Equal Charges May Attract

Let us consider the interaction between an isolated conducting sphere of radius with net charge + q +q (on its surface) and a point charge + q +q , located at a distance d > R d>R from the center of the sphere. For large distances d d , the sphere and the point charge repel. It turns out, however, that they attract if the point charge is sufficiently close to the sphere. Show that for d R < η \frac{d}{R}< \eta , the sphere attracts the point charge. Determine η \eta .


The answer is 1.618.

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.

1 solution

Anish Puthuraya
Mar 17, 2014

Here is a brief sketch of the problem
alt text alt text
This problem requires the knowledge of Method of Images , which can be referred to from this link.

Using Method of Images, we can find the images of the point charge in the sphere, which results in the following situation,
alt text alt text
Now,
Consider the total charge on the conducting sphere in the above figure.

It is obvious that, according to the figure, the total charge on the sphere is q = R d q \displaystyle q' = -\frac{R}{d}q .

But, since the sphere is isolated, the total charge on it must remain same. In other words, the final charge on the sphere must be the same as + q \displaystyle +q

Hence, to make the necessary corrections, we introduce another positive charge, which is placed at the center of the sphere (to distribute the potential uniformly) and has a magnitude of Q \displaystyle Q .
alt text alt text
Using the fact that the sphere is isolated,
Q + q = q Q+q' = q

Thus,
Q = q + R d q = ( 1 + R d ) q Q = q + \frac{R}{d}q = \left(1+\frac{R}{d}\right)q

Now, we need to find the force on q \displaystyle q due to Q \displaystyle Q and q \displaystyle q' , which is,

F = k Q q d 2 k q q ( d R 2 d ) 2 F = \frac{kQq}{d^2} - \frac{kqq'}{(d-\frac{R^2}{d})^2}

When the charges are just about to start attracting, its clear that F = 0 \displaystyle F=0 .

k q Q d 2 = k q q ( d R 2 d ) 2 \frac{kqQ}{d^2} = \frac{kqq'}{(d-\frac{R^2}{d})^2}

Substituting the values of the charges, and simplifying the expression, we get,

( 1 + d R ) ( d 2 R 2 1 ) 2 = ( d R ) 4 \left(1+\frac{d}{R}\right)\left(\frac{d^2}{R^2}-1\right)^2 = \left(\frac{d}{R}\right)^4

Solving for d R \displaystyle \frac{d}{R} , we find that,

d R = 1.618 , 0.75488 \frac{d}{R} = 1.618, 0.75488

We neglect the second solution, because if d R = 0.75488 \displaystyle \frac{d}{R} = 0.75488 , then, R > d \displaystyle R>d

But, the problem already mentions that d > R \displaystyle d>R . This is a contradiction.

Hence,
d R = 1.618 \frac{d}{R} = \boxed{1.618}

Moderator note:

This is a great solution!

Nice solution Anish! :)

I have one question. Why do you place the charge Q at centre? Can we place it somewhere else?

Pranav Arora - 7 years, 2 months ago

Log in to reply

The sphere is a conductor. Thus, it must be an equipotential surface.

So, if you place the charge elsewhere, the potential due to this charge will be distributed non-uniformly on the sphere.
In other words, the sphere would not be an equipotential surface. So, we have to place the charge at the center.

Anish Puthuraya - 7 years, 2 months ago

thank you very much what nice solution.

shreyas S K - 7 years, 2 months ago

I think that the total charge on the figure is NOT = -qR/d but -q, as can be found out by integrating the surface charge density(as a function of theta) over the surface of the sphere... its done in the very link u shared... so we go ahead and find a equation 2y^4-y^3-4y^2+2=0... y=d/R... can u pls tell how to solve this type of equations?

Subhajit Sinha - 6 years, 10 months ago

No ... the method of images applies to a GROUNDED sphere, not an isolated one. Furthermore, your assumption that the potential is uniformly distributed also seems wrong, since the external point charge will polarize the sphere, inducing a non-uniform charge distribution. I believe that there is a solution to this problem, but it is not nearly so simple to solve as shown in this answer .. I tried to solve it by finding the point where the local surface charge density at the nearest point on the sphere goes to zero. I did it using image charges to start, then realized that was almost certainly wrong, since the equations are only valid for a grounded sphere (i.e. one where the internal charge can assume any value). I then tried to use the Maxwell's equation for electric displacement with the appropriate boundary conditions, but I ran out of time.

David Moore - 6 years, 5 months ago

Log in to reply

Superposing a uniformly charged sphere over the grounded sphere is correct as then all points in the conductor will have same potential. To clarify, the resultant sphere is the superposition of the 2 spheres (grounded and uniformly charged) and it is not the case that the 2 spheres are considered to be both conducting thus changing the distribution of charges.

Dhruv G - 5 years ago

And the best thing is that , the answer results to the Golden Ratio !!!

Richeek Das - 2 years, 6 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...