Testing Coulomb's law

Suppose that a new theory predicts a small deviation of Coulomb's law. That is, imagine that the force of interaction between two charges q 1 q_{1} and q 2 q_{2} separated by a distance r r is given by F ( r ) = k q 1 q 2 r 2 + δ F ( r ) with δ F ( r ) = k q 1 q 2 r 2 α F(r)=k \frac{q_{1}q_{2}}{r^{2}}+ \delta{F}(r) \quad \text{with} \quad \delta F(r)= k' \frac{q_{1}q_{2}}{r^{2-\alpha}} where k k' and α \alpha are constants. Consider now an experiment in which we place a point charge q q with mass m m at the center O of a heavy nonconducting spherical shell of radius R R with uniformly distributed charge Q Q . It turns out that if both charges q q and Q Q are positive and α < 0 \alpha<0 the charge q q will oscillate about the point O. If the deviation δ F ( r ) \delta F(r) were zero, then the force of interaction between the shell and the charge would also be zero. Therefore, from dimensional analysis, we deduce that the period of small oscillations T T must be proportional to T 0 = 2 π m R δ F ( R ) , T_{0}=2 \pi \sqrt{\frac{m R}{|\delta{F(R)}|}}, i.e., T = c T 0 T=c T_{0} . If α = 1 \alpha=-1 , determine the constant of proportionality c c . You may assume that x R 1 \frac{x}{R}\ll 1 .


The answer is 1.73.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

Because the first part k q 1 q 2 r 2 \frac{kq_1q_2}{r^2} of the force does no force to any charge inside the shell, we only need to consider the second part δ F ( r ) \delta F(r) of the force.

Suppose when the distance from the charge q to O is x (x is very small).

Suppose the shell is made up of many very small ring, whose centers are in the same diameter.

Consider one ring of them, which is limited from z to z+dz, has radius r = R 2 z 2 r=\sqrt{R^2-z^2} , distance to charge q is l = r 2 + ( z x ) 2 R 2 2 z x l=\sqrt{r^2+(z-x)^2} \approx \sqrt{R^2-2zx} and charge d Q = Q d z 2 R dQ=\frac{Qdz}{2R} .

The force by this ring to charge q: d F = k q d Q l 3 z x l = k q Q d z ( z x ) 2 R l 4 dF=\frac{k'qdQ}{l^3} \frac{z-x}{l}=\frac{k'qQdz(z-x)}{2Rl^4}

= k q Q d z ( z x ) 2 R ( R 2 2 z x ) 2 k q Q d z ( z x ) 2 R 5 ( 1 + 4 z x R 2 ) =\frac{k'qQdz(z-x)}{2R(R^2-2zx)^2} \approx \frac{k'qQdz(z-x)}{2R^5} (1+\frac{4zx}{R^2})

k q Q d z 2 R 5 ( z x + 4 z 2 x R 2 ) \approx \frac{k'qQdz}{2R^5} (z-x+\frac{4z^2x}{R^2})

Therefore, the total force into the charge q is: F = R R k q Q d z 2 R 5 ( z x + 4 z 2 x R 2 ) F=\int \limits_{-R}^R \frac{k'qQdz}{2R^5} (z-x+\frac{4z^2x}{R^2})

= k q Q x 3 R 4 =\frac{k'qQx}{3R^4} .

Therefore, the period: T = 2 π 3 m R 4 k q Q = 3 T 0 T=2\pi \sqrt{\frac{3mR^4}{k'qQ}}=\sqrt{3} T_0 , so c = 3 c=\sqrt{3} .

how is dQ made to be Q d z 2 R \frac{Qdz}{2R} ? all the charge is Q spread evenly over the 4 π \pi R^2. little area of the ring is 2 π r d z 2\pi rdz r r is radius of ring, not sphere. so then isn't dQ equal Q 2 π r d z 4 π R 2 = Q r d Z 2 R 2 Q\frac{2\pi r dz}{4\pi R^2}=Q\frac{rdZ}{2 R^2} ?

Jindus Mayad - 7 years, 9 months ago

Log in to reply

Note that the thickness won't be dz , it would be d z s i n θ \frac{dz}{sin\theta} , where θ \theta is the angle which line joining ring to center makes with the axis of the ring , hence , sin θ = r R \sin\theta = \frac{r}{R} .

d Q = Q 4 π R 2 2 π r d z sin θ dQ = \frac{Q}{4 \pi R^2} 2\pi r \frac{dz}{\sin\theta} , where s i n θ = r R sin \theta = \frac{r}{R} , hence , d Q = Q d z 2 r dQ = \frac{Qdz}{2r}

jatin yadav - 7 years, 9 months ago
Mark Hennings
Sep 4, 2013

The electrostatic potential for a single point charge q q must be k q r + k q 2 r 2 \frac{kq}{r} + \frac{k'q}{2r^2} if we are to get the right force equation. Thus the potential at a point P P inside the sphere, where O P = r < R OP=r<R , is (writing r 1 2 = R 2 + r 2 2 R r cos θ r_1^2 = R^2+r^2-2Rr\cos\theta ): ϕ ( r ) = π π 2 π R 2 sin θ d θ 4 π R 2 Q ( k r 1 + k 2 r 1 2 ) = Q 2 1 1 ( k R 2 + r 2 2 R r u + k 2 ( R 2 + r 2 2 R r u ) ) d u = k Q 2 R + k Q 4 R r ln ( R + r R r ) = k Q 2 R + k Q 2 R n = 0 r 2 n ( 2 n + 1 ) R 2 n + 1 k Q 2 R + k Q 2 R 2 + k Q 6 R 4 r 2 + \begin{array}{rcl} \phi(r) & = & \int_{-\pi}^\pi \frac{2\pi R^2\sin\theta\,d\theta}{4\pi R^2} Q \Big(\frac{k}{r_1} + \frac{k'}{2r_1^2}\Big) \\ & = & \tfrac{Q}{2}\int_{-1}^1 \Big(\frac{k}{\sqrt{R^2+r^2-2Rru}} + \frac{k'}{2(R^2+r^2-2Rru)}\Big)\,du \\ & = & \frac{kQ}{2R} + \frac{k'Q}{4Rr}\ln\Big(\frac{R+r}{R-r}\Big) \\ & = & \frac{kQ}{2R} + \frac{k'Q}{2R}\sum_{n=0}^\infty \frac{r^{2n}}{(2n+1)R^{2n+1}} \\ & \approx & \frac{kQ}{2R} + \frac{k'Q}{2R^2} + \frac{k'Q}{6R^4}r^2 + \cdots \end{array} and hence the central force on a particle of charge at the point P P is q d ϕ d r k Q q 3 R 4 r = δ F ( R ) 3 R r -q\frac{d\phi}{dr} \; \approx \; -\frac{k'Qq}{3R^4}r \; = \; -\frac{|\delta F(R)|}{3R}r Thus a particle of charge q q obeys the following approximate differential equation for small oscillations about O O : r ¨ + δ F ( R ) 3 m R r = 0 \ddot{r} + \frac{|\delta F(R)|}{3mR}r \; = \; 0 and so the particle performs approximate SHM of period 2 π 3 m R δ F ( R ) = T 0 3 2\pi \sqrt{\frac{3mR}{|\delta F(R)|}} \; = \; T_0\sqrt{3}

Jatin Yadav
Sep 2, 2013

Clearly , the F n e t F_{net} due to traditional coulomb's law would still be zero. Now , we have to check F n e t F_{net} due to deviation .

First consider a ring having charge q q of radius r r

Deviation for field due to the ring at a distance y y = k q y ( r 2 + y 2 ) 2 \frac{k^{'}qy}{(r^2 + y^2)^2} ....... ( i ) (i) (its easy , prove yourself).

Now, consider this example.

Let us consider a ring on the sphere subtending an angle θ \theta at the center.

y f o r r i n g = R c o s θ + x y_{for ring} = Rcos\theta + x , where x x is the amount by which q q has been displaced, and radius r = R s i n θ r = Rsin\theta .

d q r i n g = Q 4 π R 2 2 π R s i n θ R d θ = Q s i n θ d θ 2 dq_{ring} = \frac{Q}{4 \pi R^2}2\pi R sin\theta Rd\theta = \frac{Q sin\theta d\theta}{2} .

Now , d F = k ( Q s i n θ d θ ) q ( R c o s θ + x ) 2 ( ( R + x c o s θ ) 2 + ( R s i n θ ) 2 ) 2 dF = \frac{k^{'}(Q sin\theta d\theta)q(R cos\theta + x)}{2( (R + xcos\theta )^2 + (Rsin\theta)^2)^2} [Using ( i ) (i) ]

= k ( Q s i n θ d θ ) q ( R c o s θ + x ) 2 ( R 2 + x 2 + 2 R x c o s θ ) 2 \frac{k^{'}(Q sin\theta d\theta)q(R cos\theta + x)}{2( R^2 + x^2 + 2Rxcos \theta)^2} .

Using ( 1 + x ) n = 1 + n x (1 + x )^n = 1 + nx for x < < 1 x << 1 for the denominator , We get :

d F = k q Q R 4 ( R c o s θ + x ) ( 1 4 x R c o s θ ) s i n θ d θ dF = \frac{k^{'}qQ}{R^4} (Rcos \theta + x)( 1 - 4\frac{x}{R} cos\theta)sin\theta d\theta

F = 0 π k q Q R 4 ( R c o s θ + x ) ( 1 4 x R c o s θ ) s i n θ d θ \Rightarrow F = \int_{0}^{\pi} \frac{k^{'}qQ}{R^4} (Rcos \theta + x)( 1 - 4\frac{x}{R} cos\theta)sin\theta d\theta

Put cos θ = t \cos \theta = t to get :

F = 1 1 k q Q R 4 ( R t + x ) ( 1 4 x R t ) d t F = \int_{-1}^{1} \frac{k^{'}qQ}{R^4} (Rt + x)( 1 - 4\frac{x}{R}t)dt

F = k q Q 3 R 4 = δ ( F ( R ) 3 R \Rightarrow F = -\frac{k^{'}qQ}{3 R^4} = -\frac{\delta(F(R)}{3R}

T = 3 m R δ ( F ( R ) = 3 T 0 \Rightarrow T = \sqrt{\frac{3mR}{|\delta(F(R)|}} = \sqrt3 T_{0}

Hence , c = 3 = 1.732 c = \sqrt{3} = \fbox{1.732}

Typo: Second last line , i missed 2 π 2 \pi , also after using binomial identity ( 1 + x ) n . . . . (1+x)^n.... , i forgot a factor of 2 in the denominator .

jatin yadav - 7 years, 9 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...