and your predictions at either end would be off.

Drug metabolism offers another common exception. Most drugs clear from the body following the exponential pattern at normal doses. But when metabolic pathways become saturated, as happens with alcohol or certain medications at high concentrations, the clearance rate becomes constant regardless of how much drug is in the blood. This is called zero-order kinetics, and it means a fixed amount is removed per hour instead of a fixed fraction.² The time it takes to halve the concentration then depends on where you started, which makes dosing predictions trickier and makes the standard half-life formula unreliable.

Even in physics, there are edge cases. The discharge of a capacitor through a resistor is usually presented as a clean exponential decay. But experiments with certain capacitor configurations show that the voltage can oscillate in a damped pattern rather than decaying smoothly. The standard exponential curve turns out to describe only the overdamped case, where resistance is high enough to suppress oscillation.¹⁰ And at the quantum level, unstable systems do not always begin decaying exponentially. Researchers have observed that trapped atoms initially show non-exponential decay before settling into the expected exponential behavior, and that the act of measuring the system during this early phase can either slow or speed the decay.¹¹

How to Spot Whether Your Data Actually Follows Exponential Decay

Before relying on the formula, you want to know whether your data truly decays exponentially. The classic check is to plot the natural logarithm of the remaining quantity against time. If the data follows exponential decay, this log-transformed plot should give you a straight line whose slope equals −λ. A curved log-plot means something else is going on: biphasic decay, a changing rate constant, or a process that does not follow the simple model.

In environmental science, researchers routinely fit first-order models to pollutant data and then check whether the fit is good enough by looking at the R² value and whether the residuals show a pattern. The wood decomposition studies mentioned earlier reported R² values above 0.97, indicating that the single exponential fit captured the data well.⁶ But soil degradation data sometimes shows more scatter, particularly when microbial activity and soil composition vary across samples.³ If your straight-line fit on the log-plot is poor, consider whether the system involves multiple overlapping processes that each decay at their own rate.

When there are clearly two phases, the standard approach is to fit two separate exponential curves to the two segments. You extract a fast rate constant from the early data and a slow rate constant from the late data. The full model becomes a sum of two exponentials: N(t) = A × e^(−λ₁t) + B × e^(−λ₂t), where A and B are the proportions that decay at each rate. This is exactly what virologists do when modeling biphasic viral decline during antiviral treatment.⁸

Practical Tips That Textbooks Skip

Unit consistency trips people up more often than any conceptual difficulty. If your half-life is in hours and you want to know the remaining amount after a number of minutes, convert one or the other before plugging in. The formula does not care what units you use, as long as λ and t match. Similarly, N₀ can be in any units: grams, atoms, volts, concentration. The result will be in the same units.

When working with very long or very short time scales, the numbers inside the exponent can become extreme. If λt is larger than about 30 or 40, e^(−λt) is effectively zero, and your calculator may round to exactly zero. This is fine for practical purposes: after enough half-lives, the remaining amount is negligible. Specifically, after 10 half-lives, less than 0.1% of the original amount remains. After 20 half-lives, you are down to about one millionth.

In pharmacology, clinicians sometimes need to estimate a drug’s decay constant from limited data. Researchers have developed methods to estimate the elimination rate constant from as few as a single blood sample, which is particularly useful for critically ill patients where repeated blood draws are burdensome.¹² For most purposes, though, two measurements at different times give you a reasonable estimate of λ using the two-point method described earlier.

Why Exponential Decline Feels Counterintuitive

People are reliably bad at intuiting exponential processes. Most of the research on this focuses on exponential growth, where studies have documented a strong bias toward underestimating how fast compounding stacks up. In experiments where people estimated the results of repeated compounding, the average participant perceived only about a third of the actual impact, treating exponential growth as though it were closer to linear.¹³

The same cognitive tendency affects how people think about decay, just in reverse. We tend to overestimate how much remains after several half-lives because we mentally subtract a fixed amount each period rather than a fixed fraction. If something starts at 1,000 and its half-life is one year, your intuition might guess roughly 500 after two years, picturing two rounds of losing 250. The actual answer is 250, because each halving applies to whatever is left, not to the original amount. After three years, you are at 125, not “about 250.” The gap between linear intuition and exponential reality widens quickly.

This matters practically whenever you are making decisions based on decay estimates. If a contaminated site has a pollutant with a 5-year half-life and you need it below 5% of its current concentration, your gut might guess 20 years. The real answer, solved with the formula, is about 21.6 years. Not far off in this case, but the mismatch grows for shorter half-lives, multiple compounding periods, or situations where precision matters.

Soil Contamination and Environmental Cleanup Timelines

One of the most practical applications of the decay formula is estimating how long environmental contamination will persist. When pesticides or pharmaceutical residues enter soil, their breakdown often follows first-order kinetics well enough that regulators use the half-life as a key planning metric.⁴ A contamination event with a chemical whose half-life is 20 days will drop to about 3% of its initial concentration in 100 days: five half-lives. A more persistent compound with a half-life of 40 days would need 200 days to reach the same level.

But the decay constant itself is not a fixed property of the chemical alone. It depends on the environment. Soil organic matter content, clay levels, temperature, moisture, and microbial communities all influence how quickly a compound breaks down. Sterilizing soil significantly slowed degradation in lab experiments, confirming that microbes do much of the work.³ This means the same chemical can have very different effective half-lives in sandy versus clay-rich soil, or in a warm humid climate versus a cold dry one. When you apply the exponential decay formula in environmental settings, the decay constant you use should come from data collected under conditions similar to the site in question, not from a single published value in a textbook.

For persistent organic pollutants like dioxins and PCBs, the picture gets more complicated because degradation occurs through multiple pathways simultaneously. The overall rate constant is the sum of contributions from photolysis, biodegradation, and other processes, and each pathway may dominate in different environmental compartments (air, water, soil, sediment).¹ A compound might degrade quickly in sunlit surface water but persist for decades in deep sediment where photolysis cannot reach. The formula itself does not change, but the λ you plug in depends on where the contamination sits.

Multiple Decay Processes Happening at Once

In many real systems, several independent decay mechanisms operate simultaneously on the same substance. A pesticide in a lake might undergo photolysis from sunlight, hydrolysis from water chemistry, and biodegradation from microbes all at the same time. Each process has its own rate constant, and because the processes are independent, their rate constants add together: the total decay constant is the sum of the individual ones. The half-life of the overall process is then ln(2) divided by this combined constant.

This additive property is useful. It means you can speed up an overall decay process by enhancing any one contributing mechanism. Bioremediation efforts, for example, work by boosting microbial activity, effectively increasing the biodegradation rate constant and shortening the total half-life. Conversely, conditions that shut down one pathway, like burying a light-sensitive compound underground where photolysis stops, lengthen the half-life because one term drops from the sum.

Understanding how the formula handles combined rates also clarifies why the same material decays at different speeds in different conditions. Forest deadwood decomposes faster in warmer, wetter climates where fungal activity is high, and slower in polluted environments where the microbial community is suppressed.⁷ The exponential formula stays the same; only the numbers you feed into it change.

1
PubMed. Degradation half-life times of PCDDs, PCDFs and PCBs for environmental fate modeling
2
PubMed Central. Physiology, Zero- and First-Order Kinetics
3
PubMed. Degradation and adsorption of selected pharmaceuticals and personal care products (PPCPs) in agricultural soils
4
PubMed. Adsorption and degradation of triazophos, chlorpyrifos and their main hydrolytic metabolites in paddy soil from Chaohu Lake, China
5
Forest Ecology and Management. Five-year decomposition of coarse woody debris in a subtropical forest: Effects of tree species and nitrogen addition
6
PLOS ONE. Decay and nutrient dynamics of coarse woody debris in the Qinling Mountains, China
7
Forest Ecology and Management. Pollution-induced slowdown of coarse woody debris decomposition differs between two coniferous tree species
8
PubMed. Biphasic clearance kinetics of hepatitis B virus from patients during adefovir dipivoxil therapy
9
PubMed. Viral kinetics and mathematical models
10
Physics Education. Discharging a capacitor
11
PubMed. Observation of the quantum zeno and anti-zeno effects in an unstable system
12
PubMed. Estimating a drug’s elimination rate-constant or half-life from a single blood sample: a practical approach with particular benefits for critically ill/vulnerable patients
13
Journal of Economic Psychology. Exponential-growth bias and overconfidence