Propellant properties — density, freezing point, toxicity, storability: Worked examples

This page is the drill floor for the parent topic. We built the ideas there; here we exercise them on every kind of case a problem can throw at you.

Before we start, two tiny reminders so no symbol arrives unexplained:

Recall What ρ, V and m mean

$ρ$ (rho) is density — how much mass is packed into each little box of volume. $m$ is mass (in kilograms, kg). $V$ is volume (in cubic metres, m³, or cubic centimetres, cm³). They are tied together by $ρ = m / V$ , which rearranges to $V = m / ρ$ . Picture a bucket: same kilograms of feathers vs lead — the lead needs a far smaller bucket because its $ρ$ is bigger.

Recall The symbols in the boil-off formula

$\dot{Q}$ (Q-dot) — rate of heat leaking in, in watts (W = joules per second). The dot means "per second".
$L_{v}$ — latent heat of vaporization: joules needed to boil 1 kg of liquid into gas.
$\overset{m}{˙}$ (m-dot) — mass boiling off per second, in kg/s.
$I_{s p}$ — specific impulse (see Specific Impulse), effectively "engine efficiency in seconds".

The one place a velocity calculation appears (Example 9) introduces the rocket-equation symbols right where they are used, so nothing is loaded on you before you need it.

The scenario matrix

Every problem this topic can pose falls into one of these cells. The worked examples below are labelled with the cell(s) they cover, so together they fill the whole grid. The figure below is that same grid drawn on the board — each cell shows its letter, its topic, and the example number that works it out, so you can trace any cell straight to its example.

Figure — Propellant properties — density, freezing point, toxicity, storability

Figure — the scenario matrix. Nine cells (A–I) span the full space of problems: straight computation, trade-offs, degenerate/zero inputs, sign traps, time evolution, limiting behaviour, ratios, a word problem, and an exam twist. The pale-yellow "Ex n" tag in each cell names the worked example that fills it, and the legend at the bottom repeats the mapping.

#	Cell class	What makes it tricky	Covered by
A	Straight density → volume	just $V = m / ρ$ , watch units	Ex 1
B	Density trade-off (dense vs high-Isp)	two effects fight; need $ρ I_{s p}$	Ex 2
C	Zero / degenerate input	$ρ \to 0$ , or $m = 0$	Ex 3
D	Freezing-point sign & unit (°C ↔ K, is it liquid?)	negative temperatures, K vs °C	Ex 4
E	Boil-off rate over time	$\dot{Q}$ , $L_{v}$ , convert to %/day	Ex 5
F	Limiting behaviour (big tank, thin insulation)	scaling as $V^{2/3}$ ; $d \to 0$	Ex 6
G	Toxicity comparison (ppm, "how many × worse")	small numbers, ratios	Ex 7
H	Real-world word problem (mission choice)	combine density + storability + toxicity	Ex 8
I	Exam twist (rocket-equation Δv, find the flaw)	catch a hidden assumption	Ex 9

Example 1 — Cell A: straight density → volume

Example 2 — Cell B: the density vs Isp trade-off

Worked example Which propellant packs more punch per litre?

Compare RP-1 ( $ρ = 820 kg/m^{3}$ , $I_{s p} = 300 s$ ) with LH₂ ( $ρ = 71 kg/m^{3}$ , $I_{s p} = 450 s$ ) using the density–specific-impulse product $ρ I_{s p}$ .

Forecast: LH₂ has 50% more Isp but only ~9% the density. Which wins on $ρ I_{s p}$ ? Guess before reading.

Compute $ρ I_{s p}$ for each. $ρ I_{s p}_{RP-1} = 820 \times 300 = 246, 000$ $ρ I_{s p}_{LH_{2}} = 71 \times 450 = 31, 950$ Why this step? $ρ I_{s p}$ is our rough figure of merit for how much "push per unit tank volume" a propellant gives. High Isp is useless if it barely fits in the tank; this product balances both.
Take the ratio. $\frac{246 , 000}{31 , 950} = 7.7$ . Why this step? A ratio tells us how many times better one is on this metric, ignoring the raw units.

Verify: RP-1 scores 7.7× higher on $ρ I_{s p}$ — this is exactly why dense fuels win in boosters (packaging-limited) while LH₂ still wins in upper stages where $Δ v$ dominates. Consistent with the parent note's boosters-vs-upper-stages rule. See Propellant Combinations.

Example 3 — Cell C: zero / degenerate input

Worked example What happens as density → 0, or mass → 0?

Two edge cases: (a) a vanishingly thin gas whose density shrinks toward zero, $ρ \to 0$ (its mass per unit volume becomes tiny); (b) an empty tank, $m = 0$ .

Forecast: One of these blows up to infinity, the other collapses to zero. Which is which?

Case (a): $ρ \to 0$ . $V = m / ρ \to \infty$ . Why this step? Dividing a fixed mass by an ever-smaller density means the volume grows without bound — a wisp of gas needs an infinite tank to hold any real mass. This is the degenerate limit that explains why gaseous propellants are stored compressed (high $ρ$ ) or liquefied (cryogenic).
Case (b): $m = 0$ . $V = 0/ ρ = 0$ . Why this step? Zero mass needs zero volume — trivially true, and a good "does my formula behave sanely at the boundary?" check.

Verify: Both limits are physically reasonable: no propellant ⇒ no tank; near-zero density ⇒ impractically huge tank. The formula never divides by mass, so $m = 0$ is safe; it does divide by $ρ$ , so $ρ = 0$ is where it breaks — precisely why we never store propellant as an ideal gas.

Example 4 — Cell D: freezing-point signs & unit conversion

Worked example Is this propellant a liquid at the given temperature?

A satellite thruster sits in shadow at $T = - 30 ° C$ . It uses NTO (freezing point $T_{f} = - 11.2 ° C$ , boiling point $+ 21 ° C$ ). Is the NTO usable? Also express all three temperatures in kelvin.

Forecast: $- 30$ is colder (more negative) than $- 11.2$ . Frozen or fine?

Compare on the number line. For a liquid we need $T_{f} < T < T_{b}$ . Here $- 30 < - 11.2$ , so $T$ is below the freezing point. Why this step? With negative numbers, "colder" means more negative. On the figure below, $- 30 ° C$ (pink dot) sits to the left of the blue "liquid range" band, outside it on the cold side — so the NTO freezes solid. Mission problem!
Convert to kelvin using $T_{K} = T_{° C} + 273.15$ .
- $T_{f} = - 11.2 + 273.15 = 261.95 K$
- $T = - 30 + 273.15 = 243.15 K$
- $T_{b} = 21 + 273.15 = 294.15 K$ Why this step? Kelvin has no negatives, so $243.15 < 261.95$ makes the "too cold" verdict obvious — no sign confusion.

Figure — NTO on the temperature line. The blue band is the usable liquid range between freezing ( $- 11.2 ° C$ ) and boiling ( $+ 21 ° C$ ); the yellow dots mark those edges. The pink dot at $- 30 ° C$ lands left of the band — colder than freezing — so the propellant is solid. The arrow shows the "colder" direction so the sign logic is visible at a glance.

Verify: In kelvin $243.15 K < 261.95 K$ confirms frozen ✓. The fix: heaters must keep the line above $- 11.2 ° C$ . This is why Hypergolic Propellants like NTO/MMH still need thermal management despite being "storable".

Example 5 — Cell E: boil-off rate over time

Worked example How fast does LH₂ boil away?

An LH₂ tank has surface area $A = 30 m^{2}$ , insulation thickness $d = 0.10 m$ , conductivity $k = 0.010 W/(m\cdotK)$ , and temperature difference $Δ T = 270 K$ (warm space vs 20 K hydrogen). Total propellant $m = 8000 kg$ , $L_{v} = 445, 000 J/kg$ . Find the boil-off in % per day.

Forecast: Good insulation gives ~1–3% per day (parent note). Will we land in that band?

Heat leak $\dot{Q} = k A Δ T / d$ . $\dot{Q} = \frac{0.010 \times 30 \times 270}{0.10} = 810 W$ Why this step? This is Fourier conduction: heat flows faster with more area ( $A$ ), a bigger temperature gap ( $Δ T$ ), and thinner insulation ( $d$ ). We need $\dot{Q}$ before we can find how much boils.
Mass boil-off rate $\overset{m}{˙} = \dot{Q} / L_{v}$ . $\overset{m}{˙} = \frac{810}{445 , 000} = 1.82 \times 1 0^{- 3} kg/s$ Why this step? Every joule that leaks in must go somewhere; it goes into evaporating liquid. Dividing power (J/s) by energy-per-kg (J/kg) gives kg/s.
Per day, as a fraction. Seconds in a day $= 86, 400$ . $\overset{m}{˙}_{day} = 1.82 \times 1 0^{- 3} \times 86, 400 = 157.2 kg/day$ $fraction = \frac{157.2}{8000} = 0.0197 = 1.97 % per day$ Why this step? "% per day" is the engineering figure people quote (see Boil-off Losses); we scale the per-second rate up to a day and divide by total mass.

Verify: $1.97 %$ /day sits right inside the quoted 1–3% band for well-insulated LH₂ ✓. Units check: $W \div (J/kg) = (J/s) / (J/kg) = kg/s$ ✓.

Example 6 — Cell F: limiting behaviour (scaling + thin insulation)

Worked example Does a bigger tank boil off faster or slower

per kilogram? And what if $d \to 0$ ? (a) Two geometrically similar tanks: tank 2 holds 8× the volume of tank 1. Compare their boil-off per unit mass. (b) What does $\dot{Q}$ do as insulation thickness $d \to 0$ ?

Forecast: More insulation surface on a big tank... but also way more propellant. Which grows faster?

Mass ∝ volume, heat leak ∝ area. $m \propto V$ , and surface area $A \propto V^{2/3}$ (parent note). Why this step? Boil-off fraction per day $= \overset{m}{˙} / m \propto \dot{Q} / m \propto A / m \propto V^{2/3} / V = V^{- 1/3}$ .
Apply the 8× factor. Ratio of per-mass boil-off $= (V_{2} / V_{1})^{- 1/3} = 8^{- 1/3} = \frac{1}{2}$ . Why this step? $8^{1/3} = 2$ , so the big tank boils off half as fast per kilogram. Bigger cryo-tanks are relatively better at holding cold — the "square-cube law" working in your favour.
Case (b): $d \to 0$ . $\dot{Q} = k A Δ T / d \to \infty$ . Why this step? Zero insulation ⇒ dividing by zero ⇒ infinite heat leak — the mathematical statement of "an uninsulated cryo-tank boils off instantly."

Figure — per-kilogram boil-off vs tank volume, following the curve $V^{- 1/3}$ . The yellow dot is the small reference tank; the pink dot at 8× volume sits on the dashed line at exactly half the reference rate. The downward slope shows why one large cryo-tank keeps its propellant colder than many small ones of the same total mass.

Verify: $8^{- 1/3} = 0.5$ ✓. The $d \to 0$ limit correctly diverges, and the $V^{- 1/3}$ trend explains why launch vehicles favour large single tanks over many small ones for cryogenics. Ties to Tank Design.

Example 7 — Cell G: toxicity comparison (ppm ratios)

Worked example How much more toxic is hydrazine than RP-1?

Compare workplace safety limits (TLV-TWA): hydrazine $= 0.01 ppm$ , RP-1 $= 200 ppm$ . How many times more dangerous is hydrazine by this measure?

Forecast: Both small vs air, but 0.01 vs 200 is a huge gap. Order of magnitude?

Recall the rule: lower TLV = more toxic. A smaller safe concentration means even a whiff is dangerous. Why this step? TLV-TWA is the concentration you may breathe for an 8-hour day; the lower it is, the worse the poison. So the ratio must be RP-1's number ÷ hydrazine's number.
Take the ratio. $\frac{200}{0.01} = 20, 000$ . Why this step? This tells us how many times lower hydrazine's safe limit is — i.e. how many times more toxic.

Verify: Hydrazine's safe exposure is 20,000× stricter than RP-1's ✓ — consistent with the parent note calling it "highly toxic." Ratio is dimensionless (ppm ÷ ppm) ✓. This is the practical push toward Green Propellants.

Example 8 — Cell H: real-world mission word problem

Worked example Choose a propellant for a 2-year deep-space cruise

A probe must fire its main engine after 2 years of coasting, storing 3000 kg of propellant. Options:

LH₂/LOX — $I_{s p} = 450 s$ , but boils off ~2%/day.
NTO/MMH — $I_{s p} = 320 s$ , storable (≈0% boil-off), toxic.

Estimate the LH₂ remaining after 2 years and decide which to fly.

Forecast: 2% per day for 730 days... does any hydrogen survive?

Model boil-off as daily compounding. Fraction remaining $= (1 - 0.02)^{730} = 0.9 8^{730}$ . Why this step? Each day removes 2% of what's left, so it multiplies, not subtracts — geometric decay, like radioactive-style loss.
Evaluate. $0.9 8^{730} \approx 4 \times 1 0^{- 7}$ , i.e. essentially 0 kg survives without active cooling. Why this step? This quantifies "100% boil-off" from the parent note: after 2 years the tank is empty.
Weigh the trade. LH₂'s higher Isp is worthless if the propellant is gone. NTO/MMH keeps all 3000 kg for 2 years; its lower Isp and toxicity are acceptable because the mission is possible at all. Why this step? Storability is a hard constraint here — it gates the whole decision before Isp is even considered.

Verify: $0.9 8^{730} \approx 4.1 \times 1 0^{- 7}$ ✓ — practically total loss. Decision: fly storable NTO/MMH (or Methalox with active cooling as the modern compromise). Matches the parent's Mars-transfer discussion.

Example 9 — Cell I: exam twist (rocket-equation Δv, find the flaw)

Before this one, meet the three symbols it needs, right where they are used:

Worked example "Higher density always beats lower density" — true or false, with numbers

A student claims: "Since RP-1 is denser than LH₂, an RP-1 upper stage always gives more $Δ v$ ." Test this on a stage with dry mass $m_{f} = 1000 kg$ and propellant $m_{p} = 4000 kg$ . Here $m_{0} = m_{f} + m_{p} = 5000 kg$ is the wet mass, so the mass ratio is $m_{0} / m_{f} = 5000/1000 = 5$ . Use the rocket equation, ignoring the (small) tank-mass difference.

Forecast: The claim ignores the exponential role of Isp. Which propellant actually gives more $Δ v$ here?

Compute $Δ v$ for RP-1 ( $I_{s p} = 300$ ). $Δ v = 300 \times 9.81 \times ln 5 = 300 \times 9.81 \times 1.6094 = 4736 m/s$ Why this step? The Rocket Equation converts efficiency and mass ratio into velocity. $ln 5 = 1.6094$ .
Compute $Δ v$ for LH₂ ( $I_{s p} = 450$ ), same mass ratio. $Δ v = 450 \times 9.81 \times 1.6094 = 7104 m/s$ Why this step? Holding the mass ratio fixed isolates the Isp effect — the very thing the student ignored.
Spot the flaw. At equal mass ratio, LH₂ gives 50% more $Δ v$ . The student's claim is false; density helps tankage, not $Δ v$ directly. Why this step? The trap is assuming density enters $Δ v$ . It doesn't — only $I_{s p}$ and the mass ratio do. Density's benefit is indirect (lighter tanks → slightly better mass ratio), and for upper stages it's outweighed by Isp.

Verify: $7104/4736 = 1.50$ ✓ — exactly the Isp ratio $450/300 = 1.5$ , confirming $Δ v \propto I_{s p}$ at fixed mass ratio. Claim busted.

Rapid self-check

Volume for 12,000 kg of LCH₄ at 423 kg/m³

28.4 m³

Which scores higher on ρ·Isp, RP-1 or LH₂

RP-1 (7.7× higher)

Is NTO liquid at −30 °C (Tf = −11.2 °C)

No — it freezes (−30 is colder)

LH₂ boil-off from 810 W leak, 8000 kg, Lv = 445 kJ/kg

≈1.97 %/day

Per-mass boil-off of an 8× larger similar tank

half (8^(−1/3))

How many times more toxic is hydrazine than RP-1 (TLV)

20,000×

LH₂ remaining after 2 yr at 2 %/day

≈0 (4×10⁻⁷ of it)

At fixed mass ratio, LH₂ vs RP-1 Δv ratio

1.5 (= 450/300)

What does TLV-TWA stand for

Threshold Limit Value – Time-Weighted Average (safe 8-hour ppm)