The unit s⁻¹ (read “per second” or “inverse seconds”) means that something is being counted, repeated, or measured for every one second of time. It is one of the most versatile units in science, showing up whenever a quantity is divided by time. You will see it attached to wildly different measurements depending on the field, from how fast a chemical reaction proceeds to how many times a wave oscillates, and recognizing what it means in each context is simpler than it first appears.
What the Superscript Notation Actually Means
The small “⁻¹” after the “s” is a mathematical shorthand. Writing s⁻¹ is identical to writing 1/s or “/s.” The negative exponent flips the unit to the denominator, so whenever you see s⁻¹, you can mentally replace it with “divided by one second” or just “per second.” A speed of 5 meters per second can be written 5 m/s or 5 m·s⁻¹. Both say exactly the same thing. Scientists prefer the exponent form because it stays tidy when you start stacking several units together. Writing kg·m²·s⁻³ is easier to read than kg×m²/s³, especially in equations where fractions inside fractions get confusing fast.
This convention applies to every SI unit, not just seconds. You will see m⁻¹ (per meter) in optics, kg⁻¹ (per kilogram) in pharmacology, and mol⁻¹ (per mole) in chemistry. The logic is always the same: a negative exponent means “for every one of that unit.” Once that clicks, the notation stops being intimidating regardless of which letters surround it.
The Most Common Meaning: Frequency
The single most frequent reason you will encounter s⁻¹ is frequency, the number of times something repeats in a second. When a tuning fork vibrates 440 times each second, its frequency is 440 s⁻¹. This particular use of s⁻¹ has its own dedicated name: the hertz, abbreviated Hz. One hertz equals exactly one s⁻¹. Radio stations, musical pitch, computer processors, and alternating-current electricity all express their repetition rates in hertz, which is simply a dressed-up version of s⁻¹.
The hertz was adopted as an SI unit in 1960 to honor the physicist Heinrich Hertz, and it exists purely for convenience and clarity. Writing “the FM station broadcasts at 98.5 MHz” is friendlier than “98.5 × 10⁶ s⁻¹.” But in research papers and technical reports, especially in physics and engineering, you will often see the bare s⁻¹ form instead of Hz, usually because the author wants to make the dimensional relationships in an equation transparent rather than hide them behind a named unit.
Rates That Are Not Frequency
Here is where things get interesting: s⁻¹ does not always describe something repeating in a cycle. It also describes how quickly something happens per unit of time, even if the event is not periodic. In these contexts, calling it “hertz” would be misleading, because hertz implies a regular oscillation. The unit is still s⁻¹, but the physical meaning is different.
Radioactive decay is a classic example. Every radioactive isotope has a decay constant, often symbolized by the Greek letter lambda (λ), measured in s⁻¹. This number tells you the probability per second that any given atom will decay. Carbon-14, for instance, has a very small decay constant, meaning each atom is overwhelmingly likely to survive any particular second. Radon-220, by contrast, has a much larger decay constant, reflecting its rapid disintegration. The unit is s⁻¹ in both cases, but nobody calls it hertz because decay is a random, one-time event for each atom rather than a repeating cycle.
Chemical reaction rate constants work similarly. A first-order reaction, where the speed depends only on how much reactant is present, has its rate constant expressed in s⁻¹. The number tells you what fraction of the starting material converts to product each second. Enzyme turnover numbers, which describe how many substrate molecules a single enzyme molecule can process per second, also carry units of s⁻¹. In all these cases, the unit says “per second” and the context tells you what is happening per second.
Angular Velocity and the Radian Question
When something spins, its angular velocity is typically expressed in radians per second, written rad/s or rad·s⁻¹. A radian is the SI unit for measuring angles, but it is defined as a ratio of two lengths (arc length divided by radius), which makes it dimensionless, meaning it has no physical unit in the strict sense. Because of that, radians per second reduces dimensionally to just s⁻¹. This creates a genuine source of confusion: if you see a value of 314 s⁻¹ in a paper, it could mean 314 hertz (cycles per second) or 314 radians per second (angular speed), which are not the same thing. A full cycle is 2π radians, so 314 rad/s corresponds to about 50 Hz, not 314 Hz.
In practice, context almost always resolves the ambiguity. A paper discussing a spinning wheel or a rotating magnetic field will mean radians per second. A paper discussing electromagnetic wave frequency will mean hertz. But metrologists, the scientists who maintain measurement standards, have debated for decades whether the radian should formally carry a unit label precisely to prevent this kind of confusion. For now, the official SI position is that the radian is dimensionless, so angular velocity and frequency share the same bare unit of s⁻¹ even though they describe different physical things.
Real Examples from Research
Seeing s⁻¹ in the wild helps solidify the concept. In atmospheric physics, the rate at which an excited molecule spontaneously emits a photon and drops to a lower energy state is called the Einstein A-coefficient, and it is measured in s⁻¹. A higher value means the molecule emits faster. For the 1.27 µm infrared band of molecular oxygen, for instance, the A-coefficient has been determined to be about 1.47 × 10⁻⁴ s⁻¹, meaning each excited molecule has roughly a 0.015 percent chance of emitting a photon in any given second.1Geophysical Research Letters. The Einstein Coefficient for spontaneous emission of the O2(a1Δg) state That may sound negligible on a per-second basis, but across billions of molecules and longer timescales, the emission produces a measurable atmospheric glow called airglow, visible from space.
In neuroscience, the firing rate of a neuron, how many electrical impulses it sends per second, is another quantity naturally expressed in s⁻¹ (or equivalently spikes per second). Researchers carefully calibrate recording techniques to make sure the measured rate reflects the neuron’s true behavior rather than an artifact of the equipment.2Journal of Neuroscience. Measuring the Firing Rate of High-Resistance Neurons with Cell-Attached Recording A resting cortical neuron might fire a handful of times per second, while a neuron in the auditory pathway can fire hundreds of times per second. In every case, the unit is s⁻¹.
In nuclear magnetic resonance (NMR) spectroscopy, a technique used to study molecular structure, various relaxation and damping rates are expressed in s⁻¹. The radiation damping rate in an NMR experiment, for example, describes how quickly the bulk magnetization of a sample returns to equilibrium due to its own emitted signal, and it is given as an inverse time in s⁻¹. These rates matter because they influence the sharpness and accuracy of the spectral lines researchers use to identify chemical compounds.
How the Second Itself Is Defined
Because s⁻¹ is built on the second, it is worth knowing what a second actually is by modern standards. Since 1967, the SI second has been defined by the behavior of cesium-133 atoms. Specifically, the second is the time it takes for the radiation associated with a particular energy transition in cesium-133 to oscillate exactly 9,192,631,770 times.3Annalen der Physik. The Hyperfine Transition for the Definition of the Second This atomic definition replaced earlier definitions based on the Earth’s rotation, which turned out to be slightly irregular. The cesium-based second is stable to better than one part in ten trillion, making s⁻¹ an extraordinarily precise unit to build on.
Metrologists are now working toward redefining the second using optical clocks, which tick at frequencies hundreds of thousands of times higher than cesium clocks and promise even greater precision. When that redefinition happens, the numerical size of the second will not change perceptibly for everyday purposes, but the definition anchoring s⁻¹ will shift to a different atomic transition. For anyone using s⁻¹ outside a national metrology lab, the practical impact will be zero.
When s⁻¹ Appears in Compound Units
You will often see s⁻¹ as one piece of a larger unit rather than standing alone. A few common compound units illustrate how versatile it is:
- m·s⁻¹: meters per second, the SI unit of speed. Divide a distance by a time and you get this.
- m·s⁻²: meters per second squared, the SI unit of acceleration. Earth’s gravitational acceleration is about 9.8 m·s⁻².
- kg·m²·s⁻³: the watt, the SI unit of power. Even familiar units like watts are built from kilograms, meters, and inverse seconds when you unpack them into their base components.
- mol·L⁻¹·s⁻¹: a common unit for reaction rates in chemistry, describing how many moles of a substance form (or disappear) per liter per second.
Each of these compound units inherits its time dimension from s⁻¹. When physicists or engineers perform dimensional analysis, checking that the units on both sides of an equation match, they break every unit down to base components. The second (and its inverse) is one of the seven SI base units that appear over and over in these breakdowns, which is why s⁻¹ is practically impossible to avoid in technical work.
Common Points of Confusion
A few misunderstandings come up regularly when people first encounter s⁻¹. The biggest is assuming it always means hertz. As covered above, hertz is one specific use of s⁻¹ reserved for cyclical frequency. Decay constants, rate constants, damping rates, and probability-per-unit-time quantities all carry the same unit without being frequencies in the everyday sense.
Another source of confusion is mixing up s⁻¹ with “per minute” or “per hour.” Revolutions per minute (rpm), for instance, is a common unit for engine speeds and centrifuge settings, but it is not an SI unit. To convert rpm to the SI angular velocity in rad·s⁻¹, you need to account for both the minute-to-second conversion and the fact that one revolution equals 2π radians. An engine spinning at 3,000 rpm is turning at about 314 rad·s⁻¹. Forgetting either conversion factor is an easy mistake.
A subtler confusion arises with quantities like bandwidth. In electronics, bandwidth can be described in hertz, but signal processing also uses the “Neper per second” (Np·s⁻¹) for describing exponential decay rates of signals. Both have s⁻¹ lurking inside, but they measure different aspects of signal behavior: one measures oscillation frequency, the other measures how fast an oscillation dies away. Recognizing which flavor of s⁻¹ you are looking at always comes down to reading the surrounding context.
Reading s⁻¹ on Labels, Displays, and Data Sheets
Outside research papers, you might run into s⁻¹ on equipment data sheets, calibration certificates, and sensor readouts. A vibration sensor on industrial machinery, for example, might report acceleration in m·s⁻² and frequency in Hz (which is s⁻¹). A Geiger counter might report counts per second, abbreviated cps, which is dimensionally identical to s⁻¹ but uses a friendlier label for field work. Laser specification sheets often list linewidth or pulse repetition rate in Hz, but the underlying physics discussions in the accompanying documentation might switch to s⁻¹ without warning.
If you are reading a technical document and suddenly encounter s⁻¹ where you expected hertz, or vice versa, the most useful thing to check is whether the quantity in question is a cyclical frequency (oscillations, waves, pulses) or a non-cyclical rate (decay, relaxation, probability per second). For cyclical quantities, s⁻¹ and Hz are interchangeable. For non-cyclical rates, only s⁻¹ is appropriate. That single question resolves the ambiguity almost every time.
Why Scientists Do Not Just Say “Per Second”
Given that s⁻¹ literally means “per second,” you might wonder why anyone bothers with the notation. Part of the answer is compactness, but the deeper reason is dimensional consistency. When you write an equation, every term must have units that balance. Writing s⁻¹ makes the time dependence explicit and machine-checkable. If someone accidentally writes s⁻² where s⁻¹ belongs, the mismatch jumps out immediately. Written longhand as “per second” buried in a sentence, the same error might slip past unnoticed.
The notation also travels across languages without translation. A researcher in Japan, Brazil, or Germany encountering s⁻¹ knows exactly what it means without needing to parse an English phrase. This universality is one of the quiet strengths of SI notation in general: a handful of letters and superscripts replaces thousands of words across dozens of languages. For a unit as ubiquitous as “per second,” that efficiency adds up fast.
Everyday Quantities You Already Understand in s⁻¹
Your resting heart rate is roughly 1 to 1.3 s⁻¹, which you would more naturally call 60 to 80 beats per minute. The refresh rate of your phone screen is 60 or 120 s⁻¹ (60 Hz or 120 Hz). The hum of the electrical grid in your walls is either 50 or 60 s⁻¹ depending on where you live. The frequency of the musical note middle C is about 262 s⁻¹. None of these examples require you to understand physics to grasp what s⁻¹ is doing: it is counting events in each second. The notation is more compact than the English phrase, and it plugs cleanly into equations, but the meaning is the one you already intuit. Something happens, and s⁻¹ tells you how often.