* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Homework 9 Answers
Survey
Document related concepts
Equations of motion wikipedia , lookup
Navier–Stokes equations wikipedia , lookup
Equation of state wikipedia , lookup
Noether's theorem wikipedia , lookup
N-body problem wikipedia , lookup
Field (physics) wikipedia , lookup
Nordström's theory of gravitation wikipedia , lookup
Perturbation theory wikipedia , lookup
Lorentz force wikipedia , lookup
Aharonov–Bohm effect wikipedia , lookup
Maxwell's equations wikipedia , lookup
Electric charge wikipedia , lookup
Partial differential equation wikipedia , lookup
Transcript
Homework 9 Answers, 95.657 Electromagnetic Theory I Dr. Christopher S. Baird, UMass Lowell Jackson 4.10 Two concentric conducting spheres of inner and outer radii a and b, respectively, carry charges ±Q. The empty space between the spheres is half-filled by a hemispherical shell of dielectric (of dielectric constant ε/ε0), as shown in the figure. +Q -Q a b (a) Find the electric field everywhere between the spheres. SOLUTION We can split the region where we want to know the electric field into two regions, the left hemisphere and the right hemisphere, solve for the field in each region separately and then apply the boundary conditions to get the final solution. In the left region, there are no dielectrics and no charges. So the electric potential obeys the Laplace equation: ∇ 2 L =0 If we align the z-axis pointing to the right, the problem has azimuthal symmetry. The general solution to the Laplace equation in spherical coordinates for azimuthal symmetry is: L r , , =∑ Al r l Bl r−l−1 P l cos l The potential on the surface of a conductor is always constant: L r =a =C C=∑ Al a l Bl a −l −1 Pl cos l Because this equation must hold for all values of the independent polar variable and because the Legendre polynomials are orthogonal, all coefficients must equate independently: C= A0 B0 and a Bl =−Al a 2l 1 for l > 0 The solution now becomes: L r , , =C− l B 0 B0 ∞ r a ∑ Al − a r a r l =1 l 1 P l cos The other surface is also a conductor and must have a constant potential as well: L r =b=D l B 0 B0 ∞ b a D=C− ∑ Al − a b l =1 a b l 1 P l cos Because this equation must hold for all values of the independent polar variable and because the Legendre polynomials are orthogonal, all coefficients must equate independently: B0 = a b D−C and a−b Al =0 for l > 1 The solution now becomes: L r , , =C a b D−C −1 1 a−b a r At this point C and D are undetermined, so we can redefine them to simplify the equation: L r , , =D C r This leads to a total electric field of: E L= C r 2 r We could have guessed this form of the solution based on the symmetry of the problem, but it is often safer and more instructive to go through all the steps. The right hemisphere is a separate region and can be now solved separately. In the region between the shells, we can tell that there are no free charges and no bound charges (the bound charges, or polarization charges, will reside along boundaries, but not in the volume where we want to know the potential). The total electric potential then obeys the Laplace equation: 2 ∇ R=0 The same process as above is applied leading to the same form of the solution: E R= F r r2 We can now apply boundary conditions where the dielectric material meets free space. The tangent component of the electric field must be continuous across the boundary: E L ×=E R× [E L⋅r =E R⋅r ]= /2 C F = 2 2 r r C=F which leads to: E R=EL = C r r2 Because all materials are linear, this directly tells us that D L =0 C C r and D R= 2 r 2 r r Because the free charges give rise to the D field, the fact that D on the left is different from D on the right means that the free charge densities on each side must be different. The last fact that we know is that there is a total free charge Q on the inner conductor surface and a total free charge -Q on the outer conductor surface. We only know the total free charge, not the free charge density. We cannot assume the free charge spreads out uniformly over the sphere, because there is nothing in the problem to suggest that. Because we don't know the charge density, we cannot use a local boundary condition, but instead an integrated boundary condition that makes use of the total charge. We draw an integration sphere with radius r where a < r < b so that it completely encloses the free charge Q and use Gauss's law in integral form. We could use Gauss's law for the E, D, or P fields. We only know the total free charge, which only gives rise to the D field, so we use: ∮S D⋅r d 3 x=Q We must be careful here. We found above that the D field is not the same over this integration sphere, so we must split the integral into two pieces: 2 / 2 2 ∫ ∫ DR⋅r r 2 sin d d ∫ ∫ D L⋅r r 2 sin d d =Q 0 0 0 / 2 2 /2 2 C ∫ ∫ sin d d 0 C ∫ ∫ sin d d =Q 0 0 0 / 2 C= Q 2 0 We now have our final solutions: D L= E= 0 Q 0 2 r 2 r and D R= Q r 2 0 2 r Q r 2 0 2 r (b) Calculate the surface-charge distribution on the inner sphere. SOLUTION: The distribution of free surface charges σ is linked to the D-field according to: ∇⋅D= We draw a Gaussian pillbox straddling the surface and use the divergence theorem to find: [D2−D1⋅n= ]on S Inside a conductor there are no fields, so D1 = 0. L =[ DL⋅r ] r=a and R=[ DR⋅r ]r=a L= 0 Q 0 2 a 2 r and R= Q r 0 2 a 2 (c) Calculate the polarization-charge density induced on the surface of the dielectric at r = a. SOLUTION: We use Gauss's law on a pillbox surface again, but instead only enclose the polarization charge. [P2−P1⋅n=− pol ]on S Inside the conductor there are no fields, so P2 = 0 and the normal now points inwards, n = -r [P⋅r =− pol ]r =a pol =[−P ]r =a For linear materials, we know P=−0 E pol =[−−0 E ]r =a pol = 0 −Q 0 2 a 2 Of course, this only applies to the right side. There is no dielectric material on the left side and therefore no polarization charge. Supplemental Problem D Consider an infinitely long, straight, cylindrical wire consisting of a core of dielectric ε1 with radius a surrounded by a layer of dielectric ε2 with radius b such that b is greater than a and the surface at radius b is held at potential V(φ). Find the potential everywhere inside the tube. SOLUTION: Because everything is uniform and infinite in the dimension parallel to the wire, this problem reduces down to a two-dimensional problem in polar coordinates. We also note that there are no free charges anywhere and the polarization charges only reside on the boundaries, so we can reduce the problem to two different problems in two different regions, the core region and the outer region. Within each region, there are neither free charges nor polarization charges. The total charge density is zero in each region so that we seek solutions to the Laplace equation for the total electric potential. The general solution to the Laplace equation in polar coordinates is: , =a 0b0 ln A0 B0 ∑ a b − A e i B e−i ,≠0 In both regions, we need a valid solution on the whole angular extent, so the single-valued requirement reduces the solution to: , =a 0b 0 ln ∑ a m mb m −m Am e i m B m e−i m m ,m ≠0 The core region includes the origin, so that the solution must reduce to: 1 ,=∑ m Am e i m B m e−i m m We can let the summation go over all integers and not just positive integers and combine terms: 1 ,= ∞ ∑ m=−∞ Am ∣m∣e i m The outer region contains no such requirement and stays as 2 , =a 0b0 ln ∑ a m m b m −m Am ei mB m e −i m m , m≠0 We can again let the summation go over all integers (except m = 0) and not just positive integers and combine terms: 2 , =a 0b0 ln ∞ ∑ a m ∣m∣b m −∣m∣ e i m m=−∞ ,m≠0 Apply the boundary condition of continuous tangential total electric field E at the interface between two dielectrics: [ E2⋅=E 1⋅ ]=a [ ∂ 2 ∂ 1 = ∂ ∂ ∞ ∑ m=−∞ , m≠0 ] =a i m a m a∣m∣bm a −∣m∣ e i m = ∞ ∑ m=−∞ ,m ≠0 i m Am a∣m∣e i m Due to the orthogonality of the complex exponentials, all the coefficients must match up, leading to: Am =a m bm a −2∣m∣ The second boundary condition is: [2 E2⋅ =1 E1⋅ ]=a [ 2 2 ∂2 ∂ 1 =1 ∂ ∂ ] =a ∞ ∞ b0 2 ∑ a m∣m∣a∣m∣−1−∣m∣b m a−∣m∣−1 e i m =1 ∑ ∣m∣Am a∣m∣−1 e i m a m=−∞ , m≠0 m=−∞, m≠0 b0 =0 Am = 2 a −b a −2∣m∣ 1 m m Apply the last boundary condition: 2 =b=V V =a 0 ∞ ∑ a m b∣m∣b m b−∣m∣ e i m m=−∞, m≠0 Multiply both sides by a negative complex exponential, integrate, and use the orthogonality of complex exponentials to pick out a specific term from the sum: 2 ∫ V e−i m d = a m b∣m∣bm b−∣m∣ 2 0 2 a 0= 1 ∫ V d 2 0 The m = 0 terms are uniquely determined. For m ≠ 0, we have the three equations in boxes and three unknowns. We can thus solve this system of equations to get a unique solution. We find: b m=2 −1 a∣m∣C m a m = 21 a −∣m∣C m Am =2 2 C m 2 where we define the known constant C m = a∣m∣b∣m∣ ∫ V e−i m d 2∣m∣ 2∣m∣ 2 a 2−1 b 21 0 The final solution for the total electric potential then becomes: ∞ 1 ,=2 2 ∑ Cm m=−∞ ∞ ∣m∣ e a i m ∣m∣ , = ∑ C m 2 1 a m=−∞ 2 e a 2− 1 ∣m∣ im 2 where a∣m∣b∣m∣ Cm= ∫ V e−i m d 2∣m∣ 2∣m∣ 2 a 2−1 b 21 0 One way to check this is to let a approach zero and check that we get the solution to the problem presented in an earlier lecture: ∣m∣ e 1 ∞ , = ∑ 2 m=−∞ b 2 2 i m ∫ V e−i m ' d ' 0