See also: Course-wide Conventions & Notation
Run the code: Notebook for Section 2.5, in which you can change and rerun the Python cells of this section in your browser.
Overview and Learning Objectives¶
A particle in a box is the simplest quantized model of translation and provides a clean route to the translational partition function. We review the 1D and 3D energy levels, derive in the high-temperature/large-volume limit, and use to recover ideal-gas thermodynamic relations.
Learning objectives:
Write the quantized energy levels for a particle in a 1D box and generalize to a 3D rectangular box and cube.
Derive the single-particle partition function and its continuum approximation in the limit of small level spacing.
Define the thermal de Broglie wavelength and express .
Use to obtain , , and .
Core Ideas and Derivations¶
Review of the Particle in a Box¶
Particle in a One-Dimensional Box¶
Figure 2.5.1 shows the first four energy levels and wavefunctions of a particle in a one-dimensional box.
Source
import sys
import matplotlib.pyplot as plt
import numpy as np
sys.path.append("..") # the book folder, which holds figstyle
import figstyle as fs
def draw(c):
fig, ax = plt.subplots(figsize=(4.8, 3.7), constrained_layout=True)
x = np.linspace(-0.5, 0.5, 200) # position in units of L
levels = range(1, 5)
# Energy levels in units of E_1 = h^2 / (8 m L^2), with each wavefunction drawn on its level
for n in levels:
ax.plot([-0.5, 0.5], [n**2, n**2], color=c.gray, lw=1.2)
ax.plot(x, n**2 + np.sin(n * np.pi * (x + 0.5)), color=c.blue)
ax.text(0.53, n**2, rf"$\psi_{n}$", color=c.blue, va="center", clip_on=False)
# Walls and floor of the box
ax.plot([-0.5, -0.5, 0.5, 0.5], [18, 0, 0, 18], color=c.ink, lw=1.6, solid_capstyle="butt", zorder=3)
ax.set(xlabel="Position $x$", ylabel="Energy", xlim=(-0.5, 0.5), ylim=(0, 18),
xticks=[-0.5, 0, 0.5], yticks=[n**2 for n in levels])
ax.set_xticklabels(["$-L/2$", "0", "$L/2$"])
ax.set_yticklabels(["$E_1$", "$E_2 = 4E_1$", "$E_3 = 9E_1$", "$E_4 = 16E_1$"])
ax.tick_params(axis="y", length=0, pad=6)
for side in ("left", "bottom"):
ax.spines[side].set_visible(False)
return fig
fs.show(draw, alt="Box with walls at x equal to minus L over 2 and L over 2. Four horizontal energy levels lie at 1, "
"4, 9, and 16 times the lowest energy. On each level, the wavefunction is drawn as a sine curve "
"that is zero at both walls, with one half-wave for the first level, two for the second, three for "
"the third, and four for the fourth.")Figure 2.5.1:Energy levels (gray lines) and wavefunctions (blue curves) of a particle in a one-dimensional box with walls at . Each wavefunction is drawn on its energy level, and the energy increases as .
For a particle in a one-dimensional box of length , the quantized energy levels are
where is Planck’s constant, is the particle mass, and is a positive integer.
Particle in a Three-Dimensional Box¶
In three dimensions, the energy levels are
where , , and are positive integers, and , , and are the respective side lengths of a rectangular box.
Particle in a Cube¶
For a cube of side , . The energy simplifies to
where is the cube volume.
Partition Function for a Particle in a Cube¶
The single-particle partition function for a three-dimensional cube is
Define . Then
Approximation of the Sum¶
The parameter is the ratio of , the energy that sets the spacing between the levels, to . For helium at 300 K in a volume of 1.00 L, (see the worked example below). When (large or large ), the sum is well approximated by an integral:
Substituting gives
We often define the thermal de Broglie wavelength as
Hence,
Cubing this result, as in Eq. (2.5.5), gives the single-particle partition function. Because the error of the approximation is negligible when , the result is written as an equality:
Partition Function for Particles in a Cube¶
For identical, non-interacting, indistinguishable particles, Section 2.4 gives the total partition function as
The factor of corrects for particle indistinguishability: permutations of identical particles do not produce new states.
Ensemble Averages for Particles in a Cube¶
Natural Logarithm of the Partition Function¶
Taking the logarithm of gives
Often, the Stirling approximation () is used for large . The internal energy, heat capacity, and pressure derived below require only derivatives of . The entropy is obtained from in Section 4.3 and evaluated for this gas in Section 5.1.
Internal Energy¶
Because is proportional to , depends on as
The internal energy then follows from the derivative in Section 2.3:
Heat Capacity at Constant Volume¶
From , it follows that
Pressure¶
To find the pressure, we use the expression from Section 2.3,
recovering the ideal gas law.
Translational Partition Function¶
Because the particle-in-a-box spectrum describes translational motion, the single-particle partition function
is called the translational partition function. It underpins the classical description of dilute gases when , the mean spacing between particles, so quantum effects are negligible at ordinary densities and temperatures.
Figure 2.5.2 draws the two lengths to scale for helium. In the gas at and , the conditions of the worked example below, the mean spacing is about 69 times . In liquid helium at its normal boiling point of , is larger than the mean spacing, which is the case that requires the full quantum treatment of identical particles.
Source
from matplotlib.patches import Circle
from scipy.constants import h, k, u
m_He = 4.002602 * u # mass of a helium-4 atom in kg
def thermal_wavelength(T):
"""Thermal de Broglie wavelength of a helium atom in m."""
return h / np.sqrt(2 * np.pi * m_He * k * T)
T_gas, P_gas = 300, 1.00e5 # K, Pa
spacing_gas = (k * T_gas / P_gas) ** (1 / 3) # (V/N)^(1/3) of an ideal gas in m
T_liquid, density_liquid = 4.22, 125 # K, kg/m^3 (liquid helium at its normal boiling point)
spacing_liquid = (m_He / density_liquid) ** (1 / 3)
def draw_atoms(ax, c, spacing, wavelength):
"""Atoms on a grid at the mean spacing, each with a disc of diameter Lambda. Lengths are in units of the spacing."""
radius = wavelength / (2 * spacing)
for i in range(4):
for j in range(3):
if radius > 0.1:
ax.add_patch(Circle((i, j), radius, fc=c.tint(c.blue, 0.72), ec=c.blue, lw=1.0, alpha=0.85))
else: # no outline, which would make the dot larger than Lambda
ax.add_patch(Circle((i, j), radius, fc=c.blue, ec="none"))
ax.add_patch(Circle((i, j), 0.11, fc="none", ec=c.muted, lw=0.8)) # a ring to find the dot
# the mean spacing, between the centers of two neighbors
for x in (0, 1):
ax.plot([x, x], [-0.68, -0.92], color=c.muted, lw=0.9)
ax.annotate("", xy=(0, -0.8), xytext=(1, -0.8),
arrowprops=dict(arrowstyle="<|-|>", color=c.ink, lw=1.2, mutation_scale=9, shrinkA=0, shrinkB=0))
ax.text(1.18, -0.8, f"$(V/N)^{{1/3}}$ = {1e9 * spacing:.3g} nm", ha="left", va="center")
# the thermal wavelength: a dimension line where the disc is large enough, and a leader where it is a dot
label = f"$\\Lambda$ = {1e9 * wavelength:.3f} nm"
if radius > 0.1:
for x in (-radius, radius):
ax.plot([x, x], [2.68, 2.92], color=c.muted, lw=0.9)
ax.annotate("", xy=(-radius, 2.8), xytext=(radius, 2.8),
arrowprops=dict(arrowstyle="<|-|>", color=c.ink, lw=1.2, mutation_scale=9, shrinkA=0, shrinkB=0))
ax.text(radius + 0.18, 2.8, label, ha="left", va="center")
else:
ax.plot([0.015, 0.3], [2.03, 2.62], color=c.muted, lw=0.9)
ax.text(0.22, 2.8, label, ha="left", va="center")
ax.set(xlim=(-0.85, 3.85), ylim=(-1.05, 3.05), aspect="equal")
ax.axis("off")
def draw(c):
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(fs.WIDTH, 3.35), constrained_layout=True)
draw_atoms(ax1, c, spacing_gas, thermal_wavelength(T_gas))
draw_atoms(ax2, c, spacing_liquid, thermal_wavelength(T_liquid))
fs.panel(ax1, "a", f"Helium gas at {T_gas:g} K and {P_gas / 1e5:g} bar")
fs.panel(ax2, "b", f"Liquid helium at {T_liquid:g} K")
return fig
fs.show(draw, alt="Two panels, each with twelve helium atoms on a square grid at the mean spacing. In panel a, the gas "
"at 300 K and 1 bar, the mean spacing is 3.46 nanometers and the thermal wavelength is 0.050 "
"nanometers, so the disc around each atom is a barely visible dot inside a gray ring that marks its place. In panel b, the liquid at 4.22 K, "
"the mean spacing is 0.376 nanometers and the thermal wavelength is 0.425 nanometers, so the discs "
"of neighboring atoms overlap.")Figure 2.5.2:Thermal de Broglie wavelength and mean spacing of helium atoms, drawn to scale. The atoms are placed on a square grid at the mean spacing, and each disc has the diameter . The two panels have different scales. (a) Gas at and , where the mean spacing is 69 times . Each disc is a barely visible dot, and a gray ring marks its place. (b) Liquid at its normal boiling point of , with a density of , where is 1.13 times the mean spacing and the discs overlap.
Computational Studio: Ideal Gas¶
In Figure 2.5.3, you can change the number of atoms , the volume , and the temperature of a small sample of a noble gas and follow the quantities that its partition function gives.
Figure 2.5.3:A small sample of an ideal monatomic gas. Panel (a) shows atoms in a cube of volume , seen from one side. Panel (b) lists the thermal de Broglie wavelength , the mean distance between atoms, the translational partition function (Eq. (2.5.17)), (Eq. (2.5.11)), the internal energy (Eq. (2.5.14)), the pressure (Eq. (2.5.16)), and the entropy, which Section 4.3 obtains from . Because is small, is evaluated exactly and not with the Stirling approximation.
You can also open the studio in a new tab: Ideal Gas Studio.
Worked Example¶
Estimating and (helium, 300 K)¶
For a particle of mass , the thermal de Broglie wavelength is
Take helium: , .
Compute
Compute the single-particle translational partition function in
Compute for the same volume, for which
Compare the mean spacing between atoms with at
The ideal gas law gives the number of atoms in the volume,
so the mean spacing between atoms is
and the number of thermally accessible translational states per atom is
Result. In a macroscopic volume, and , consistent with the continuum approximation used to derive . At 1 bar, is also much larger than , which is the condition assumed for in Section 2.4.
Concept Checks¶
Why does the sum over quantum numbers become well-approximated by an integral at high or large ?
What does the condition “interparticle spacing ” mean physically?
Why does translation give but not for an ideal gas?
Where does the factor enter when connecting microscopic states to macroscopic entropy?
Key Takeaways¶
Quantized translation in a box leads, in the continuum limit, to .
sets the scale of quantum wavepacket “size”; classical behavior emerges when is small compared to typical spacings.
For an ideal monatomic gas, reproduces and .
Partition functions connect microscopic spectra to thermodynamic equations of state via derivatives of .