A Dipole in an Electric Field — Question 6

Section 22-8 · Physics pages 3622–3623

Question 6 A cubic field exposes the first finite-size correction to dipole force A rigid axial dipole has −q at x −d/2 and +q at x + d/2, with p = qd. The external field on the x axis is Ex(X) = AX3, where A > 0. Its orientation is held fixed. The usual point-dipole approximation is Fx = p dEx/dx. X Ex Positive branch of the cubic field; its negative branch is odd under X reversal. (a) Calculate the exact finite-dipole force and compare it with the point-dipole result. (b) For x̸ = 0, find the correction relative to the point-dipole result. Explain the force at x = 0, where the local first derivative vanishes.
Show solutionHide solution
Question 6 - Solution A cubic field exposes the first finite-size correction to dipole force X Ex (a) Method and calculation Fx = qA[(x + d/2)3 −(x −d/2)3]. Fx = pA ( 3x2 + d2 4 ) , Fpoint = 3pAx2. (b) Interpretation and further calculation Fx −Fpoint Fpoint = d2 12x2 (x̸ = 0). Fx(0) = pAd2/4 > 0. At the origin both charges feel positive-x forces: the negative charge is in a negative field and the positive charge is in a positive field. The local gradient alone misses this finite-separation effect. Physical check With p fixed and d →0, the finite-size correction vanishes as d2; for fixed d it dominates sufficiently close to the origin.

Question 6

A cubic field exposes the first finite-size correction to dipole force

A rigid axial dipole has −q-q at x−d/2x-d/2 and +q+q at x+d/2x+d/2, with p=qdp=qd. The external field on the x axis is Ex(X)=AX3E_x(X)=AX^3, where A>0A>0. Its orientation is held fixed. The usual point-dipole approximation is Fx=pdEx/dxF_x=p\,dE_x/dx.

Diagram shown in the original SVG worksheet.

  1. Calculate the exact finite-dipole force and compare it with the point-dipole result.

  2. For x≠0x\ne0, find the correction relative to the point-dipole result. Explain the force at x=0x=0, where the local first derivative vanishes.

Question 6 - Solution

A cubic field exposes the first finite-size correction to dipole force

Diagram shown in the original SVG worksheet.

(a) Method and calculation

Fx=qA[(x+d/2)3−(x−d/2)3].F_x=qA[(x+d/2)^3-(x-d/2)^3].Fx=pA(3x2+d24),Fpoint=3pAx2.F_x=pA\left(3x^2+\frac{d^2}{4}\right),\quad F_{\mathrm{point}}=3pAx^2.

(b) Interpretation and further calculation

Fx−FpointFpoint=d212x2(x≠0).\frac{F_x-F_{\mathrm{point}}}{F_{\mathrm{point}}}=\frac{d^2}{12x^2}\quad(x\ne0).Fx(0)=pAd2/4>0.F_x(0)=pAd^2/4>0.At the origin both charges feel positive-x forces: the negative charge is in a negative field and the positive charge is in a positive field. The local gradient alone misses this finite-separation effect.

Physical check

With p fixed and d→0d\to0, the finite-size correction vanishes as d2d^2; for fixed d it dominates sufficiently close to the origin.