0% found this document useful (0 votes)
217 views15 pages

The Random Walk Problem

This document discusses the random walk problem. It introduces the concept of a random walk where a particle takes random steps in arbitrary directions and sizes over many iterations. It then provides the key results for a one-dimensional random walk: 1) The average total distance traveled after N steps is equal to N times the average step size (Equation 1). 2) The total standard deviation of the distance traveled after N steps is equal to the square root of N times the standard deviation of a single step (Equation 2). It then provides two numerical examples to demonstrate how to apply these results to calculate the average distance and fluctuation for a particle undergoing a random walk over time.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
217 views15 pages

The Random Walk Problem

This document discusses the random walk problem. It introduces the concept of a random walk where a particle takes random steps in arbitrary directions and sizes over many iterations. It then provides the key results for a one-dimensional random walk: 1) The average total distance traveled after N steps is equal to N times the average step size (Equation 1). 2) The total standard deviation of the distance traveled after N steps is equal to the square root of N times the standard deviation of a single step (Equation 2). It then provides two numerical examples to demonstrate how to apply these results to calculate the average distance and fluctuation for a particle undergoing a random walk over time.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

Chapter 3

The Random Walk Problem


3.00 Introduction

The random walk problem, while important in its own right, is another
example of a problem in which we can easily pass from the low statistics
limit to the high statistics limit. So we study this problem to see,
again, how statistical systems can be deterministic when they are large
enough.

Consider a system in which a single particle takes steps which are


arbitrary in direction and in size. A single particle in a room full of
air is one such example. Another example is that of a completely
inebriated person trying to walk home. In the first case we cannot
predict the distance between the particle collisions or in which
direction the particle will go next. In the second case we cannot
predict the size or direction of the inebriate’s next step with any
certainty and that is why we refer to this as the random walk problem.
Our goal will be to make rather precise predictions about each of these
systems after many steps (or collisions) have occurred.

3.01 One-Dimensional Random Walk

We shall begin by considering a one dimensional system in which the


particle can move either to the right or the left an arbitrary distance.
We shall use the small letter "s" to represent a possible single step.
To find the average single step size, +s,, we must sum over all the steps,
s = ∑ s Ps ,
s
where PS is the probability of undergoing a step of size s. If we are to
consider arbitrary steps then this summation should be replaced by an
integral over all step sizes:

s =
∫ s P ( s) ds.

Recalling the definition for the standard deviation and rewriting it in


integral form too, yields,

σ 12 =
∫ ( s − s ) 2 P( s) ds.

Thus given a P(s) function we can integrate to get +s, and σ1. The type
of function used for P(s) would depend on the details of the interactions
and because these can be very complicated we shall assume that we know
+s, and σ1 for our system of interest. This assumption is not as great as
it may seem because both of these quantities can be measured
experimentally for real systems.

In one important case we know +s, exactly without doing any measurements.
If there is no driving force in the system causing the particle to prefer
one direction over the other then +s, must be exactly zero. Examples of
RANDOM WALK Page 47

driving forces are a fan blowing in the room full of air or a sloped
sidewalk that the inebriate attempts to navigate along. Keep this case
in mind when reading random walk problems as no numerical value need be
specified for +s, if a driving force does not exist.

Let us first calculate the total distance traveled after N steps. We


label each step with the letter i and si will be the size of step i. The
total distance traveled, capital S, is,
N

S= ∑ s.i =1
i

We now take the average of this total distance and recognize that the
average of a sum is the sum of the averages:

∑ ∑ ∑
N N N

<S> = si = si = s.
i =1 i =1 i =1
In the last step we used +s,, the average value of a single step. The
last summation is just summing +s, up N times. So we have the first
result for the random walk problem:

Average Total Distance:


(One Dimension)
S = N s (1)

Our next goal will be to find the standard deviation for the total trip
distance. Here we have to average the square of the difference between
S and +S,:


N

σ 2 = (S − S )2 = ( s − s )]2 .
i =1
We now rewrite the square of the quantity to be summed as a product of
two summations and pull the summation signs out in front of the averages:

∑ ∑ ∑∑
N N N N

σ2 = ( si − s ) (s j − s ) = ( si − s ) ( s j − s ) .
i =1 j =1 i =1 j =1

Now we regroup the terms in the summations according to whether i=j or


i=
/j:
RANDOM WALK Page 48

∑∑ ∑∑
N N N N

σ2 = ( si − s ) ( si − s ) + ( si − s ) ( s j − s ) ,
j =i i =1 j ≠ i , j =1 i =1

∑ ∑∑
N N N

σ2 = ( si − s ) 2 + ( si − s ) ( s j − s ) .
i =1 j ≠ i , j =1 i =1

The first term can be simplified by replacing the average of the sum by
the sum of the average. The last term can be simplified using the theorem
that says the average of product is the product of the averages when the
events are statistically independent. This condition is meet because
this sum contains products referring to different step numbers. Then we
have,

∑ ∑∑
N N N

σ2 = ( si − s ) 2 + ( si − s ) ( s j − s ) .
i =1 j ≠ i , j =1 i =1

∑ ∑∑
N N N

σ2 = ( si − s AV ) 2 + ( si − s ) ( s j − s ).
i =1 j ≠ i , j =1 i =1

∑ ∑∑
N N N

σ2 = ( si − s AV ) 2 + ( s − s ) ( s − s ).
i =1 j ≠ i , j =1 i =1


N

σ2 = ( si − s AV ) 2 + 0.
i =1
In the next to last line we used +s, for the average of a single step.
The quantity remaining in the last summation is just the square of the
standard deviation for a single step. Since we are summing this N times
we obtain,

σ 2 = N σ 12
Taking the square root yields our second key result for the random walk
problem:
RANDOM WALK Page 49

Total Standard Deviation:


(One Dimension)
(2)
σ= N σ1

3.02 Two Numerical Examples

Example #1: The valence electrons in most metals are essentially free
to wander about. They possess kinetic energy and move with velocities
of the order of 106 m/sec. As they move around the metal they undergo
random collisions with the positive ion cores, the surfaces of the sample
or with other electrons. In a particular metal (copper in this case)
they travel an average distance of 3 x 10-8 meters between collisions and
undergo an average of 5 x 1013 collisions per second. If one could follow
a particular electron for one minute, how far would that electron go on
average and what is the fluctuation (standard deviation) in that
distance? Treat the metal as a one dimensional sample.

We begin the solution of this problem by noting that the electrons


undergo random collisions and that there is no force directing them in
any one direction or another. Therefore +s, must be 0. The average
distance given in the problem, 3 x 10-8 meters, represents the fluctuation
in the distance traveled per collision. Therefore σ1 = 3 x 10-8 meters.
In one minute an average electron undergoes (5 x 1013 collisions per
second) x (60 seconds) = 3 x 1015 collisions. Up to this point we have
identified three relevant variables from a reading of this random walk
problem: +s,, σ1, and N.

Using the results of Equations (1) and (2) yields,

S = N s = (3 x 1015 ) (0) = 0,

σ = (3 x 1015 ) (3 x 10 −8 meters),
σ = 1643
. meters.
So after one minute the electron is where it started from but with an
uncertainty of 1.643 meters!

Example #2: Suppose a sample of the same metal from Example #1 was to
have an electric field applied whose strength was sufficient to cause the
electron to move along the line of the field an average of 10-14 meters
between each collision. How far would an electron go on average after
one minute and what is the fluctuation in that distance? Again, treat
the metal as a one dimensional sample.

Solution: We begin as before by identifying the three relevant variables


+s,, σ1, and N. Since the time period, one minute, is the same as before,
N = 3 x 1015 collisions. The fluctuation in single steps, σ1, is
determined by the collisions (which are random) not by the field so it,
too, remains the same, σ1 = 3 x 10-8 meters. The presence of the field
RANDOM WALK Page 50

causes the electron to move a finite amount with each single step;
therefore +s, = 10-14 meters.

Using Equations (1) and (2),


S = N s = (3 x 1015 ) (10− 14 meters) = 30 meters,

σ= N σ1 = (3 x 1015 ) (3 x 10 −8 meters) = 1643


. meters.

Thus in one minute an average electron moves 30 meters with an


uncertainty of 1.643 meters in this distance.

3.03 Statistical Determinism in the Random Walk Problem

Example #2 provides a convenient way of gaining insight into the concept


of statistical determinism. If we calculate the relative fluctuation in
single steps we obtain,
σ13 x 10−8 meters
ℑrel = = − 14
= 3 x 106 ,
s 10 meters

which is comparatively large. Clearly the random motion which causes the
fluctuations, 3 x 10-4 meters, dominates over the driving motion due to
the field, 10-14 meters. Thus our knowledge of what happens in a single
step is very uncertain. When we do a similar calculation for the entire
period of one minute we obtain,
σ 1643
. meters
ℑrel = = = 0.055,
S 30 meters
which is comparatively small. After one minute we have rather precise
knowledge of the position of the electron: it has moved a distance of 30
meters with an uncertainty of only 1.643 meters for a relative
uncertainty of 5.5%. This is characteristic of statistical systems; as
the size of the system increases (more collisions in our case) the
precision with which we can calculate results increases.

If we take the ratio of Equations (1) and (2) we obtain a general result:
σ N σ1 1 σ1
ℑrel = = = ,
S N s N s
Thus as
N → ∞ , ℑ rel → 0.
Hence the relative fluctuation decreases as the size of the system
increases. One caveat should be noted here. This limit does not follow
in the case of no driving force as +s, = 0. The relative fluctuation
cannot be defined and therefore the system is not statistically
deterministic.
RANDOM WALK Page 51

3.04 Three-Dimensional Random Walk

We now extend our discussion to the case of random motion in three


dimensions. We allow for the possibility that in a single step the
particle can move in any direction and will undergo a (vector)
displacement s. This displacement can be resolved into its rectangular
components, s = sx ^
i + sy ^
j + sz k
^. To find the average displacement for
a single step we must average (integrate) over all directions:

∫∫∫
G
s = ( sx i + sy j + sz k ) P( x , y , z) dxdydz,

where P(x,y,z) is the probability of moving in a direction x,y,z. If the


steps are truly random and can occur in any direction then the
probability of moving in the x direction is independent of both the y
direction and the z direction and similarly the y direction is
independent of the z direction. This means that the directions are
statistically independent and as we saw earlier on that means we can
write the probability as a product of individual probability functions,
P(x,y,z) = Px(x) Py(y) Pz(z). Each of these probability functions is
normalized over its particular coordinate,

G
∫ P ( x)dx = 1, etc.
x

Then the integral for s can be written as,

=
∫∫∫ ( sx i + sy j + sz k ) Px ( x ) Py ( y ) Pz ( z ) dxdydz,

∫ ∫ ∫
G
s = sx iPx ( x )dx Py ( y )dy Pz ( z)dz

+
∫ ∫ ∫
s y jPy ( y )dy Px ( x )dx Pz ( z)dz

+
∫ ∫ ∫
 ( z )dz
sz kPz Py ( y )dy Px ( x )dx ,

∫ ∫ ∫
G  ( z )dz.
s = sx iPx ( x )dx + sy jPy ( y )dy + sz kPz

In general there will be only one preferred direction for a given driving
force. Let us assume that this force is along the z-axis. Then the
average displacement along the x and y-axes must be zero. Thus the
average displacement for a single step is,


G
s = sz k Pz ( z ) dz = sz k.

In three dimensions the standard deviation for single steps takes the
form,
RANDOM WALK Page 52

∫∫∫ ( sG − sG )
2
σ 12 = P( x , y , z ) dxdydz,

or,

σ 12 =
∫∫∫ [( x − x ) 2 + ( y − y ) 2 + ( z − z ) 2 ] P( x , y , z) dxdydz,

which is an integration over all coordinates, x,y,z. As before, the


x,y,z coordinates are completely independent of each other so we can
factor the probability function into individual ones for each coordinate:

σ 12 =
∫∫∫ [( x − x ) 2 + ( y − y ) 2 + ( z − z ) 2 ] Px ( x ) Py ( y ) Pz ( z ) dxdydz.

Separating this into three integrals, and then using the fact that each
of the probability functions is normalized over its own coordinate
yields,

σ 12 =
∫ ( x − x ) 2 Px ( x ) dx +
∫ (y − y ) 2 Py ( y ) dy +
∫ ( z − z ) 2 Pz ( z ) dz.

Each of these integrals is the square of the standard deviation along


each of the respective axes. So we can write,

σ 12 = σ 12,x + σ 12, y + σ 12,z


Because the collisions are random in all directions we expect each of the
terms in this expression to be identical. Thus we have,

σ 12 = 3 σ 12,z ,
σ 1 = 3 σ 1,z .
Then we have,

For Single Steps:


(Three Dimensions)
G
s = sz k , (3a)

σ 1 = 3 σ 1,z . (3b)

In order to find the average and standard deviation for a total trip of
N steps we need to perform averaging of the kind that we did in Section
3.01 above. The only difference now is that we have vector quantities
instead of the single coordinate. Because vectors are sums of their
components the results of this averaging process yields a vector for SAV
which is the sum of the averages of the individual one dimensional
components. Therefore we have:
RANDOM WALK Page 53

G
S = N sx i + N s y j + N sz k.
Since the driving force only exists along one axis (which we have chosen
to be the z-axis), this reduces to,

Average Total Displacement:


(Three Dimensions)
G
S = N sz k.
(4)

The corresponding proof for the standard deviation follows likewise to


give,

Total Standard Deviation:


(Three Dimensions)

(5)
σ= N σ1 = 3 N σ 1,z .

3.05 Additional Problems

1. In the absence of an external electric field the electrons in a certain


metal under go 3 x 1012 collisions per second. As the electrons undergo
these collisions, 68% of the time the distance traveled in a single
collision is within 5 Angstroms. a) What is the average distance
traveled by a single electron in one minute? b) What is the maximum
distance traveled by 68% of the electrons in one minute? c) What is the
maximum distance that any one electron could travel in one minute? d)
Show in detail how your answers to a) and b) would change if there were
an electric field present which would give each electron an average
displacement of 0.3 Angstroms for each collision.

[Answers: a) 0, b) 0.0671 meters, c) 9 x 104 meters, and d) 5400 meters,


5400 _
+ 0.671 meters.]

2. An ammonia bottle is opened briefly in the center of the room releasing


many ammonia molecules into the air. These molecules go at most 10-5
meters in either direction between collisions with air (and other
ammonia) molecules 68% of the time, and collide on the average 107 times
per second. After each collision they are equally likely to go in either
direction. a) What is the average distance traveled in one step
(collision)? b) What is the standard deviation about the average in a)?
c) What is the average distance traveled after 15 seconds? d) What is the
standard deviation about the average in c)? e) If you were 6 meters from
the bottle, how long would it take before 32% of the ammonia molecules
were farther from the bottle than you?

[Answers: a) 0, b) 10-5 meters, c) 0, d) 0.123 meters, and e) 10 hours.]


RANDOM WALK Page 54

3. A hole is migrating through a semiconductor that has no external field,


so the hole is equally likely to go in either direction after collision
with a lattice site. Such a collision occurs roughly 1013 times per
second, and the hole goes an average of 10-8 cm between collisions. a)
What is the average displacement in one dimension for any one step? b)
What is the standard deviation about this value? c) What would be the
average displacement in this dimension after 10 µs? d) What would be
the standard deviation about the value in part c)?

[Answers: a) 0, b) 10-8 cm, c) 0, and d) 10-4 cm.]

4. Ammonia is a rather strong smelling substance. If a small vial is opened


in a room it will not be too long before you can smell it. Suppose a vial
is opened two meters away from you and that you begin to smell it after
2 minutes. You may assume that ammonia molecules undergo the same
collision rate as the nitrogen molecules in the atmosphere, and that the
standard deviation for single collisions is 10-5 meters. You may also
assume that your nose can detect the molecules when 32% of them are at
a distance greater than your distance from the vial. a) What is the
collision rate for the nitrogen molecules in the room? b) What is the
average distance traveled by an ammonia molecule after two minutes? c)
What is the standard deviation in the distance traveled after 10 minutes?

[Answers: a) 3.33 x 108 seconds-1, b) 0, and c) 4.472 meters.]

5. A sample of copper metal is 10 cm long. In a metal such as copper the


free electrons undergo 1012 collisions per second. For single steps 68%
of the electrons go a distance of 0.1 nanometers. In the presence of an
electric field the electrons undergo an average single step of 0.0001
nanometers. a) On the average how long would it take a single electron
to pass through this piece of metal? b) What is the fluctuation in the
distance traveled in this time? c) On the basis of your answers to a)
and b), how accurate is your prediction that the electron traveled 10 cm
in that time. Express your answer as a percent.

[Answers: a) 1 second, b) 10-4 meters, and c) 0.1%.]

6. Holes in semiconductors typically undergo collisions at the rate of 1013


per second. The root mean square distance they travel is 2 x 10-10
meters. In the presence of an electric field they travel an average
distance of 10-14 meters in the direction of the electric field for single
steps. Answer the following questions assuming a time period of five
minutes. a) What is the average total distance traveled in the
direction of the electric field? b) What is the standard deviation
about the distance in part a)? c) What is the expected range of
distances actually traveled in five minutes? d) What is the drift
velocity of the holes? e) What is the root mean square speed of the
holes?

[Answers: a) 30 meters, b) 0.011 meters, c) 30.000 _


+ 0.011 meters, d)
0.1 meters/second, and e) 2 x 103 meters/second.]

7. An electron is migrating through a semiconductor that has a small


electric field applied. The strength of the electric field is such as
to cause the electron to go an average distance of 10-2 nanometers for
each single step between collisions. Such a collision occurs roughly 1013
RANDOM WALK Page 55

times per second, and the electron goes a root-mean-square distance of


10-8 cm between collisions. a) What is the average displacement along
the field direction for any one step? b) What is the standard deviation
about this value? c) What would be the average total displacement in the
direction of the field after ten seconds? d) What would be the standard
deviation about the value in part c)? e) If you were asked to express
the accuracy of predicting how far the electron moved in ten seconds what
would you say?

[Answers: a) 10-2 nm, b) 10-8 cm, c) 103 meters, d) 10-3 meters, and e)
1000.000 _
+ 0.001.]

8. In the presence of an electric field the electrons in a metal go an


average distance of 10-3 angstroms, with a fluctuation of 3 angstroms.
The collision rate between electrons is 1014/minute. a) How far does an
electron travel on the average after one hour? b) What is the fluctuation
in your answer to a)? c) Recalculate your answers for a) and b) for the
case when the electric field is turned off.

[Answers: a) 600 meters, b) 0.023 meters, and c) 0, 0.023 meters.]

9. My love wears a particularly exotic perfume which I can detect from great
distances. She is sitting on the beach 100 meters north of me. A gentle
wing is blowing out of the north so that on the average each perfume
molecule travels a distance of 4x10-6 cm between collisions. The
molecules in the atmosphere undergo a rather standard fluctuation of 10-5
m per collision. The typical rate of collisions between air molecules
is 107/sec. a) On the average how long does it take for her perfume to
reach my nose? b) What is the fluctuation in the answer to a)? c) What
is the relative fluctuation in the answer to a)? d) What is the relative
fluctuation for single collisions? e) What is the wind speed (on the
average)?

[Answers: a) 4.167 minutes, b) 0.5 meters, c) 0.5%, d) 250, and e)


0.8948 miles/hour.]

10. Energy produced at the center of the sun has a hard time finding its way
out. In this problem you are going to calculate how long it takes as
average photon to get out. On the average in the sun, a photon will go
about 1 centimeter between collisions with hydrogen atoms, and on the
average it is held about 10-8 seconds by the hydrogen atom before being
re-emitted in a completely random direction. Hence, the photon takes 108
"steps" per second. a) what is the average distance traveled in any one
dimension per step? b) What is the standard deviation about the value
obtained in part a)? c) The radius of the sun is about 7 x 108 meters.
About how many "steps" must a photon take before having a 32% chance of
being outside the sun? d) At the rate of 108 steps per second, how many
years does the number of steps in part c) correspond to?

[Answers: a) 0, b) 0.66 cm, c) 4.9 x 1021, and d) 1.55 x 106 years.]

11. A sample of metal is 2 cm in length. The conduction electrons in this


metal undergo 1.5 x 1012 collisions per second. The root mean square
displacement between collisions is 0.2 nm. When an electric field is
applied the average displacement along the direction of the field is 30
pm. a) How many collisions would an electron undergo in traversing the
RANDOM WALK Page 56

entire length of this sample? b) How long would it take an electron to


traverse the sample? c) After traveling the number of collisions obtained
in part a), what is the fluctuation in the distance traveled?

[Answers: a) 6.66 x 108, b) 4.44 x 10-4 seconds, and c)5.16 x 10-6.]

12. In a field free region, the electrons in a certain metal undergo 2 x 1011
collisions per second. As the electrons undergo these collisions, 68%
of the time the distance traveled in a single collision is 1 angstrom.
a) What is the average distance traveled in one hour by an electron?
b) What is the maximum distance traveled by 68% of the electrons in one
hour? c) What is the maximum distance traveled by any of the electrons
in one hour?

[Answers: a) 0, b) 2.68 x 10-3 meters, and c) 7.2 x 104.]

13. A slightly inebriated man can take steps with a standard deviation of 0.3
meters. The sidewalk the man is walking on has a small slope so that his
average single step size is 0.03 meters. The man takes an average of
five steps per minute and has a half kilometer walk home downhill
(fortunately). a) How long does it take the man to get home? b) What
is the standard deviation for his total trip home. c) What is the
relative fluctuation in the trip.

[Answers: a) 2.315 days, b) 38.73 meters, and c) 0.077.]

14. A perfume bottle is opened in the center of the room releasing many
perfume molecules into the air. These molecules go at most 4 x 10-5
meters in either direction between collisions with air molecules 95% of
the time [95% means two standard deviations], and collide on the average
108 times per second. Answer the following questions assuming that this
is a one-dimensional problem. a) What is the average distance traveled
in one step (collision)? b) What is the standard deviation about the
average in a)? c) What is the average distance traveled after 1 minute?
d) What is the standard deviation about the average in c)?

[Answers: a) 0, b) 2 x 10-5 meters, c) 0, and d) 1.5492 meters.]

15. The electrons in a certain metal travel an average distance of 3 x 10-3


angstroms between collisions in the presence of an electric field, and
the fluctuation about this average is 0.5 angstroms. The (drift)
velocity with this field applied is 0.2 mm/sec. The length of the sample
of metal is 11 cm. a) How many collisions does a typical electron in this
metal undergo per second? b) In the time it takes an average electron to
pass entirely through this sample what is the fluctuation in the distance
traveled? c) In the time it takes an average electron to pass entirely
through this sample what is the relative fluctuation in the distance
traveled? d) What is the relative fluctuation in the distance traveled
between single collisions?

[Answers: a) 6.667 x 108, b) 3.028 x 10-5 meters, c) 2.652 x 10-4, and d)


166.67.]

16. In the presence of an electric field the electrons in a metal go an


average distance of 10-3 angstroms, with a fluctuation of 3 angstroms.
The collision rate between electrons is 1014/minute. a) How far does an
RANDOM WALK Page 57

electron travel on the average after one hour? b) What is the


fluctuation in your answer to a)? c) Recalculate your answers for a) and
b) for the case when the electric field is turned off.

[Answers: a) 600 meters, b) 0.023 meters, and c) 0 meters, and 0.023


meters.]

17. In the interior of stars light is continually absorbed and re-emitted,


in an arbitrary direction, by the hydrogen atoms. In a dense star the
light will go about one millimeter before being absorbed. The re-
emission process takes about 10-9 seconds for a collision rate of about
109 collisions/second. a) What is the average distance traveled by a
photon (particle of light) in a single step? b) What is the standard
deviation in your answer to a)? c) If the radius of the star is 4000 km,
how long will it be before a photon has a 32% chance of being outside the
star?

[Answers: a) 0 meters, b) 1 millimeter, and c) 507 years.]

18. The drift speed of the electrons in copper wire is 3.7 x 10-4 meters/sec
for a given applied electric field. The number of electron collisions
occurring per second is 5 x 1023. The fluctuation in the size of the
single steps is 10-8 meters. a) How far does an electron go in 1000
seconds? b) What is the average size of a single step (distance between
collisions)? c) What is the fluctuation in the distance after 1000
seconds? d) To what accuracy (%) do we know how far the electrons travel
in 1000 seconds? e) After 1,000,000 seconds what is the accuracy in the
distance traveled? f) How are your answers to part d) and e) related to
the question of statistical determinism?

[Answers: a) 0.37 meters, b) 7.4 x 10-28 meters, c) 2.236 x 105 meters, d)


6.04 x 107 %, e) 7.07 x 106 %, and f) As the number of collisions
increases the accuracy improves, and in the limit of an infinite number
of collisions the accuracy would be zero.]

19. Electrons in copper have a typical range of 3 x 10-6 cm between collisions


in a (typical) time of 2 x 10-14 seconds for the case of no applied
electric field. a) What is the average speed of the electrons between
collisions in copper? b) How many collisions are occurring per second in
copper? In the presence of an electric field of 1 volt/cm, the drift
speed of the electrons is measured to be 30 cm/sec. c) What is the
average distance traveled in one step in the presence of the electric
field? d) What is the fluctuation in the distance traveled after one
second and what is the relative uncertainty? e) What is the fluctuation
in the distance traveled after one hour and what is the relative
uncertainty? f) What is the fluctuation in the distance traveled after
one day and what is the relative uncertainty? g) If the electric field
were doubled to 2 volts/cm, which basic quantities would change in this
problem? h) If the range were redefined to be that distance covered in
95% of the collisions, which basic quantities would change in the
problem?

[Answers: a) 1.5 x 106 m/sec, b) 5 x 1013 /sec, c) 6 x 10-15 m, d) 0.212 m,


0.70711, e) 12.728 m, 0.01179, f) 62.354 m, 0.00241, g) vdrift, savg, and h)
6 x 10-8 m.]
RANDOM WALK Page 58

20. I'm capable of taking steps of a least one meter in size 68% of the time.
On a particularly day I leave class knowing (because my stomach tells me
so) that I want to go to lunch at Mrs. B's which is approximately 500
meters away. But in my indecisiveness (my head keeps trying to develop
new problems for my upcoming Physics 406 exam that distracts me from my
walking forward all the time) I only average 30 cm per step. Yet I still
step twice every second (a basic muscular response). Clever students
plot my path along the way. a) What is the relative uncertainty in my
single steps. b) How long will it take for me to get to Mrs. B's on the
average? c) What is my average speed for the trip? d) What is the
certainty and relative uncertainty in my trip to Mrs. B's? e) How far
would I have to walk to get my relative uncertainty for a trip under
0.01%?

[Answers: a) 3.33, b) 13.89 min, c) 0.6 m/sec, d) 40.82 m, 0.08, and e)


3.33 x 108 m.]

21. The free electrons in silver metal experience collisions at the rate of
2.5 x 1015/second and 68% of the time, travel a distance of 5.7 x10-8
meters between collisions. a) What is the average speed of a free
electron between collisions assuming there is no electric field present?
b) After one second what is the average distance traveled and what is the
fluctuation in that distance? In the presence of a certain electric field
the electrons acquire an average drift speed of 1.4 x 104 m/sec. c) What
is the average distance for single steps between collisions? d) How many
collisions would an electron experience on the average in going a
distance of one meter? e)How far would an electron have to travel so the
the relative fluctuation in the distance was 0.01%?

[Answers: a) 1.425 x 108 m/sec, b) 2.85 m, c) 5.6 x 10-12 m, d) 1.786 x


1011, and e) 5.802 x 104 m.]

22. An inebriated man (drunk, in mid-westerner's lingo) walks out of a bar


in San Francisco (a very hilly town, unlike Detroit). After taking 1000
steps he is 300 meters (downhill) from the bar. An observant physicist
nearby notes that 68% of the time his steps are within one meter size.
a) What is the average size for his single steps. b) What is the standard
deviation in size of his single steps? c) What is the relative
fluctuation in his single steps? d) What is the standard deviation after
1000 steps? e) What is the relative fluctuation after 1000 steps? f) By
what factor has the relative fluctuation decreased from the one step
value to the 1000 step value? g) How many steps would the man have to
take to reduce his relative fluctuation to 0.01?

[Answers: a) 30 cm, b) 1 m, c) 3.33, d) 31.62 m, e) 0.11, f) 31.62, and


g) 111111.11.]

23. The free electrons in the metal rubidium have a mean free path of 2.2 x
10-8 meters, that is, this is size of the standard deviation for single
steps in any dimension. a) What is the "average" time between
collisions based on the data given so far in this problem? b) How many
degrees of freedom does a single free electron have? c) What is the
average kinetic energy of a single free electron in a metal at room
temperature? d)What is the average speed of a single free electron? e)
What is the average time between collisions based on the speed in part
c)? f) What is the collision rate based on your answer to part a)? g)
RANDOM WALK Page 59

Within what distance range could one expect to find a single electron
after one second? h) What is the average distance traveled by a single
free electron in one second?

Answers: a) 1.88 x 10-13 sec, b) 5.31 x 1012/sec, c) 0.051 m, and d) 0.]

24. Experimental data for sodium metal has the average speed of a free
electron at 1.07 x 106 meters/second, with an average distance (by this
we mean that the distance traveled by a free electron was within 68% of
the following value) at 3.5 x 10-6 meters . a) What is the time between
collisions on the average? b) What is the collision rate, i.e., the
number of collisions per second? c) What is the fluctuation in the
distance traveled after one collision? d) How far has a single electron
traveled after 1,000,000 collisions? e) What is the fluctuation in the
distance traveled after 1,000,000 collisions?

[Answers: 3.27 x 10-12 sec, b) 3.06 x 1011/sec, c) 3.5 x 10-6 m, d) 0 m, and


e) 0.0035 m.]

25. Ice (frozen water, H2O) is a protonic semi-conductor, that is, when an
electric field is applied to ice, the hydrogen ions (protons) carry the
electric current by transferring from one molecule to the next. The
number of protons available for conduction processes depends
significantly on the temperature of the ice which is a characteristic of
semiconductors as opposed to pure conductors whose conduction is
relatively temperature independent. Suppose that for a given electric
field the average drift speed of the protons is 10-5 m/sec, and the rate
of collisions (number of protons transfers per second) is 106 /sec. Also
suppose that the range of single steps is within 10-8 meters 68% of the
time. a) How far does a proton travel on the average in one minute?
b) How many transfers does a proton experience on the average in one
minute? c) What is the average distance traveled in a single transfer?
d) What is relative fluctuation in single steps? e) What is the
relative fluctuation in the total distance traveled after one minute?

[Answers: a) 6 x 10-4 m, b) 6 x 107, c) 1 x 10-11 m, d) 1000, and e) 0.129.]

26. Sulfur Dioxide (S2O) is a particularly evil smelling gas which gives
rotten eggs their putrid odor. Supposed that some sulfur dioxide gas is
released into the atmosphere and mixes with the oxygen and nitrogen
molecules in the air. On the average air molecules experience collisions
at a rate of 107/sec. Assume that the fluctuation in the single steps
traveled by air molecules is 10-4 meters. a) After five minutes what is
the average distance traveled by the sulfur dioxide molecules? b) What
is the standard deviation in your answer to part a)? c) What is the
relative fluctuation in your answer to part b)? d) Repeat the
calculations of a), b), and c) for a time period of five hours. e) If
you were standing ten meters from the source of the gas how long would
it take for 32% of the molecules to be farther from the bottle than you?

[Answers: a) 0, b) 0, c) 5.477 m, d) indeterminate, and e) 0.278 hours.]

27. Air molecules collide on the average 107/second. Suppose that a small
breeze is blowing at 8 meters/second. 68% of the time air molecules
travel a distance of less than 10-5 meters per single step. a) What is
the average time between collisions? b) What is the average distance
RANDOM WALK Page 60

traveled for single steps (collisions)? c) What is the relative


fluctuation in your answer to part b)? d) After ten hours what is the
average distance traveled by a single molecule? e) What is the
fluctuation in your answer to part d)? f) What is the relative
fluctuation in your answer to part d)? g) Compare your answers to part
c) and f). What can you say about the relative precision of the two
measurements? Is this an expected result?

[Answers: a) 10-7 seconds, b) 8 x 10-7 meters, c) 12.5, d) 2.88 x 105


meters, 6 meters, f) 2.083 x 10-5, and g) The relative fluctuation in part
f) is much smaller that in part e). This is what is expected for a
larger system.]

28. When a particular aluminum wire is carrying an electric current of 2.4


amperes the (average) drift speed of a single electron is 24
centimeter/hour. The range of an electron between collisions is 3.3 x
10-9 meters with an average speed of 2.1 x 106 meters/second. a) What is
the average time between collisions and how many collisions does each
electron experience per second? b) How long does it take an electron to
transverse the sample of copper if it is ten centimeters in length? c)
On the average, how many collisions does an electron experience in
traversing the sample? d) What is the fluctuation and relative
fluctuation in the electron’s transit through this sample? Compare these
to the fluctuation and relative fluctuation for single steps.

[Answers: a) 1.571 x 10-15 seconds, 6.364 x 1014/second, b) 1.5 x 103


seconds, c) 9.545 x 1017, and d) 3.224 meters, 32.241, 3.3 x 10-9 meters,
3.15 x 1010.]

29. The heating system in a certain room causes the air molecules to move on
the average with a speed of 2.5 centimeters/second. The range of an air
molecule between collisions is 5x10-4 meters with an average speed of 500
meters/second. a) What is the average time between collisions? b) How
many collisions does an air molecule experience per second? c) After one
hour how far would a molecule travel? d) What is the fluctuation in the
answer to part c)? e) Find the relative fluctuation for single steps and
for the one hour trip.

[Answers: a) 10-6 seconds, b) 106 sec-1, c) 90 meters, d) 30 meters, and


e) 2 x 104, 0.3333.]

30. Compressed air is flowing along a pipeline at a speed of 10


meters/seconds. The molecules within the gas are colliding with
each other at a rate of 3 x 107/seconds, and travel a distance
between collisions of, at most, 5 x 10-6 meters between collisions.
By “at most” I mean that 95% of the time the distance is less than
this amount. a) What is the average distance traveled in a single
step? b) What is the fluctuation in the above average? c) What is
the average distance traveled in one hour? d) What is the
fluctuation in the above average? e) Is the position of a single
molecule statistically determinant after one hour? Explain.
[Answers: a) 3.333 x 10-7 m, b) 2.5 x 10-6 m, c) 3.6 x 104 m, d) 0.822 m
and e) the relative fluctuation is 2.282 x 10-5 so this is approaching
statistical determinism.]

You might also like