1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Entropy of the universe

  1. Aug 9, 2008 #1
    1. The problem statement, all variables and given/known data

    A model universe comprises 100 atoms in system 1 and 1500 atoms in system 2. Compute the entropy for the universe when there are 3 atoms on in system 2 and 97 atoms on in system 1 (using sterlings approximation).

    3. The attempt at a solution
    I am able to find the entropy of systems 1 and 2 by initially finding the number of microstates and then the equation:

    entropy = boltzmanns * ln(microstates)

    Just wondering how I would get the entropy of the universe though.
  2. jcsd
  3. Aug 10, 2008 #2


    User Avatar

    Well, if you know how many microstates there are for systems 1 and 2, then how many microstates are possible for the combined system (1+2)?
  4. Aug 10, 2008 #3
    I was unsure of the terminology. So taking universe as systems 1 and 2:

    I can find number of microstates using Ω = C(N,n)
    (i.e. the number of ways of moving n atoms from N sites)

    Thus for system 1:
    Ω = C(1500,3)?
    This is a massive number!

    Am I on the right track?
  5. Aug 10, 2008 #4


    User Avatar
    Science Advisor
    2017 Award

    Yup that's the correct way of doing the calculation. The calculation may be made easier by finding a formula for C(N,n) in terms of factorials, then applying Stirling's approximation.
  6. Aug 11, 2008 #5
    sterlings approximation only helps once I get a value of Ω though correct?
    I cannot even compute C(1500,3)=1500!/3!1497! due to overflow!
  7. Aug 11, 2008 #6


    User Avatar
    Science Advisor
    2017 Award

    Try to calculate ln(Ω) instead.
  8. Aug 12, 2008 #7
    Ok, so I managed to compute C(1500,3) and C (100,97).
    I got 561375500 and 646800 respectively.

    (I wish to keep in terms of boltzmanns)
    Therefore for system 1:
    entropy = K*ln(561375500) = 20.15K

    and system 2:
    entropy = K*ln(646800) = 13.38K

    total = 33.53K?

    Where does Sterling's Approximation come into this?
  9. Aug 12, 2008 #8


    User Avatar
    Science Advisor
    2017 Award

    ln(Ω) = ln(1500!/3!1497!) = ln(1500!) - ln(3!) - ln(1497!) = 1500ln(1500) - 1500 - ln(6) - 1497ln(1497) + 1497

    Here's where sterling's approximation saves you from evaluating horrible factorials.
  10. Aug 13, 2008 #9
    yeah i completed it, thanks for help
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook