Quantum Field Theory II (8.

324) Fall 2013 Assignment 8

Nov. 26, 2013 Due Tuesday Dec. 10, 2013

Please remember to put your name at the top of your paper. Readings Peskin & Schroeder chapters 10, 12, 13. Weinberg Vol. 1 chapter 12 and Vol. 2 chapter 18.

Problem Set 8 1. Renormalization group properties (30 points) (). (a) Consider a coupling constant and a redened coupling constant Find the general transformation law for the beta function, namely the relation between () and (). If we think of as a coordinate we see that transforms as a tensor. What kind of tensor ? (b) Assume that () = b2 2 + b3 3 + b4 4 + and consider the perturbatively dened and invertible coupling constant redenition: () = + a2 2 + a3 3 + . ( ) writing it in the form Calculate ( ) = 2 + 3 + 4 + b2 b3 b4 Verify that: i. b2 = b2 and b3 = b3 . ii. It is possible to make b4 anything you want by such a coupling redenition. = F is also a xed iii. Let = F denote a xed point. Show that point. How are the derivatives and related at the xed point?

(c) Consider the dierential equation for a massless coupling g dg = bg 2 cg 3 dg 4 d (1)

with b, c, d numerical constants. Show that one can write a solution to the above equation in the form ln 1 c = + 2 ln bg () + O(g ()) bg () b (2)

where is an integration constant, which can be considered as a dynamically generated scale. Argue that is renormalization group invariant. (d) More generally, show that in a renormalizable theory with a dimensionless coupling constant g () and no other dimensional parameter (like in a nonAbelian gauge theory), dimensional transmutation happens. That is, show that g () can be written in a form g () = f log (3)

with a universal scale and f a function depending on the specic theory. (Inverting (2) gives a specic example of f .) 2. Running coupling near a xed point (20 points) Suppose g (0) = g0 . Find the behavior of g () as for the following situations: (a) -function has a simple zero: (g ) = b(g a), (b) (g ) has a higher zero (g ) = b(g a)n , (c) (g ) is given by (g ) = g (a2 g 2) . 3. Decoupling a massive eld (50 points) Consider the following Lagrangian for two scalar elds 1 and 2 1 1 11,0 4 12,0 2 2 22,0 4 1 2 1 2 2 . L = ( 1 )2 ( 2 )2 m2 1,0 1 m2,0 2 2 2 2 2 4! 1 2 2 1 2 4! 2 (7) 2 (6) b > 0, n>1 (5) b>0 (4)

(a) Renormalize the general interactions by writing some relation between the bare parameters 11,0 , 22,0 , 12,0 and the renormalized parameters 11 , 22 , 12 . Use dimensional regularization and minimal subtraction. Find the beta functions. (b) Suppose we took 11,0 = 22,0 = 0 and 12,0 = 0 /3 in the original Lagrangian, so that 1 1 1 1 2 2 0 2 2 2 L = ( 1 )2 ( 2 )2 m2 ( + 2 1,0 1 m2,0 2 2 ) . (8) 2 2 2 2 4! 1 Show that the relations between bare and renormalized parameters, as well as the beta functions in (a) are consistent with having a well-dened renormalization and beta function for the single coupling 0 . Determine this beta function and use it to write () in terms of (M ). (c) Now consider the limit where the mass m2 becomes much larger than external momenta and the xed mass unit (of dimensional regularization), we will show that 2 essential decouples up to a nite renormalization of the remaining coupling, as follows. Compute 1 1 elastic scattering in (7) up to one-loop and take the limit of the amplitude as m2 2 becomes large. Show that the resulting scattering amplitude is the same as what follows from the following low-energy eective Lagrangian where only 1 remains 0 2 = 1 ( 1 )2 1 m 2 4 (9) L 2 2 0 1 4! 1 provided we make the identication (to order 2 )
1 1 m2 2e ( ) = ( ) + 2 ( ) ln . 6 (4 )2 42

(10)

0 in (9) In the above equation () denotes the renormalized coupling for while () denotes that for 11,0 in (7). (d) Argue that equation (10) implies that (m2 ) = (m2 ) + O(2 ) , (11)

where O(2 ) has no large logarithms. Equation (11) means that the running coupling of the original theory and that of the eective theory must be adjusted to match at the mass scale m2 . (e) Verify the consistency of (10) with the values of the beta functions for and for .