Intuition The one core idea
A gas is billions of molecules flying at random speeds, and the Maxwell-Boltzmann distribution is just a hill-shaped graph telling you how common each speed is . Everything in the derivation is built from two facts: fast molecules are energetically expensive (they get rare), but there are more ways to be fast (a bigger "shell" in velocity space) — and the whole page below teaches you the exact symbols that encode those two facts.
This page assumes nothing . Before you touch the full derivation , every letter, squiggle, and picture it uses is built here from the ground up, in an order where each idea leans on the one before it.
Definition Velocity vector
v and its components
A molecule doesn't just have a speed; it moves in a direction . We write its motion as an arrow v (the little arrow on top means "this is a directional quantity, not just a number").
To pin the arrow down with numbers, we drop it into a 3D grid of three perpendicular axes labelled x , y , z . The components v x , v y , v z are how fast the molecule moves along each axis.
Picture a molecule shooting across a room. Its shadow on the floor moving east-west is v x , its shadow moving north-south is v y , and how fast it climbs is v z .
v (a plain number)
Speed is the length of the arrow — how fast, ignoring direction. If you know the three components, Pythagoras in 3D gives:
v = v x 2 + v y 2 + v z 2 .
Why the square root of a sum of squares? Because that is just the 3D version of the ruler-distance formula: the length of the diagonal of a box with sides v x , v y , v z . Look at figure s01 — the red arrow's length is exactly this diagonal.
Intuition Why the topic needs BOTH
The derivation starts with components (they are independent — no force ties east-motion to up-motion) and only later assembles them into speed (the thing we actually measure and plot). Keeping the two separate is the whole trick.
Before we can build the energy-cost factor, we need the three physical quantities it is made from.
Definition The three physical inputs
T = absolute temperature in kelvin (K). T = 0 means no thermal motion at all; bigger T = more violent jiggling.
k B = Boltzmann's constant = 1.38 × 1 0 − 23 J/K . It is just a unit converter : it turns temperature (kelvin) into energy (joules), because k B T has units of energy.
m = mass of one single molecule in kilograms. Not molar mass! (Molar mass M is per mole ; divide by Avogadro's number to get m .)
m vs M
Why it trips people: tables list molar mass M in g/mol, so it feels ready to use.
The fix: either use per-molecule m with k B , or molar mass M (in kg/mol) with the gas constant R . Never mix m with R or M with k B .
The combination k B T appears everywhere: it is the natural energy scale of the gas — roughly "how much kinetic energy one degree of freedom carries." That's the message of the Equipartition theorem .
e and the function e − x
e ≈ 2.718 is a special constant. In e − x , the symbol x is just a placeholder input — any real number you feed in (x ranges over all reals, though we will only ever plug in non-negative quantities like an energy ratio). The function e − x is a curve that:
equals 1 when x = 0 ,
shrinks smoothly toward 0 as x grows,
never actually reaches 0 and never goes negative.
Think of it as "how much is left" — like a hot cup cooling: it drops fast, then slower, forever approaching room temperature.
Why this exact curve and not, say, a straight line down? Because e − x has a magical property: it turns a sum in the exponent into a product of separate factors . That is, e − ( a + b ) = e − a ⋅ e − b : add inside, and the result splits into a multiplication of two independent pieces. (Equivalently, its inverse — the logarithm — turns products back into sums.) Hold onto this: it is exactly why the derivation can split three axes bundled together in one exponent into three independent factors, one per axis.
Definition The Boltzmann factor
e − E / k B T
Now we plug something real into the placeholder x : the ratio E / k B T , using k B and T from §2. Whenever a state costs energy E , nature makes it rarer by exactly the factor e − E / k B T . Here the "energy" of a moving molecule is its kinetic energy 2 1 m v 2 (with m the molecular mass from §2), so the factor becomes e − m v 2 /2 k B T .
This is the energy-cost half of the tug-of-war.
You'll meet this factor properly in Boltzmann factor and partition function — for now just read it as "big energy ⇒ small probability."
Intuition Why the sum in the exponent splits into three component factors
Kinetic energy uses the sum v x 2 + v y 2 + v z 2 , so the Boltzmann factor is e − m ( v x 2 + v y 2 + v z 2 ) /2 k B T . Using the property above — a sum in the exponent becomes a product of factors — this splits cleanly:
e − 2 k B T m ( v x 2 + v y 2 + v z 2 ) = e − 2 k B T m v x 2 ⋅ e − 2 k B T m v y 2 ⋅ e − 2 k B T m v z 2 .
Each factor involves only one axis . So the joint density falls apart into three independent single-axis pieces — that is why the derivation can treat v x , v y , v z separately, and why each one gets its own bell-shaped curve.
Intuition You can't ask "what's the probability of EXACTLY
v = 422 m/s?"
Speed is continuous, so any single exact value has probability zero (there are infinitely many nearby speeds). Instead we ask about a tiny band of speeds, from v to v + d v .
f ( v ) and the interval d v
d v = an infinitesimally thin sliver of speed, "a tiny width."
f ( v ) = the probability density : probability per unit speed . It is not itself a probability.
f ( v ) d v = the actual fraction of molecules whose speed lands in the sliver [ v , v + d v ] .
Picture a histogram of speeds. As you make the bars infinitely thin, the tops trace a smooth curve f ( v ) . The area of each thin bar (height f ( v ) × width d v ) is the fraction of molecules in that bar.
Definition The integral sign
∫ and normalization
∫ 0 ∞ f ( v ) d v means "add up the areas of all the thin bars from speed 0 to infinity." Since every molecule has some speed, these fractions must total 1 (i.e. 100% ):
∫ 0 ∞ f ( v ) d v = 1.
This "must add to 1" rule is called normalization — it fixes the leftover constants in the formula.
Read ∫ as a stretched-out "S" for S um. It is the tool for adding infinitely many infinitely thin pieces — exactly what a continuous distribution demands.
Definition The single-component density
g ( v x )
g is the probability density of one velocity component : g ( v x ) d v x = the fraction of molecules whose x -component of velocity lands in the sliver [ v x , v x + d v x ] . It is the one-axis cousin of f ( v ) from §4, and it must normalize: ∫ − ∞ ∞ g ( v x ) d v x = 1 . (There is an identical g for the y and z axes; §3 showed why the three separate.)
Definition A Gaussian (bell curve)
A e − b v x 2
From §3, each component's factor is e − m v x 2 /2 k B T — a curve symmetric about 0 , peaking in the middle and decaying on both sides (positive and negative v x are equally likely, since no direction is preferred). We name its two pieces:
b = the width parameter in the exponent. Comparing with §3, b = 2 k B T m : bigger mass or colder gas ⇒ larger b ⇒ a narrower bell.
A = the height constant out front, chosen so the area equals 1 .
So g ( v x ) = A e − b v x 2 with those meanings fixed.
Definition The spherical-shell factor
4 π v 2
This is the geometry half of the tug-of-war. Imagine all velocity arrows of the same length v : their tips form a hollow sphere of radius v in velocity space. A sphere's surface area is 4 π v 2 , so there is more room to be fast. This is why f ( 0 ) = 0 : a sphere of radius 0 has no surface at all.
Intuition Putting the two heroes together
Multiply three identical Gaussians (one per axis) by the shell factor. The three height constants combine into A 3 = ( 2 π k B T m ) 3/2 , and that is exactly the prefactor of the final speed distribution:
f ( v ) = more room (grows) 4 π v 2 ( 2 π k B T m ) 3/2 energy cost (shrinks) e − m v 2 /2 k B T .
One factor rises, one falls — their product is a hill. That hill is Maxwell-Boltzmann, and the messy-looking ( 2 π k B T m ) 3/2 out front is nothing more than three copies of the single-axis normalization constant A from above.
velocity components vx vy vz
Gaussian per component g of vx
exponential e to the minus x
sum in exponent splits into product
temperature T and kB and mass m
Gaussian integral fixes A
spherical shell 4 pi v squared
probability density f of v and integral
Notice the shape: independent components each get a Gaussian , the Gaussian integral and normalization nail down the constants, the shell factor converts components into speed , and the Boltzmann factor supplies the energy cost. Multiply — done.
Test yourself: cover the right side and answer before revealing.
What does the little arrow in v signify? That v is directional (a vector), not just a plain number.
How do you get speed v from components? v = v x 2 + v y 2 + v z 2 — the 3D Pythagoras / diagonal length.
What is the value and role of k B ? 1.38 × 1 0 − 23 J/K; it converts temperature into energy so k B T has energy units.
Why can't we ask for the probability of an exact speed? Speed is continuous, so any single value has probability zero; we use bands f ( v ) d v instead.
What is the difference between f ( v ) and f ( v ) d v ? f ( v ) is probability per unit speed (a density); f ( v ) d v is the actual fraction in the sliver [ v , v + d v ] .
What does g ( v x ) represent? The probability density of one velocity component; g ( v x ) d v x is the fraction of molecules with x -component in [ v x , v x + d v x ] .
What does ∫ 0 ∞ f ( v ) d v = 1 mean physically? Every molecule has some speed, so all fractions add to 100%.
What key property of the exponential does the derivation use? e − ( a + b ) = e − a ⋅ e − b — a sum in the exponent splits into a product of separate factors (one per axis).
What are A and b in g ( v x ) = A e − b v x 2 ? b = m /2 k B T (width in the exponent);
A = b / π = m /2 π k B T (height, fixed by normalization).
How is A pinned down? By the Gaussian integral
∫ e − b v x 2 d v x = π / b plus area = 1, giving
A π / b = 1 .
What is the full prefactor of f ( v ) and where does it come from? ( 2 π k B T m ) 3/2 — it is A 3 , three copies of the single-axis normalization constant.
Where does 4 π v 2 come from? The surface area of the sphere of velocity vectors all having length v .
Why is f ( 0 ) = 0 even though e 0 = 1 ? The shell factor 4 π v 2 vanishes at v = 0 — a zero-radius sphere has no surface.
What is the Gaussian integral result? m or molar mass M — which pairs with k B ?Per-molecule mass m pairs with k B ; molar mass M pairs with R . Never mix.
Ready? Then head to the full derivation . Related tools live in Kinetic theory of gases , Effusion and Graham's law , and the comparison Maxwell-Boltzmann vs Fermi-Dirac vs Bose-Einstein .