Suppose we have a 2D polymer model described by a set of 2D vectors {$\mathbf{t}_i$} ($i=1,2,\dots N$) of length $a$.

The energy of the polymer is given by: $$ \mathcal{H}~=~-k\sum^N_{i=1}\mathbf{t}_i\cdot\mathbf{t}_{i+1} ~=~-ka^2\sum^N_{i=1}\cos\phi_i. $$

The constant $k$ is a measure of bending rigidity so that a probability of finding any configuration of the polymer is proportional to $e^{-\frac{\mathcal{H}}{k_BT}}$.

Essentially, this is a discrete version of the wormlike-chain model by Kratky and Porod.

  1. What is the expression for propability density $p(\mathbf{R})$, where $\mathbf{R}$ is an end-to-end vector $\sum_i\mathbf{t_i}$, for this model? Let's assume that $N \rightarrow \infty$.

  2. How do I find linear force extension relation,when a force $\mathbf{F}$ is applied to the ends of the chain? Let's assume that the force is weak.

  3. Can anyone point out useful literature?

  • $\begingroup$ Cool question, +1. I don't know the answer but it looks like the probability distribution should be a function of $|\mathbf{R}|^2$ only, since there are no external fields to break rotational invariance. Maybe try to relate the Hamiltonian to $|\mathbf{R}|^2$. For part 2), once you have the probability distribution p($\mathbf{R}$) you can find the expectation value of the energy $<U(\mathbf{R})>$. This should have a minimum at the equilibrium position. Expand about the minimum to second order; the spring constant is $\partial_R^2<U(\mathbf{R}_0)>$. I look forward to seeing the answer! $\endgroup$ Feb 6, 2013 at 23:36
  • $\begingroup$ Correction: you want to find the minimum of the free energy $F(\mathbf{R}) = <U(\mathbf{R})> - T S(\mathbf{R})$, since your chain is in contact with a heat bath. You can calculate this from the partition function by the usual procedure. $\endgroup$ Feb 7, 2013 at 0:30
  • $\begingroup$ I think $\mathbf{R}$ follows Gaussian distribution with <$\mathbf{R}$> =0 and <$\mathbf{R}^2$> to be determined. Am I right? $\endgroup$
    – molkee
    Feb 7, 2013 at 3:56
  • $\begingroup$ Seems equivalent to asking the magnetization of a classical XY chain. This is certainly well known (and probably amenable to a transfer matrix computation). Or am I missing something? $\endgroup$ Feb 7, 2013 at 13:01
  • $\begingroup$ A good book is Statistical Physics of macromolecules, by Grosberg and Khokhlov, amazon.com/… Already the first chapter addresses the question of the end-to-end distance, although from a more general prospective. $\endgroup$
    – Roger V.
    Apr 13, 2020 at 12:31

1 Answer 1


This question has been worked out in the first chapter of "Statistical Physics of DNA: An Introduction to Melting, Unzipping and Flexibility of the Double Helix", by Nikos Theodorakopoulos. The first chapter is freely available and can be found here: https://www.worldscientific.com/doi/pdf/10.1142/9789811209543_0001. As can be seen, there is no closed form for this problem. Nevertheless, since you are pointing out the limit of $N \rightarrow \infty$, a good approximation is to rescale your average intermonomer distance $a$ to the value of the persistence length $l_p = \kappa/(k_BT)$ and then use it as a freely jointed chain. The bending stiffness will then not be important anymore making this a valid approximation. For the freely jointed chain, the exact expression is also given in the link. Notice also that the freely jointed chain for $N \rightarrow \infty$ becomes a Gaussian.

So to wrap up and give a short answer: For your problem, a Gaussian distribution with an intermonomer distance of $a = \kappa/(k_BT)$ would be a good approximation for $N \rightarrow \infty$. Hope this answers your question :)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.