How can the symmetry factor for a diagram be quickly determined and verified?

  • Level: Graduate 
  • Thread starter Thread starter ChrisVer
  • Start date Start date
  • Tags Tags
    Calculation
Join the discussion
Registration is free. Start your own thread to ask a follow-up.
2 replies · 2K views
Science Advisor
Messages
3,372
Reaction score
465
Can someone check the symmetry factor I've found in the following diagram and verify it?
Do you know if there is any way to determine the SF of a diagram fast?

So I have [itex]\phi_x[/itex] which can be contracted with 4 [itex]\phi_w[/itex] : 4
Then [itex]\phi_y[/itex] which can be contracted with 3 [itex]\phi_w[/itex] : 3
Then I have 2 [itex]\phi_w[/itex] which can be contracted with 4 [itex]\phi_u[/itex]: 2x4
Then I have 3 [itex]\phi_u[/itex] which can be contracted with 4 [itex]\phi_z[/itex]: 3x4
The rest [itex]\phi_u[/itex] get contracted together (only 2 left), whereas the [itex]\phi_z[/itex] we get a factor of 2 since there are 2 possibilities to contract 3 fields.
Finally the last contraction gives just a factor 1 (no possible alternative choices).

So the result from contractions is 4x3x2 x4 x3 x2 x 4 = 4! x 4! x 4
The diagram is of order 3 (3 vertices) so there is a factor 1/3! and also the 1/4! from the coupling constant for [itex]\phi^4[/itex] theory.
So is it correct to say that the symmetry factor is afterall SF=16?
 
Attachments
  • test414.png
    test414.png
    3.4 KB · Views: 513
Physics news on Phys.org
Your factor of 2 (in blue) should be 3. From three fields, there are three possible distinct pairs to construct (AB, AC and BC).
But there is also a second point: there should be a factor of 1/4! for each vertex, so you should have 1/(4!)^3 instead of 1/4!
As for a shortcut: usually people drop the denominator and simply calculate the symmetry factor of the graph (the number of automorphisms in mathspeak). I think there is a nice discussion in Peskin and Schroeder if I recall correctly.
 
  • Like
Likes   Reactions: ChrisVer
In Peskin they say that they drop the 1/4! factor and write the vertex factor as [itex]\int d^4 z (-i \lambda)[/itex] because they say that to a generic vertex has four lines coming in from four different places so the various contractions of [itex]\phi \phi \phi \phi[/itex] is 4!...Also the n! from the Taylor expansion will cancel because of interchanging of vertices...
But I don't understand either of these explanations...
So in case I used (1/4!)3 I would have obtained the correct result?

[itex]A= \frac{16}{4! 4!}= \frac{1}{3! 3!} = \frac{1}{36}[/itex]
So the symmetry factor is: 36?