Scientific Notation Explained: From Standard Form to a × 10ⁿ

The first time most people meet scientific notation, it shows up as a wall of zeros they would rather not count. The speed of light is 299,792,458 m/s. A hydrogen atom is about 0.0000000001 m across. Writing those out is error-prone, and reading them back is worse — miscount one zero and your answer is off by a factor of ten. Scientific notation fixes that by splitting every number into two honest pieces: a coefficient you can read at a glance, and a power of ten that tells you where the decimal point really sits.

This guide walks through the conversion both ways, the one rule that makes a coefficient "correct," and the gotchas that bite people when numbers get astronomically large or subatomically small.

The mantissa × 10^exponent format

Every number in scientific notation looks like this:

a × 10ⁿ

The a is the coefficient (sometimes called the mantissa). The n is the exponent. The defining rule is short and worth memorizing: a number is written as a coefficient between 1 and 10 — formally 1 ≤ |a| < 10 — times a power of ten. One non-zero digit sits to the left of the decimal point, and everything else trails behind it.

That single constraint is what separates real scientific notation from its lookalikes. 4.7 × 10⁴ is scientific notation. 47 × 10³ is not, even though it equals the same value, because 47 is not between 1 and 10. (That second form is engineering notation, which I'll come back to.) When you keep the coefficient pinned to that range, every number has exactly one scientific form, and comparing two numbers becomes a matter of glancing at their exponents.

Converting standard form to scientific notation

The mechanic is always the same: move the decimal point until exactly one non-zero digit sits in front of it, then count how many places you moved.

Move the decimal left (the number was large) → the exponent is positive.
Move the decimal right (the number was small) → the exponent is negative.
The number of places you moved is the exponent.

Two worked examples:

92500 → 9.25 × 10⁴. Start at the implied decimal after the last zero: 92500.. Slide it left until it sits after the 9: 9.2500. That took four hops, so the exponent is 4. Drop the now-meaningless trailing zeros and you have 9.25 × 10⁴.

0.00045 → 4.5 × 10⁻⁴. Here the number is smaller than one, so the decimal moves right. 0.00045 → slide past the leading zeros until the 4 sits in front: 4.5. That took four hops to the right, so the exponent is −4, giving 4.5 × 10⁻⁴.

Going the other direction is just the reverse: a positive exponent tells you how many places to push the decimal right (padding with zeros), and a negative exponent pushes it left. 4.5 × 10⁻⁴ expands back to 0.00045; 9.25 × 10⁴ expands back to 92500. If you want to check your hop-counting against a reliable answer, paste either form into the Scientific Notation Converter and it shows the plain decimal, the scientific form, the engineering form, and E-notation side by side.

Significant figures travel with the coefficient

One reason scientists insist on scientific notation is that it makes precision explicit. The plain number 4700 is ambiguous — those trailing zeros might be measured digits or might just be placeholders. Is it good to two figures or four? You cannot tell.

Write it as 4.70 × 10³ and the ambiguity vanishes: the coefficient 4.70 says three significant figures, full stop. Counting them is mechanical. Start at the first non-zero digit and count every digit after it, including trailing zeros that follow a decimal point. In 0.00420 the significant figures are 4, 2, 0 — three of them. The leading zeros before the 4 are not significant; they only place the decimal. A common lab-report deduction is claiming more precision than your instrument delivers, and that mistake almost always comes from miscounting those leading zeros.

Engineering notation and E-notation

Scientific notation has two close cousins worth knowing.

Engineering notation uses the same a × 10ⁿ shape but forces the exponent to be a multiple of three. That lines the numbers up with SI prefixes: 10³ is kilo, 10⁶ is mega, 10⁻⁶ is micro. So 47000 becomes 47 × 10³ (i.e. 47 k-somethings), and the coefficient is allowed anywhere in 1 ≤ |a| < 1000. Engineers reach for it because the exponent maps straight onto the k / M / G / m / µ / n labels printed on a multimeter or a datasheet — no mental arithmetic at the bench.

E-notation is the compact form your code and spreadsheets speak: 4.5e-4 means 4.5 × 10⁻⁴, and 9.25E4 means 9.25 × 10⁴. The e (case-insensitive) literally stands for "times ten to the power of." It exists because keyboards have no superscript key, so 6.022e23 is how Avogadro's number lands in a Python file, a JSON payload, or a CSV cell. When you copy a constant out of a paper written as 6.62607015 × 10⁻³⁴ and need it as a code literal, you want 6.62607015e-34 — same value, no Unicode superscript to fight with.

Why huge and tiny numbers break naive tools

Here is the part I learned the hard way. Early on I wrote a quick conversion helper that just called parseFloat on the input string and reformatted it. It worked beautifully for 92500 and 0.00045, so I shipped it. Then someone fed it 1.23456789012345678e-300, and it handed back a number that had quietly lost its last several digits. The reason is that JavaScript — like almost every language — stores numbers as 64-bit IEEE-754 binary floats, and most decimals have no exact binary representation. That is the same quirk behind the infamous 0.1 + 0.2 = 0.30000000000000004. For a converter, silent rounding is a real bug, not a curiosity, and at the extremes a float can overflow a giant value to Infinity or flush a tiny one to 0.

The fix is to never let the value become a float at all. Parse the digit string, track the exponent as an integer, and shift the decimal point with plain integer arithmetic. That way Avogadro's number 6.022 × 10²³ and the electron mass 9.109 × 10⁻³¹ kg convert digit-for-digit, and a 600-digit exponent stays exact instead of collapsing. That string-and-integer approach is exactly what the converter linked above does under the hood, which is why I trust it for the values a calculator quietly mangles.

A quick reference

A number in scientific notation is a coefficient 1 ≤ |a| < 10 times a power of ten.
To convert from standard form, move the decimal until one non-zero digit leads, then count the moves: left is positive, right is negative.
92500 = 9.25 × 10⁴; 0.00045 = 4.5 × 10⁻⁴.
Significant figures are read off the coefficient — count from the first non-zero digit, leading zeros never count.
Engineering notation locks the exponent to multiples of three; E-notation (4.5e-4) is the code-friendly spelling.
For very large or very small numbers, use a tool that works from the digit string, not a 64-bit float.

If you find yourself converting numbers between units as well — say, that micro-ohm impedance into a different scale — the Unit Converter handles the dimensional side while the notation converter handles the exponent side. Between them, you should never have to count a row of zeros by hand again.

Made by Toolora · Updated 2026-06-13