Problem set #1 for 5.72 (spring 2012) Cao
I.
Consider a random walk on a linear three-site model. π and π are the probabilities of moving right and left respectively. ( π + π = 1 )
π΄
1 π π
π΄
2 π π
π΄
3
1) Write the transition matrix π .
2) Find the stationary distribution and show that it satisfies detailed balance.
3) For the special case of π = π =
1
2
, compute the probability at π = 3 after π steps, π π
(π ) , given π π
(0) = πΏ π1
.
4) *Repeat the calculation in 3) for π ≠ π .
II.
A one-dimensional random walk on an infinite lattice is described by a master equation ππ π ππ‘ = π
(π π−1
− π π
) + π(π π+1
− π where π and π are the forward and backward rate constants.
1) Calculate: π
),
〈π(π‘)〉 = οΏ½ ππ π
(π‘) π
,
〈π 2 (π‘)〉 = οΏ½ π 2 π π
(π‘) , π and
〈πΏπ 2 〉 = 〈π 2 (π‘)〉 − 〈π(π‘)〉 2
2) *Given that the random walker starts at π = 0 , i.e. π π
.
(0) = πΏ π0
, find π π is initially at site π΄
1
(π‘) .
III.
For the three-site model the forward rate is π and the backward rate is π . The walker
. π
π΄ π
2
π΄
1
π΄
3
1
1) Calculate the probability that the walker has not arrived at π΄
3
.
by time π‘ .
2) Calculate the mean first passage time from π΄
1
to π΄
3
IV.
*Consider a random walk on an infinite π -dimensional lattice with transition rate to
. the next neighbor being πΌ . The random walker starts at
1) Derive a formal expression for π π π = 0 , i.e. π π
(π‘) .
2) Show the long-time behavior is diffusive with diffusion constant πΌ and
− π
2
(0) = πΏ π0 lim
3) π
0 π‘→∞ that π
0
(π‘) ∝ π‘ .
(π‘) is the probability that the walker arrives at site 0 for the first time. Show
1
4) where πΜ
0
and πΜ
0 πΜ
0
(π ) = 1 −
π < 1 for π > 2 .
(Note: references Reichl and Montroll.)
0
∞
π = 1 for π = 1 or 2 πΌπΜ
0
are the Laplace transforms of π
0
Define the total return probability as
Show that and
π = ∫ π
0
* honorary problems will not be counted towards grade.
(π ) ,
and π
0
(π‘)ππ‘ .
.
2
MIT OpenCourseWare http://ocw.mit.edu
5.72 Statistical Mechanics
Spring 2012
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms .