5. Molecular PhysicsSavchenko Formulas, chapter 5 of 14, 56 formulas
Sections follow the book, and within a section the most used formulas come first. An italic problem number means the formula appears in its statement. Rest the cursor on a number to see the statement.
A solid angle is the area a cone cuts out on the unit sphere. The whole sphere subtends $4\pi$, a half space $2\pi$. The fraction of molecules heading for a distant target is its solid angle over $4\pi$.
$\displaystyle v \sim \sqrt{\frac{kT}{m}}\ \text{(по порядку величины)}$
$v_{\text{кв}}$
root of the mean square speed
$\mu$
molar mass
R
gas constant
The thermal speed of a molecule follows from $m v^2/2 = 3kT/2$. Through the molar mass the same formula reads $\sqrt{3RT/\mu}$, so light gases move faster at the same temperature. For estimates $v \sim \sqrt{kT/m}$ is enough.
Through a hole small against the mean free path molecules leave one by one, and the flux is proportional to $n v$, that is to $p/\sqrt{mT}$. So a light gas leaks faster than a heavy one, and two vessels at different temperatures settle at $p_1/\sqrt{T_1} = p_2/\sqrt{T_2}$, not at equal pressures.
Temperature measures the mean kinetic energy of random motion. Each translational degree of freedom carries $kT/2$, three of them $3kT/2$. The same holds for a Brownian particle and any body in thermal equilibrium, so $kT$ estimates the thermal swing of a pendulum or a torsion fibre.
$$dN = N f(v)\,dv, \qquad \int_0^\infty f(v)\,dv = 1$$
$f(v)$
probability density of the speed
$dN$
number of molecules with speeds from $v$ to $v + dv$
N
total number of molecules
The fraction of molecules within a speed interval $dv$ is $f(v)\,dv$, and the whole area under the graph is one. Averages and fluxes are computed through $f$. If every speed is multiplied by $l$, the distribution stretches and normalization asks for a factor $1/l$.
$$n = n_0\,\exp\!\left(-\frac{U}{kT}\right), \qquad n(h) = n_0\,\exp\!\left(-\frac{\mu g h}{RT}\right)$$
$\displaystyle p = p_0 e^{-\mu g h/RT}\ \text{(барометрическая формула)}$$\displaystyle \frac{n_1}{n_2} = \exp\frac{U_2 - U_1}{kT}$
U
potential energy of a molecule
n0
concentration where $U = 0$
h
height
In a force field the concentration of molecules falls with their potential energy as $e^{-U/kT}$. In gravity this is the barometric formula, and pressure falls with height as the concentration does. The ratio of concentrations at two points depends only on the energy difference.
$$\Delta v = \frac{F\tau}{m}, \qquad \frac{m v'^2}{2} = \frac{m v^2}{2} - F l$$
F
force on the molecule
$\tau$
flight time
l
path against the force
While a molecule crosses a region with a field, the force changes its momentum by $F\tau$ and the field's work changes its kinetic energy. This is how a molecular beam is deflected and sorted by speed, including in a rotating selector where the speed is tied to the pitch and the angular velocity.
A velocity component is Gaussian with variance $kT/m$. The fraction of molecules with $v_x$ above a threshold $b$ is an error integral, falling as $e^{-b^2}$ for a high threshold. For the speed the Gaussian gets an extra factor $4\pi v^2$.
A molecule flying through a gas hits every molecule inside a cylinder of cross-section $\sigma = \pi d^2$, so the path between collisions is about $1/(n\sigma)$. The factor $\sqrt{2}$ accounts for the other molecules moving too, with a relative speed $\sqrt{2}$ times larger on average. The free path is inversely proportional to pressure.
$$\sigma = \pi(r_1 + r_2)^2, \qquad Z = \sqrt{2}\,n\sigma\langle v\rangle$$
r1, r2
radii of the colliding molecules
Z
collisions of one molecule per second
$\langle v\rangle$
mean speed
Two molecules collide when their centres come within the sum of the radii, hence the cross-section $\pi(r_1+r_2)^2$. In a second a molecule covers $v$ and hits $n\sigma v$ neighbours, with the $\sqrt{2}$ correction for their motion. The time between collisions is $\lambda/v$.
$$q = -\kappa\frac{dT}{dx}, \qquad \kappa = \frac{1}{3}\lambda\langle v\rangle n c_1$$
$\displaystyle \kappa \sim \frac{1}{r^2\sqrt{\mu}}\ \text{(не зависит от давления)}$$\displaystyle W = \kappa S\frac{\Delta T}{d}$
q
heat flux density
$\kappa$
thermal conductivity
c1
heat capacity of one molecule, $\frac{i}{2}k$
Molecules from the hot side bring more energy to a surface than those from the cold side take away, the difference building up over one free path. The product $n\lambda$ does not depend on pressure, so a gas conducts heat independently of pressure as long as the free path is shorter than the vessel. Light gases with small molecules conduct best.
$$D = \frac{1}{3}\lambda\langle v\rangle, \qquad \frac{dm}{dt} = -D S \frac{d\rho}{dx}$$
D
diffusion coefficient
$\lambda$
mean free path
$d\rho/dx$
density gradient of the carried gas
The flux of matter is proportional to the concentration gradient, and the coefficient is $\lambda v/3$ in order of magnitude. Molecules arrive at a surface from one free path on either side, carrying the density difference. Through $\lambda \propto 1/p$ the diffusion coefficient is inversely proportional to pressure.
With random steps in independent directions it is the squared displacements that add, so after $N$ steps a molecule has strayed $\lambda\sqrt{N}$. Hence the diffusion time over a distance $L$ grows as $L^2/D$, with $D \sim \lambda v$. This estimates how long a smell takes to cross a room.
$$F \approx p S v\sqrt{\frac{\mu}{RT}} \sim n m v_{\text{т}} v S$$
p
gas pressure
S
cross-section of the body
v
speed of the body
$\sqrt{RT/\mu}$
thermal speed of the molecules
When the free path exceeds the body, molecules hit it independently. More of them arrive at the front, and with more momentum, than at the back, and the pressure difference is $p$ times $v/v_{\text{th}}$ in order of magnitude. The force grows linearly with speed while the body is slower than the molecules.
When the free path exceeds the gap, a molecule flies wall to wall without collisions and carries $\frac{i}{2}k\,\Delta T$ per crossing. The heat flux is proportional to the number density and hence to pressure, unlike in a dense gas. The Pirani gauge and the Dewar flask rest on this.
$$\frac{p_1}{\sqrt{T_1}} = \frac{p_2}{\sqrt{T_2}}, \qquad p \approx \frac{F\,T_1}{S(T_2 - T_1)}$$
p1, p2
pressures in the vessels
T1, T2
temperatures of the vessels
F
force on the plate
When vessels join through a hole smaller than the free path, equilibrium means equal opposite fluxes $n\sqrt{T}$, not equal pressures. Molecules leaving the hot side of a plate carry more momentum, and in a rarefied gas the plate feels a force that measures the pressure.
A molecule leaving a rotating wall carries its tangential speed and keeps its angular momentum on the way to the other wall. The momentum the molecules bring balances only when $\omega r^2$ agrees for both cylinders, and a rarefied gas transmits rotation this way rather than by viscosity.
$\displaystyle p = nkT$$\displaystyle \rho = \frac{p\mu}{RT}$
p
pressure
V
volume
$\nu$
number of moles
$\mu$
molar mass
R
gas constant, $8.31$ J/(mol K)
The equation of state ties the pressure, volume and temperature of any amount of ideal gas. It holds while molecules rarely interact, that is for ordinary gases at moderate densities. All the isoprocesses follow from it, and for a mixture it is written for each gas separately.
$\displaystyle p = \sum_i n_i kT\ \text{(закон Дальтона)}$$\displaystyle n = \frac{N}{V}$
n
number density, molecules per unit volume
k
Boltzmann's constant, $1.38\cdot10^{-23}$ J/K
The equation of state per molecule. The pressure of a mixture is the sum of the partial pressures, since each kind of molecule pushes on the wall independently. Through $n$ one computes the free path, fluxes and the number of molecules in a volume.
$$p = p_0 + \frac{Mg}{S}, \qquad p = p_0 + \rho g h$$
p0
atmospheric pressure
M
mass of the piston or load
S
area of the piston
h
height of the liquid column
The gas pressure under a piston or a liquid column follows from the equilibrium of the piston or the column, and the volume and temperature then follow from the equation of state. A column of mercury or water in a tube adds $\rho g h$ to the atmosphere or takes it away, depending on which side the gas is.
isobar, a free piston or the atmosphere sets the pressure
In a rigid vessel pressure is proportional to absolute temperature, under a free piston the volume is. Both lines pass through the origin on the Kelvin scale, which makes converting states easy.
From $pV = \frac{m}{\mu}RT$ the density is proportional to pressure and molar mass and inversely proportional to temperature. At one pressure and temperature the densities of gases are as their molar masses. This gives the lift of a balloon and the mass of gas in a cylinder.
At constant temperature the product of pressure and volume of a given mass of gas stays the same. Air trapped in a tube by a column of mercury or water behaves so when moved slowly enough for heat to escape.
The molar mass in grams per mole equals the relative molecular mass numerically. Air has 29, nitrogen 28, oxygen 32, carbon dioxide 44. SI formulas take kilograms per mole.
A balloon is lifted by the buoyant force less the weight of the gas inside, both expressed through densities at one pressure, that is through molar masses. The envelope adds its own weight, so for a small balloon it wins and there is a smallest radius. Hot air is lighter than cold air at the same pressure.
A mole holds $N_A$ molecules and weighs $\mu$. The gas constant and Boltzmann's constant are tied by $R = kN_A$. A mixture's molar mass is its total mass over its total number of moles.
For a fixed mass of gas $pV/T$ is the same in every state, because the right side of the equation of state is $\nu R$. The isoprocesses are the cases with one parameter fixed.
$\displaystyle A = \nu R\,\Delta T\ \text{(изобара)}$$\displaystyle A = \frac{(p_1 + p_2)(V_2 - V_1)}{2}\ \text{(линейный процесс)}$$\displaystyle A = \frac{\nu R(T_1 - T_2)}{\gamma - 1}\ \text{(адиабата)}$
A
work done by the gas on its surroundings
p
gas pressure
$dV$
change of volume
The work of a gas is the area under the process on the $pV$ diagram, positive on expansion. On an isobar it is $p\Delta V = \nu R\Delta T$, on a straight-line process the area of a trapezoid, over a cycle the area inside the loop. The work of outside forces on the gas is equal and opposite.
$\displaystyle \delta Q = dU + p\,dV$$\displaystyle \Delta U = Q - A$
Q
heat supplied to the gas
$\Delta U$
change of internal energy
A
work done by the gas
Heat received by a system raises its internal energy and does work against outside forces. Signs matter, an ideal gas's $\Delta U$ depends only on temperature while the work depends on the path on the $pV$ diagram. In an adiabatic process $Q = 0$ and the work comes out of the internal energy.
$\displaystyle \Delta U = \frac{i}{2}\nu R\,\Delta T$$\displaystyle U = \frac{i}{2}pV$$\displaystyle \varepsilon = \frac{i}{2}kT\ \text{(на молекулу)}$
i
degrees of freedom, 3 for a monatomic, 5 for a diatomic gas
$\nu$
number of moles
CV
molar heat capacity at constant volume
The internal energy of an ideal gas is the sum of the molecules' kinetic energies, $kT/2$ per degree of freedom. It depends only on temperature, not on volume or pressure, so its change in any process is $\nu C_V\Delta T$. Through the equation of state $U = \frac{i}{2}pV$.
The ratio of heat capacities enters the adiabatic equation and the speed of sound. A monatomic gas has $\gamma = 5/3$, a diatomic one $7/5$. Through $\gamma$ the internal energy reads $pV/(\gamma-1)$.
The heat needed to warm a body is proportional to its mass and to the temperature change. In an insulated system the heat given up by some bodies equals the heat received by others, and this equation together with the latent heats solves every mixing problem. A heater of constant power delivers heat proportional to time.
On an isotherm $p = \nu RT/V$, and the integral $\int p\,dV$ gives the logarithm of the volume ratio. The internal energy stays fixed, so all the heat supplied turns into work. On compression the gas's work is negative.
At constant volume all the heat goes into internal energy, at constant pressure also into the work $R\,\Delta T$ per mole, hence $C_p - C_V = R$. A monatomic gas has $C_V = \frac{3}{2}R$, a diatomic one $\frac{5}{2}R$. The specific heat is the molar one over $\mu$.
$\displaystyle T p^{(1-\gamma)/\gamma} = \text{const}$$\displaystyle A = \frac{\nu R(T_1 - T_2)}{\gamma - 1} = -\Delta U$
$\gamma$
adiabatic exponent $C_p/C_V$
Without heat exchange $\nu C_V\,dT = -p\,dV$, and with the equation of state this gives $pV^\gamma = \text{const}$. An adiabat is steeper than an isotherm, and adiabatic compression heats the gas. A process is adiabatic when fast or the vessel is insulated, and the work equals the drop in internal energy.
$$Q = \nu C_V\Delta T + p\,\Delta V, \qquad p = p_0 + \frac{F}{S}$$
F
force on the piston, friction or a load
S
piston area
p0
atmospheric pressure
While the piston stands, the gas heats at constant volume, and once the gas's force beats friction and the atmosphere it moves isobarically at $p_0 + F/S$. The heat is split between internal energy, work against the atmosphere, lifting the load and friction, and heat from the piston's friction may partly return to the gas.
work per cycle, the area inside the loop on the $pV$ diagram
$Q_{\text{подв}}$
heat taken from the heater
$Q_{\text{отд}}$
heat given to the cooler
Over a cycle the internal energy returns to its value, so the work equals the heat taken minus the heat given up and the area of the loop. The work is summed over the legs, the heat on each from the first law with its sign. In a cycle of two isobars and two isochores the corner temperatures obey $T_2^2 = T_1T_3$.
$$pV^n = \text{const}, \qquad C = C_V + \frac{R}{1-n}$$
$\displaystyle C = 2\nu R\ \text{(газ под пружиной, }p \propto V)$$\displaystyle n = 0, 1, \gamma, \infty\ \text{(изобара, изотерма, адиабата, изохора)}$
n
polytropic index
C
heat capacity of the gas in that process
Any process with constant heat capacity is a polytrope, and the converse holds. The heat capacity is found by writing the first law for a small step and dividing by $dT$. For a gas under a spring-loaded piston, where $p \propto V$, a monatomic gas has $2\nu R$.
A rising parcel of air expands adiabatically, since heat exchange with its surroundings is slow, and cools. With $dp = -\rho g\,dz$ and the adiabat in the form $\nu C_p\,dT = V\,dp$ this gives a drop of about one degree per hundred metres. It estimates the temperature on a summit and the height of clouds.
The largest jet speed comes when the whole enthalpy of the gas in the vessel turns into kinetic energy, that is on discharge into vacuum. Discharging into a medium at pressure $p_2$ the factor $1 - (p_2/p_1)^{(\gamma-1)/\gamma}$ reduces it. A rocket's thrust is the mass flow times the exhaust speed.
$\displaystyle c_pT + \frac{v^2}{2} = \text{const}$$\displaystyle H = u + \frac{p}{\rho}\ \text{(энтальпия единицы массы)}$
$p, \rho$
pressure and density of the gas
v
flow speed
$\gamma$
adiabatic exponent
In steady flow without heat exchange the kinetic energy per unit mass plus its enthalpy $u + p/\rho$ is conserved, the flow work $p/\rho$ being added to the internal energy. For an ideal gas the enthalpy is $c_pT$, so heating of the jet and its acceleration trade one for the other.
$$\rho v S = \rho' v' S', \qquad F = \rho v S\,(v' - v)$$
$\rho v S$
mass flow through a cross-section $S$
$v, v'$
speeds before and after the section
F
force on the gas over the section
In a steady jet the same mass passes every cross-section each second. That mass's momentum gain per second equals the total force, including the pressure difference $pS$ at the ends of the section. Together with the energy equation this handles flow with heat addition.
$$p = p_1 p_2 \cdots p_N, \qquad p = \left(\frac{1}{2}\right)^N$$
pi
probabilities of the separate independent events
N
number of molecules or trials
The probability that several independent events happen together is the product of their probabilities. One molecule sits in half the vessel with probability $1/2$, all $N$ at once with probability $2^{-N}$, and over an observation time $\tau$ such a state lasts $\tau\,2^{-N}$ on average. For macroscopic $N$ that is never.
Entropy is the logarithm of the number of ways a macrostate can be realized. For an ideal gas that number grows as $V^N$ in the coordinates and $T^{iN/2}$ in the momenta, so the system heads for the state of largest statistical weight. Through $k\ln\Omega$ the thermodynamic $\Delta S = Q/T$ gets its meaning.
$$dN = n f(v)\,v\,S\,dt\,dv, \qquad j \sim \frac{1}{4}n\langle v\rangle$$
n
number density
$f(v)$
velocity distribution
S
area of the surface
In a time $dt$ the surface $S$ is crossed by molecules from a cylinder of height $v\,dt$, and fast molecules are weighted more in the flux than in the bulk. For estimates the flux per unit area is $n v/4$ or, more roughly, $n v/6$. This gives the outflow through a small hole and the speed of molecules that reached a neighbouring vessel.
The number of ways $k$ molecules out of $n$ can be in a chosen part equals the binomial coefficient, and the probability of that state is this times the probability of one particular arrangement. The most probable states are those with the most ways, that is with an even split.
$$\Delta S = \int_1^2\frac{\delta Q}{T}, \qquad \Delta S = C\ln\frac{T_2}{T_1}$$
$\displaystyle \Delta S = \nu C_V\ln\frac{T_2}{T_1} + \nu R\ln\frac{V_2}{V_1}$$\displaystyle \Delta S = \frac{\lambda m}{T}\ \text{(фазовый переход)}$
$\delta Q$
heat received on a small step
T
temperature at which it is received
C
heat capacity of the body
Entropy is a state function, so its change is computed along any reversible path between the same states, dividing each bit of heat by the temperature. Warming a body of constant heat capacity gives the logarithm of the temperature ratio, a phase transition gives $\lambda m/T$. An ideal gas's entropy depends on $T$ and $V$.
No engine between two temperatures beats the Carnot cycle's $1 - T_2/T_1$, and only a reversible engine reaches it. Run backwards the Carnot cycle is the ideal refrigerator or heat pump, moving $A\,T_2/(T_1 - T_2)$ of heat per unit work. For an actual cycle the efficiency is computed from the heat on each leg.
The entropy of a closed system does not decrease, and in a reversible process it stays constant. So heat never flows by itself from cold to hot, and a reversible engine gives the cooler at least $Q_1 T_2/T_1$. The condition $\Delta S = 0$ gives the largest work obtainable from two bodies and their final temperature $\sqrt{T_1T_2}$.
specific heat of fusion, $3.3\cdot10^5$ J/kg for ice
L
specific heat of vaporization, $2.3\cdot10^6$ J/kg for water
m
mass that changed phase
During melting and boiling the temperature does not change, and the heat supplied breaks bonds and is proportional to the mass. In the heat balance these terms sit next to $cm\,\Delta T$. The reverse transitions, freezing and condensation, release the same heat.
vapour density, for saturated vapour a function of temperature alone
$\phi$
relative humidity, $p/p_{\text{sat}}$
Vapour mixed with air behaves as an ideal gas at its own partial pressure, and the total pressure is the sum. Saturated vapour over its liquid has a pressure fixed by temperature alone, on compression it condenses rather than rising in pressure. Humidity is the fraction of saturation.
$$p = p_0 + \frac{2\sigma}{r}, \qquad \frac{p_r}{p_\infty} = \exp\frac{2\sigma\mu}{\rho R T r}$$
$\sigma$
surface tension
r
radius of the drop or bubble
p0
pressure outside, for a soap bubble $4\sigma/r$
A curved surface presses on the liquid beneath it with an extra $2\sigma/r$, a bubble with two surfaces with $4\sigma/r$. Because of it the saturated vapour pressure over a small drop exceeds that over a flat surface, so small drops evaporate while large ones grow. The exponent is the work against the Laplace pressure per mole over $RT$.
With a constant supplied power the heat is proportional to time, so the time to reach boiling and the time to boil away the same mass stand for $cm\,\Delta T$ and $Lm$. The ratio of the times gives the ratio of the heats, which is how a kettle measures the heat of vaporization. The power spent on evaporation is $\lambda\,dm/dt$.
$\displaystyle F = \frac{2IS}{c}\ \text{(зеркало)}$$\displaystyle F = \frac{\Phi S}{4\pi r^2 c}\ \text{(чёрная пластинка у точечного источника)}$
I
intensity, power per unit area
$\rho$
reflectivity, 0 for black, 1 for a mirror
u
energy density of the radiation
Radiation carries momentum $E/c$, so an absorbing wall feels a pressure $I/c$ and a mirror twice that, since the reflected light carries as much momentum back. In a cavity of isotropic radiation the wall gets a third of the energy density, as with a gas. This compares the Sun's light pressure with its gravity for dust grains.
$$\Phi = \varepsilon\sigma S T^4, \qquad \sigma = 5{,}67\cdot10^{-8}\ \frac{\text{Вт}}{\text{м}^2\,\text{К}^4}$$
$\displaystyle u = \frac{4\sigma}{c}T^4\ \text{(плотность энергии равновесного излучения)}$$\displaystyle \Phi = \sigma S(T_1^4 - T_2^4)\ \text{(обмен между телами)}$
$\Phi$
radiated power
$\varepsilon$
emissivity, 1 for a black body
S
surface area
T
temperature of the body
A black body radiates $\sigma T^4$ per unit surface, a grey one $\varepsilon$ times less and absorbs less by the same factor. A body in surroundings absorbs $\sigma T_0^4$ from the walls, so the net flux goes as $T^4 - T_0^4$. A screen between hot and cold halves the flux, and in equilibrium sits at $T/\sqrt[4]{2}$.
$\displaystyle T = T_\odot\sqrt{\frac{R_\odot}{2L}}\ \text{(равновесная температура на расстоянии }L)$
$\Phi$
total power of the source
R
distance from the source
I
power per unit area across the rays
A source's power spreads over a sphere of area $4\pi R^2$, so the intensity falls as the square of the distance. Hence the equilibrium temperature of a planet or a dust grain, whose absorbed power $I\pi r^2$ equals the emitted $\sigma T^4\cdot4\pi r^2$, depends only on the distance and not on the size.