IPhO 1977 · Theory problem 2

Colours of a soap film

Dipping the frame in a soap solution, the soap forms a rectangle film of length \(b\) and height \(h\). White light falls on the film at an angle \(\alpha\) (measured with respect to the normal direction). The reflected light displays a green color of wavelength \(\lambda_0\).

a) Find out if it is possible to determine the mass of the soap film using the laboratory scales which has calibration accuracy of \(0.1\ \mathrm{mg}\).

b) What color does the thinnest possible soap film display being seen from the perpendicular direction? Derive the related equations.

Constants and given data: relative refractive index \(n = 1.33\), the wavelength of the reflected green light \(\lambda_0 = 500\ \mathrm{nm}\), \(\alpha = 30^\circ\), \(b = 0.020\ \mathrm{m}\), \(h = 0.030\ \mathrm{m}\), density \(\varrho = 1000\ \mathrm{kg\,m^{-3}}\).

Official English text of the problem.

Answers

  • a) The permitted thicknesses and masses are \[ d_j=(2j+1)\frac{\lambda_0}{4\sqrt{n^2-\sin^2\alpha}}, \qquad M_j=(2j+1)(0.0609\ \mathrm{mg}), \quad j=0,1,2,\ldots. \] The thinnest film has \(d_{\min}\simeq0.10\ \mu\mathrm m\) and \(M_{\min}\simeq0.061\ \mathrm{mg}\), so its mass cannot be determined reliably with the stated balance. Higher orders give larger masses; without specifying the order, there is no unconditional yes-or-no answer for an arbitrary green film.

  • b) With illumination remaining at \(30^\circ\), there is no specularly reflected light along the normal, hence no reflected interference colour in that direction. The ideal film is dark in reflected light. A \(540\ \mathrm{nm}\) green maximum would arise if the illumination were also normal, as derived below.

Solution

Treat the film as uniform, plane-parallel, and surrounded by air. Neglect absorption, refractive-index dispersion, and edge diffraction. Interpret the stated green wavelength as a maximum of the reflected intensity.

a) Let \(d\) be the film thickness and \(r\) the refracted angle inside it. Snell’s law gives \[ n\sin r=\sin\alpha, \qquad r=\arcsin\!\left(\frac{\sin30^\circ}{1.33}\right) \simeq22.1^\circ. \]

The reflected light contains a contribution from the front surface and another that enters the film, reflects from the rear surface, and emerges. Comparing their paths at a common outgoing wavefront, their geometrical optical-path difference is \[ \Delta =\frac{2nd}{\cos r}-2d\tan r\sin\alpha. \] Using \(\sin\alpha=n\sin r\), \[ \Delta =\frac{2nd(1-\sin^2r)}{\cos r} =2nd\cos r. \]

Reflection at the front air–soap interface introduces a phase reversal of \(\pi\); reflection at the rear soap–air interface does not. Consequently, constructive interference in reflection requires \[ \frac{2\pi}{\lambda}\,2nd\cos r+\pi=2\pi(j+1), \] or \[ \boxed{2nd\cos r=\left(j+\frac12\right)\lambda}, \qquad j=0,1,2,\ldots. \]

For the observed maximum at \(\lambda_0\), \[ d_j =\frac{(2j+1)\lambda_0}{4n\cos r} =\frac{(2j+1)\lambda_0}{4\sqrt{n^2-\sin^2\alpha}}. \]

The minimum thickness corresponds to \(j=0\): \[ d_{\min} =\frac{500\times10^{-9}} {4\sqrt{1.33^2-\sin^230^\circ}} =1.014\times10^{-7}\ \mathrm m \simeq0.10\ \mu\mathrm m. \]

The mass is density times volume: \[ M_j=\varrho bhd_j. \] Thus \[ M_{\min} =(1000)(0.020)(0.030)(1.014\times10^{-7}) =6.09\times10^{-8}\ \mathrm{kg} =\boxed{0.061\ \mathrm{mg}}. \]

This is smaller than the balance’s \(0.1\ \mathrm{mg}\) calibration accuracy. Therefore, the minimum-thickness film cannot be weighed reliably with that balance.

The interference order matters: the next allowed mass is \[ M_1=3M_{\min}\simeq0.18\ \mathrm{mg}, \] which exceeds the stated accuracy, although the fractional uncertainty would still be large. The green maximum alone therefore does not establish whether an arbitrary film is sufficiently massive for a useful weighing.

b) The incident direction stays fixed at \(30^\circ\).

The front-surface reflected ray leaves at the same angle to the normal as the incident ray. For the ray reflected inside the film, reflection preserves its internal angle \(r\), and refraction upon emergence gives \[ \sin\theta_{\mathrm{out}} =n\sin r =\sin\alpha. \] Hence \[ \boxed{\theta_{\mathrm{out}}=\alpha=30^\circ}. \]

All the interfering reflected rays therefore travel in the \(30^\circ\) specular direction. An observer looking along the normal, \(\theta_{\mathrm{out}}=0\), receives none of these rays. Within the stated approximations, \[ \boxed{I_{\mathrm{reflected}}(0,\lambda)=0}. \] There is consequently no normal-view reflected interference colour, regardless of which allowed thickness the film has. Additional illumination or scattering would require further information.

For completeness, the same minimum-thickness film under normal illumination and normal observation would satisfy \[ 2nd_{\min}=\frac{\lambda_\perp}{2}. \] Its first reflected maximum would then be \[ \lambda_\perp =4nd_{\min} =\frac{\lambda_0}{\cos r} =\frac{n\lambda_0}{\sqrt{n^2-\sin^2\alpha}} \simeq5.40\times10^{-7}\ \mathrm m =\boxed{540\ \mathrm{nm}}, \] which is green. This value applies to that changed illumination geometry.

The solution was checked against the official solution.