Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Scalar field

  1. May 16, 2017 #1
    Hello! Can someone explain to me how does a scalar field changes under a Lorentz transformation? I found different notations in different places and I am a bit confused. Thank you!
     
  2. jcsd
  3. May 16, 2017 #2

    vanhees71

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    If ##x'=\Lambda x##, where ##\Lambda## is a Lorentz-transformation matrix, then a scalar field obeys by definition the transformation law
    $$\phi'(x')=\phi(x)=\phi(\Lambda^{-1} x').$$
     
  4. May 16, 2017 #3
    Thank you for your reply. This makes sense. However I found in Peskin's book on QFT a definition that is different from yours by a prime ( ' ) - I attached a screenshot of it. That is what got me confused. Do you know what does he mean by his notation?
     

    Attached Files:

  5. May 16, 2017 #4

    George Jones

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The expression that vanhees71 wrote is equivalent to the expression in Peskin and Schroeder, and both expressions are equivalent to

    $$ \phi'(Fred)=\phi(\Lambda^{-1} Fred).$$

    Why?
     
    Last edited: May 16, 2017
  6. May 17, 2017 #5
    Sorry I am a bit confused. What is primed and what is unprimed?
     
  7. May 17, 2017 #6

    vanhees71

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Just read the equation as a whole. My final equation was
    $$\phi'(x')=\phi(\Lambda^{-1} x').$$
    Now rename ##x'## back to ##x##, and you get Peskin Schroeder's formula. You can name it "Fred" as suggested in #4 (although that's a bit unusual ;-)).
     
  8. May 17, 2017 #7
    What do you mean by rename? If we have a frame S stationary with respect to the field and S' moving with respect to the field (so ##\phi'## and x' are measured in S') then we have by definition ##\phi'(x')=\phi(x)## and by the Lorentz transformation we also have ##\phi(x)=\phi(\Lambda^{-1} x')##. This make sense. But in Peskin he has a mix of both ##\phi'(x)## and this is what confuses me. How does he get to a mix of primed and unprimed indices without a ##\Lambda## factor somwhere? And I am not sure how can you rename x' to x, once you decided which moves and which is fixed. (so to be clear, I understand that choosing prime and unprime as moving or not moving is arbitrary, but Pesking seems to mix them, which confuses me). Thank you for help!
     
  9. May 17, 2017 #8

    vanhees71

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The formula of itself is unique. You can name the argument as you like. You can as well write the law as
    $$\phi'(y)=\phi(\Lambda^{-1} y).$$
    Of course, the prime at the field symbol on the left-hand side is crucial!
     
  10. May 17, 2017 #9
    But the argument of ##\phi'## is always seen from S', right?
     
  11. May 17, 2017 #10

    vanhees71

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Yes, sure.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Loading...