Why this conversion has higher stakes than it looks

Plug 25 °C into PV = nRT and you will get an answer that is wrong by 92%. Plug −10 °C into the Arrhenius equation and you will get a negative rate constant, which is meaningless. Temperature conversion looks like the most boring chapter in the textbook until you skip it once in a real calculation, and then it becomes the chapter that ate your problem set. Kelvin is non-negotiable for any equation that treats temperature as a multiplicative factor — gas laws, kinetics, thermodynamics, anything with a Boltzmann factor.

The Celsius–Fahrenheit conversion is the one most people memorized in middle school, but the offset trips up adults too. The fix is to understand the structure: Fahrenheit and Celsius differ in both interval size (9:5 ratio) and zero point (offset of 32). Kelvin and Celsius share the interval size but differ in zero point. Once you know which transformation you are doing, the formula falls out.

The four scales and what they actually measure

Celsius (°C) anchors at water’s phase changes at 1 atm: 0 freezing, 100 boiling. Default in lab work outside the US.

Fahrenheit (°F) has water freezing at 32 and boiling at 212 — the same span occupies 180 degrees instead of 100. The smaller degree size is why a 1° change in Fahrenheit feels less significant than a 1° change in Celsius.

Kelvin (K) is the SI scale. 0 K is absolute zero. Same degree size as Celsius — the only difference is the offset. No negative values exist by definition. Always required for ratio-based thermodynamic calculations.

Rankine (°R) is to Fahrenheit what Kelvin is to Celsius — an absolute scale with the smaller degree. 0 °R = 0 K. Used in some US engineering contexts; rare in chemistry.

The formulas in one place

Celsius ↔ Fahrenheit:

F = (C × 9/5) + 32
C = (F − 32) × 5/9

Celsius ↔ Kelvin:

K = C + 273.15
C = K − 273.15

Fahrenheit ↔ Kelvin (chain through Celsius):

K = (F − 32) × 5/9 + 273.15
F = (K − 273.15) × 9/5 + 32

Fahrenheit ↔ Rankine:

R = F + 459.67
F = R − 459.67

Kelvin ↔ Rankine:

R = K × 9/5
K = R × 5/9

Worked examples

Body temperature, C to F. F = (37.0 × 1.8) + 32 = 66.6 + 32 = 98.6 °F.

Oven, F to C. C = (350 − 32) × 5/9 = 318 × 0.5556 = 176.7 °C. Subtract 32 first, then scale — getting the order wrong gives 226 °C.

Liquid nitrogen, C to K. K = −196 + 273.15 = 77.15 K. Most lab work rounds to 77 K.

Sun’s surface, K everywhere else. 5778 K → C = 5778 − 273.15 = 5505 °C → F = (5505 × 9/5) + 32 = 9941 °F. At extreme temperatures the +32 offset is rounding error and people sometimes drop it. Don’t — keep the formula intact and let the magnitudes sort themselves out.

The −40 fixed point. Set C = F: C = (C × 9/5) + 32 → −0.8C = 32 → C = −40. −40 °C = −40 °F, the only temperature where the two scales agree. Useful as a sanity-check anchor.

Why Kelvin is mandatory in chemistry

Every equation where temperature appears as a multiplicative factor needs absolute temperature:

Ideal gas law: PV = nRT. T must be Kelvin. Plugging in Celsius blows up at 0 °C (you would get a finite product equal to zero).
Combined gas law: P₁V₁/T₁ = P₂V₂/T₂. Kelvin only — ratios make no sense with negative or arbitrary-zero temperatures.
Arrhenius: k = A·exp(−Ea/RT). The exponential demands an absolute T, otherwise you get nonsense at low temperatures.
Gibbs free energy: DeltaG = DeltaH − T·DeltaS. T in Kelvin, DeltaH and DeltaS in matching units.
Boltzmann distribution and partition functions: any −E/kT factor.

The fix is mechanical: every time you read a temperature off a thermometer in °C, your first move before plugging into any of these equations is +273.15.

Reference table

Reference	°C	°F	K
Absolute zero	−273.15	−459.67	0
Dry ice sublimation	−78.5	−109.3	194.7
Water freezes	0	32	273.15
Body temperature	37.0	98.6	310.15
Water boils (1 atm)	100	212	373.15
Lead melts	327.5	621.5	600.6
Iron melts	1538	2800	1811

Traps to watch

Celsius in gas-law equations. The most expensive mistake in this section. Always convert first.
273 vs 273.15. Textbook work usually allows the truncated value; precise calorimetry or thermodynamics requires the full 273.15.
Dropping the +32. The Celsius–Fahrenheit conversion has both a scale and an offset. Skip the offset and your answer is wrong by ~32 degrees at room temperature.
Temperature vs temperature change. A change of 10 °C equals a change of 10 K but a change of 18 °F. The offset only applies to absolute conversions, not differences.
Operation order in F→C. Subtract 32 first, then multiply by 5/9. Reverse the order and you get garbage.

The Thermochemistry Calculator handles all four scales and pairs nicely with heat-transfer and calorimetry workflows where you’ll be converting back and forth.

How to Convert Between Temperature Scales