doPhysics.in
All chapters

Newton, Laplace and the Speed in a Gas

14 min read3 ideas4 checks2 figures to playBy Asheshanand Acharya
What you will learn 0 / 3
The bulk modulus of a gas depends on the process
Look around
The warm bicycle pump

Pump up a bicycle tyre quickly, and then touch the barrel of the pump. It is warm. Each rapid stroke squeezes the air before the heat has time to leave. Pump very slowly, and the barrel stays cool. The same air, squeezed at two different rates, behaves in two different ways.

you will see why the heating of a squeezed gas makes sound travel faster, and by how much.

The bulk modulus of a gas

To use \(v = \sqrt{B/\rho}\) for a gas, we must find the right value of the bulk modulus \(B\). It must be the value that holds under the physical conditions of a sound wave. The bulk modulus measures how strongly a material resists a uniform compression. For a fluid it is defined as

\[ B = -V\,\frac{dP}{dV} \]

where \(V\) is the volume and \(P\) the pressure. The negative sign makes \(B\) positive. An increase in pressure causes a decrease in volume, so \(dV < 0\) when \(dP > 0\).

RememberBulk modulus
\[ B = -V\,\dfrac{dP}{dV} \]
Holds whena fluid under a small, uniform change of pressure.
Symbols\(B\) bulk modulus, \(V\) volume, \(P\) pressure
Units\(B\) in Pa, the same unit as pressure

For a sound wave in a gas, the link between pressure and volume depends on the thermodynamic process. The process is the way the gas is squeezed and released in the rapid compressions and rarefactions of sound. So we must decide which process it is. It may be an isothermal process, at constant temperature \(T\). Or it may be an adiabatic process, with no exchange of heat.

CheckpointUnanswered

Why does the definition \(B = -V\,dP/dV\) carry a minus sign?

Newtons isothermal speed is too low
Maths you need

Newton's isothermal assumption

Newton first assumed that the oscillations of sound in air are slow enough. Each element of gas would then keep a constant temperature. For such an isothermal process, the ideal gas law \(PV = nRT\) holds at constant \(T\). Here \(n\) is the amount of gas in moles and \(R\) the gas constant. Then \(PV\) is a constant, and the bulk modulus follows.

Deriving \(B_{iso} = P\) step 1 of 5
1

\[ PV = \text{constant} \implies P = \dfrac{\text{const}}{V} \]

Why. the ideal gas law \(PV = nRT\) with \(n\) and \(T\) fixed.

So under Newton's assumption \(B = P\), and this leads to

\[ v_{\text{iso}} = \sqrt{\frac{P}{\rho}} \]

For air at \(0\,°\text{C}\), \(P = 1.013 \times 10^5\,\text{Pa}\) and \(\rho = 1.29\,\text{kg/m}^3\). This gives \(v_{\text{iso}} \approx 280\,\text{m/s}\). The measured speed is \(331\,\text{m/s}\). Newton's value is about \(15\,\%\) too low.

Checkpoint · numericUnanswered

Air at \(0\,°\text{C}\) has a pressure of \(1.013 \times 10^5\,\text{Pa}\) and a density of \(1.29\,\text{kg/m}^3\). Find the speed of sound that Newton's isothermal formula predicts, in m/s.

m/s± 2 %
Laplaces adiabatic correction
Maths you need

Laplace's adiabatic correction

Laplace saw the flaw. The pressure changes in sound are extremely rapid. There is not enough time for the gas to exchange heat with its surroundings. So the process is essentially adiabatic. For an adiabatic process in an ideal gas,

\[ PV^{\gamma} = \text{constant} \]

where \(\gamma = C_P/C_V\) is the ratio of specific heats, at constant pressure and at constant volume.

Deriving \(B_{adiabatic} = \gamma P\) step 1 of 6
1

\[ \dfrac{d}{dV}\left(PV^{\gamma}\right) = 0 \]

Why. \(PV^{\gamma}\) is a constant, so its derivative with respect to \(V\) is zero.

Thus the correct bulk modulus for sound in a gas is \(B = \gamma P\). For diatomic gases such as air, \(\gamma\) is about \(1.4\). The adiabatic gas is stiffer than the isothermal one, by the factor \(\gamma\). Fig 4.1 shows why.

RememberBulk modulus of a gas for sound
\[ B = \gamma P, \qquad v = \sqrt{\dfrac{\gamma P}{\rho}} \]
Holds whensound in an ideal gas, whose compressions and rarefactions are adiabatic.
Symbols\(B\) bulk modulus, \(\gamma = C_P/C_V\) ratio of specific heats, \(P\) pressure, \(\rho\) density
Units\(B\) and \(P\) in Pa, \(\rho\) in kg/m³, \(v\) in m/s
Isothermal and adiabatic compression of a gasA graph of the pressure P of a gas against its volume V, with the starting state marked at V0 and P0. Two falling curves pass through that state: the isothermal curve, PV = constant, and the steeper adiabatic curve, P V to the power gamma = constant, with gamma = 1.4. Dashed tangent lines at the starting state show their slopes, minus P over V for the isothermal curve and minus gamma P over V for the adiabatic curve. A readout under each name gives the bulk modulus, B = P and B = gamma P.OVP V0 P0 isothermal B = P adiabatic B = γP
Fig 4.1Two ways to squeeze the same gas from the state \(V_0\), \(P_0\). The adiabatic curve \(PV^{\gamma} = \text{constant}\) is steeper than the isothermal curve \(PV = \text{constant}\). The dashed tangents have slopes \(-\gamma P/V\) and \(-P/V\), so \(B = -V\,dP/dV\) is \(\gamma P\) and \(P\).

Read Fig 4.1 at the starting state. Squeeze the gas a little, and the pressure rises along one of the two curves. When no heat leaves, the gas warms as it is squeezed, so its pressure climbs faster. That steeper climb is a larger \(|dP/dV|\), and so a larger bulk modulus. A stiffer gas hands on each push more quickly.

Newton, Laplace and the measured speed of soundA bar chart of the speed of sound in air at 0 degrees Celsius, in metres per second, from 0 to 400. Three bars: Newton, isothermal, at 280 metres per second; Laplace, adiabatic, at 332 metres per second; and the measured value, 331 metres per second. The value is printed above each bar.100200300400Ov (m/s) 280 m/s Newton isothermal 332 m/s Laplace adiabatic 331 m/s measured
Fig 4.2The speed of sound in air at \(0\,°\text{C}\). Newton's isothermal value \(\sqrt{P/\rho}\) is about \(15\,\%\) below the measured \(331\,\text{m/s}\). Laplace's adiabatic value \(\sqrt{\gamma P/\rho}\) with \(\gamma = 1.40\) agrees with the measurement.

With the numbers of Fig 4.2, Laplace's value is \(\sqrt{1.40 \times 1.013 \times 10^5/1.29} \approx 332\,\text{m/s}\). The measured speed is \(331\,\text{m/s}\). The correction multiplies Newton's value by \(\sqrt{\gamma} = \sqrt{1.40} \approx 1.18\), and that closes the gap.

CheckpointUnanswered

Why are the compressions and rarefactions of a sound wave in air adiabatic rather than isothermal?

Checkpoint · numericUnanswered

For air, \(\gamma = 1.40\). By what factor does Laplace's correction multiply Newton's speed of sound?

± 2 %
Story · Isaac Newton and Pierre-Simon Laplace, 1687
The gap in the Principia

In the Principia of 1687, Isaac Newton gave the first calculation of the speed of sound. He treated air as a spring that kept its temperature. His answer fell well short of the speeds measured with cannon and distant observers. Newton tried to explain the gap away, and the puzzle stayed open for more than a century. In 1816 Pierre-Simon Laplace pointed out that the rapid squeezes of sound heat the air. With the adiabatic stiffness \(\gamma P\), theory and measurement agreed.

Now return to the bicycle pump. Rapid strokes warm the air because the heat has no time to escape. Sound squeezes the air hundreds of times faster still. So each compression is adiabatic, and the air behaves as a stiffer spring. That stiffness, \(\gamma P\) instead of \(P\), raises the speed of sound by a factor \(\sqrt{\gamma}\), about \(18\,\%\).

Exam traps
  1. Using \(B = P\) for sound in a gas.
    Sound is adiabatic. Use \(B = \gamma P\), so \(v = \sqrt{\gamma P/\rho}\).
  2. Forgetting the product rule when differentiating \(PV^{\gamma}\).
    Both \(P\) and \(V^{\gamma}\) change with \(V\), so there are two terms: \(V^{\gamma}\,dP/dV + \gamma PV^{\gamma - 1}\).
  3. Taking the minus sign of \(B = -V\,dP/dV\) to make \(B\) negative.
    \(dP/dV\) is itself negative, so \(B\) comes out positive.