Solutions To Jaan Kalda's Problems in Thermodynamics: With Detailed Diagrams and Walkthroughs Edition 1.2.1

Solutions to Jaan Kalda’s Problems in
Thermodynamics
With detailed diagrams and walkthroughs
Edition 1.2.1
Rakshit, Ashmit Dutta, QiLin Xue, Kushal Thaman
Q1 Q2
4
T2
2
Q3
Q4
1 T1
V1 V2 V
Updated September 11, 2020

Kalda Thermodynamics 1
Preface
Jaan Kalda’s handouts are beloved by physics students both in for a quick challenge, to students preparing for
international Olympiads. As of writing, the current thermodynamics handout has 84 unique problems and 20
main ‘ideas’.
This solutions manual came as a pilot project from the online community at artofproblemsolving.com.
Although there were detailed hints provided, full solutions have never been written. The majority of the
solutions seen here were written on a private forum given to those who wanted to participate in making
solutions. In an amazing show of an online collaboration, students from around the world came together to
discuss ideas and methods and created what we see today.
This project would not have been possible without the countless contributions from members of the com-
munity. Online usernames were used for those who did not wish to be named:
Jonathan Qin, Anant Lunia, Ameya Deshmukh and Hermab Podar.
Structure of The Solutions Manual

Each chapter in this solutions manual will be directed towards a section given in Kalda’s mechanics handout.
There are three major chapters: statics, dynamics, and revision problems. If you are stuck on a problem,
cannot make progress even with the hint, and come here for reference, look at only the start of the solution,
then try again. Looking at the entire solution wastes the problem for you and ruins an opportunity for yourself
to improve.
Contact Us
Despite editing, there is almost zero probability that there are no mistakes inside this book. If there are any
mistakes, you want to add a remark, have a unique solution, or know the source of a specific problem, then
please contact us at [email protected]. The most current and updated version can be found on our website
physoly.tech
Please feel free to contact us at the same email if you are confused on a solution. Chances are that many
others will have the same question as you.
1
1 Solutions to Heat and Motion Problems

This section will consist of the solutions to problems from problem 1-8 of the handout. Heat and temperature is
typically the analysis of objects interacting with each other via thermal energy. These problems usually involve
conservation of energy, heat flux, Stefan-Boltzmann’s Law, analysis of graphs, and more.
pr 1. Applying idea 1, we have:

dT
P =C
dt
where
dT aT0
= (1 + a(t − t0 ))−3/4
dt 4
Since we have that,
T (t) = T0 [1 + a(t − t0 )]1/4
dividing by T0 gives us
T
= [1 + a(t − t0 )]1/4 .
T0
We then get
dT
= T04 (a/4)(1/T 3 )
dt
Plugging everything back in, we get:
4P T 3
C=
aT04
pr 2. The ice melts when the temperature of the kettle begins to drop. All the heat that was supplied
to the kettle was used in melting the ice and bringing it up to the temperature of the rest of the water.
Meanwhile the water already present lost some heat to the surroundings, and thus the graph dips at that
point.
The total time for the temperature to recover to T = 75◦ C is approximately t = 37 s. The heating rate
of water is the slope of the graph or dT /dt. This means that the power at T = 70◦ C is P 0 = P tan 75◦ ≈
500 W. Therefore, from energy conservation, we have that
Pt
mL + mc∆T = P t =⇒ m = ≈ 28 g .
L + c∆T
pr 3. Over one complete oscillation of the voltage, the heat lost by the filament must equal the heat
U2
gained by it. Let the resistance of the filament be R. The heat gained by the filament is R1 T2 (because
the voltage is applied only for T2 ). Let the rate at which heat is lost to the surrounding be r. The heat
lost to the surroundings is rT therefore
U12 T U2
rT = =⇒ r = 1 .
R 2 2R
From t = 0.5T to T , the heat lost takes the temperature from the maximum temperature to the minimum
temperature, a change of 2∆T (beware, the ∆T is the amplitude of the temperature while T is time
2
period of voltage oscillations). This implies that
T U 2T
r = 2mc∆T =⇒ ∆T = 1 .
2 8Rmc
ρel `
However, R = A and m = ρ`A, where A is cross-section area of the wire. Substituting these values
gives
U12 T (17)2 (0.01)
∆T = = = 33.8 K
8cρel ρ`2 8(235)(9.95 × 10−7 )(18200)(0.05)2
pr 4. Applying fact 6, we find that the initial heat flux is proportional to ∆T1 = T1 − T0 as the temper-
ature change is minimal (in fact smaller than one Kelvin). Likewise, the final heat flux is proportional
to ∆T2 = T2 − T0 . Therefore, we have that the ratio of powers is
P + Pman ∆T2
=
P ∆T1
Solving for Pman and plugging in our numbers, we find that

T2 − T1
Pman = P = 52.6 W .
T1
P
pr 5. a) The heat flux (or energy flux) density is Φ = S and the thermal resistivity is:
1 dT S dT
ρ= =
Φ dx P dx
Separating variables, we have:
Z d
ρP
∆T = dx = 11.7 K
0 S
b) Again, we separate variables. This time however, our expression becomes:

Z `
∆T S
= ρ(x)dx.
P 0
The integral can be approximated as the area under the curve. In this case, we can see that the average
value is approximately 0.14 K · m/W and then use this value to approximate the integral as a rectangle.
Solving for P from here gives us
P = 1.8 × 10−2 W.
pr 6. According to Stefan-Boltzmann’s Law, the power per unit area emitted from the surface of a
blackbody at temperature T is σT 4 .
The total power emitted from the sun, considered a blackbody for the sake of the problem, is therefore,
P = σT4 (4πR
2
)
Due to the inverse square law, the solar flux stays constant through any closed surface. The portion of
energy that reaches the satellite is given by the ratio between the cross-sectional area of the satellite and
3
the area of an imaginary sphere centered around the Sun with a radius of L (R⊕ is the radius of the
satellite)
2
R⊕ 2

πR⊕
γ= =
4πL2 2L
According to Prevost’s theory of exchange, in order to maintain thermal equilibrium, any object must
emit the same energy as it receives. If this was not true, then it would continuously lose or gain energy
until it is at equilibrium. Now let the emissivity and absorptivity factors of the satellite be and a. Since
we know that these two parameters essentially have the same value at a particular wavelength, we have
Pin = Pout
By Stefan-Boltzmann Law, we have

2
Poutput = 4πR⊕ σT 4
Equating Pin to Pout gives us
Poutput = P γeff
2
2 R⊕
× 4πR⊕ σT 4 =a× σT4 (4πR
2
)
2L
2
2 4 R⊕
× 4πR⊕ σT = a × σT4 (4πR
2
)
2L
2
4 1
T = T4 (R
2
)
2L
T4 R2
T4 = 2
4Ls
R
T = T
2L
Plugging in our given constants gives us ≈ 290 K .
pr 7. We solve the system for the heat shield made up of N thin black plates thermally isolated from
each other in stationary conditions. Of course, since the plates are black and we deal with the system
when stationary conditions have been obtained. Let the ith plate from the left plane have a temperature
Ti , and W0 be the initial heat radiation flux density. Hence, since the flux density remains constant
between the planes, we have
W = σ(Th 4 − T1 4 )
W = σ(T1 4 − T2 4 )
W = σ(T2 4 − T3 4 )
.
.
.
W = σ(TN 4 − Tl 4 )
4
Adding all the equations, we have

W0
(N + 1)W = σ(Th 4 − Tl 4 ) = W0 ⇒ W =
N +1
1 1
which means that the heat flow has reduced by a factor of ξ = . For 2 plates, this is simply ξ = 3
N +1
pr 8. According to Prevost’s theory of exchange, in order to maintain thermal equilibrium, any object
must emit the same energy as it receives. The power recieved is IAeff and the power emmited is σ6s2 T 4 ,
where s is the sidelength of the cube. This then tells us that
r
2 4 4 IAeff
IAeff = σ6s T =⇒ T = .
σ6s2
We now have to analyze the extremas of Aeff .
1. Aeff is minimum when thercube recieves the beam directly on the center of its face. This means
I
that Amin = s2 =⇒ T = .
6σ
2. The maximum is achieved when two space opposite vertices are aligned with the direction of the
beam. This would then mean that the projection is a regular hexagon as shown below.
However, the side length of this hexagon is not s because the edges are not parallel to the projection
plane. However, if you take the three vertices of the hexagon closest to you and connect these points
to make an equilateral triangle, the sides of this triangle will be parallel√to the projection plane. The
sides of the equilateral triangle are face diagonals and have length s√ 2. The area of the hexagon
√ √
is twice the area of the equilateral triangle and therefore has area 23 (s 2)2 = s2 3. Therefore
r √
4 I 3
Tmax =
σ6
Solution 2. Once again, according to Prevost’s theory of exchange, in order to maintain thermal
equilibrium, any object must emit the same energy as it receives. Let the power recieved be αIA and
5
the power emmited is σ6s2 T 4 . We then have that

r
2 4 4 αI
αIA = σ6s T =⇒ T = .
6σ
The projection of area A = A~n ·~i where ~i is unit vector in direction of incoming flux, and ~n is unit vector
in direction perpendicular to plane containing the mentioned area. The projection of area is A = A~n · ~i
where ~i is unit vector in direction of incoming flux, and ~n is unit P vector in direction perpendicular to
plane containing the mentioned area. Therefore, we have that α = ~n ·~i. There can be at most 3 faces
with positive projections of area. Let these 3 faces be in x, y, z directions. Then α = (~nx + ~ny + ~nz ) · ~i
where nx = ny = nz = 1 as these are unit vectors. Let i = ~ix + ~iy + ~iz ). Then α = ix + iy + iz . Since i
is unit vector i2x + i2y + i2z = 1. Also, 0 ≤ ix , iy , iz ≤ 0 as β 2 ≥ 0 for β ∈ R (Note that ix , iy , iz cannot be
all equal to zero). Therefore, i2x ≤ ix , i2y ≤ iy , i2z ≤ iz =⇒ i2x + i2y + i2z ≤ ix + iy + iz . We then have by
power mean inequality,
!1/2
ix + iy + iz i2x + i2y + i2z
≤
3 3
This then means that q
i2x + i2y + i2z ≤ ix + iy + iz ≤ 3(i2x + i2y + i2z )
which then tells us √

1≤α≤ 3.
Therefore, s√ r
4 3I 4 I
Tmax = and Tmin =
6σ 6σ
Credits to https://commons.wikimedia.org/wiki/File:Hexagon_in_cube.svg for the image source.
6
2 Solutions to Gases
This chapter will focus on problems 9-14 of the handout.
pr 9. Let us assume that −dm0 mass sublimes at some instant. (m0 is the total mass of the cup.)
Intially it is moving with v, the velocity of the vessel
q Finally its velocity with respect to the vessel,
3RT q
in the direction of motion of the vessel becomes − √3 = − RT
µ a
µ . Thus, it provides an impulse,
q
−(−dm0 ∆v) = −dm0 RT µ to the vessel
q
−dm0 RT s s
Z Z M
µ RT m RT m
∴v= dv = = ln 1 + ≈
M +m m0 µ M µ M
a
Since the rms speed is randomly directed in x, y, z directions and we only want one speed, we divide by the magnitude
of the unit vectors.
pr 10. During each process for a short time interval ∆t, the number of molecules passing through the
hole will be proportional to:
n∗
N ∼ An∗ ux ∝ √
M
where n∗ is the number density M is the molar mass of the molecule. For the two compounds, U235 F6
1
n∗ 2
and U238 F6 , the ratio of their number densities, n1∗ , will change by a factor of M
M1
2
after each process.
2
N
[U235 F6 ] [U235 F6 ]

M2 2
∴ =
[U238 F6 ] f [U238 F6 ] i M1
Substituting and solving gives us

N
238 + 19 · 6 2
1.4 ≤ 0.7 =⇒ N ≥ 2 · log 352 2 ≈ 161.96 =⇒ N = 162 .
235 + 19 · 6 349
pr 11. Let us consider the time period during which all the molecules complete one round trip. Consider
a cross-section between the plates: all the molecules would have crossed it during this time interval.
(λ L) The net energy transfer through this section is equal to ((K.E.)T1 − (K.E.)T2 )N towards the
colder plate.
L L
∴ ΦA + = ((K.E.)T1 − (K.E.)T2 )N
vx,T1 vx,T2

3R
m √2πM (T2 − T1 ) N
ΦA = q .
M √1 1
L R T
− T√
2 1
Using the fact that mN = mNA nAL = M nAL, we can get
3Rn √ √ p
r
R
∴Φ= √ T 2 + T 1 ( T2 T1 )
2π M
7
√ √ √
Given that, T1 T2 =⇒ T2 + T1 ≈ T1
r
3RnT1 RT2
Φ≈ √ .
2π M
pr 12. We can take s = E, because the only restriction is the distance between two steel balls, the
angle between them can be changed. As there are E edges, number of restrictions is E. By Euler’s
formula,
N = E − F + 2 ⇒ 3N = 3F − 3E + 6
3N − E = 2E − 3F + 6 =⇒ j = 2E − 3F + 6
Every edge is part of exactly 2 faces. Every face contains atleast 3 edges.
∴ 2E ≥ 3F with equality in case of only triangles
∴ j ≥ 6 with equality in case of only triangles

In case of rigid body 6 coordinates are needed to describe it. If the number of degrees of freedom were
larger, it would be needed to use more parameters than in the case of a rigid body to describe its state,
i.e. it cannot be rigid. Therefore, a convex polyhedron is rigid then and only then if all the faces of the
polyhedron are triangles.
pr 13. The total mass of the balloon is M g + mH2 g which is equal to the mass of the air that is holding
the balloon up ρair V g. From the ideal gas law, we can write
pµair
ρair =
RTair
and similarly, we can write the mass of the hydrogen gas mH2 as
pµH2 V
mH2 = .
RTair
We can then write that
pµair pµH2 V
ρair V g = M g + mH2 g =⇒ V g = Mg + g.
RTair RTair
This means that
pV0
M0 = (µair − µH2 ).
RTair
The mass of the balloon when it reaches a volume V1 is then given by
pV1
M1 = (µair − µH2 )
RTair
and by Charle’s law, note that
V1 V0
=
Tair T1
which means that
pV0
M1 = (µair − µH2 ).
RT1
The ballast needed to be thrown out is then

pV0 1 1
∆m = M0 − M1 = (µair − µH2 ) − .
R Tair T1
8
pr 14. Initially, the gas doesn’t take up any volume inside the balloon. In the final state of our system,
all of the gas occupies Vb and therefore we have our change in volume ∆V = Vb . This means that the
work done is W = −p∆Vb = −pVb . By the ideal gas law, we have that pVb = nRT . We then have by the
first law of thermodynamics that
∆U = Q − W
but since the heat flux through the balloon and the heat capacity of the balloon is neglected, then the
heat flown into the system is 0 which means that
∆U = −W =⇒ νCV (T 0 − T ) = nRT.
1
Substituting CV = γ−1 R, we have that
1
(T 0 − T ) = T =⇒ T 0 = γT = 1.4T .
γ−1
where γ = 1.4 for air.
9
3 Solutions to Adiabatic processes
This chapter will focus on problems 15-24 of the handout.
pr 15. Since the wavelength of the universe grows proportionally to the linear size of the Universe,
then the wavelength λ will be two times larger if the size of the universe becomes two times larger. The
wavelength is inversely proportional to temperature by Wien’s Law. In other words,
T0
λ ∝ T −1 =⇒ Tf = .
2
Since we have that
P V γ = T V γ−1 = const.
then, we can substitute values to get (note that V becomes 8 times larger since all three dimensions get
two times larger)
T0 V γ−1 = Tf (8V )γ−1 .
Simplifying gives us
1 4
8γ−1 = 2 =⇒ γ − 1 = =⇒ γ = .
3 3
pr 16. Using Idea 11, we have:

T γ ∝ P γ−1
Therefore, a slight change in temperature would imply:

δT δP
γ = (γ − 1)
T P
The crux of this problem is that the pressure change is so small thus by idea 13, it can be approximated
with ∆p = −ρgh. Therefore, we can isolate for δT to be:

γ − 1 −ρ0 gH0
δT = T0
γ P0
pr 17. From the prelude, we have the work as:

m 2
W = (v − v12 ) + ∆U
2 2
The internal energy changes both in gravitational energy and in internal energy so:
∆U = mg∆h + nCv ∆T
Cv
= mg∆h + (nR∆T )
R
Cv
= mg∆h + ∆(P V )
R
The compression work W is:

W = −∆(P V )
10
so putting everything together gives:

m Cv + R
0 = (v22 − v12 ) + mg∆h + ∆(P V )
2 R
Having CP = CV + R, ∆(P V ) = nR∆T , and letting the molar mass µ ≡ m/n, we can simplify this to:
1 CP (T2 − T1 ) 1
0 = (v22 − v12 ) + g(h2 − h1 ) + =⇒ v 2 + gh + cp T = 0
2 µ 2
pr 18. a) Let the height of the streamline flow far away from the wing to be h0 and the height of the
streamline flow at point P to be hP . Therefore, the continuity expression gives us:
h0
u0 h0 = uP hP =⇒ uP = u0
hP
here we are assuming that the airflow is completely two-dimensional, which means the area is proportional
to the height. Note that we can only use this continuity expression with assumption (d) as the dynamic
pressure is much smaller, which means effects caused by compression can be negligible. We can measure
h0 /hP = 0.77 so we have u0 = 77 m/s. However, this is the speed of the air as measured by the plane.
We can relate these quantities through:
~vair relative ground = ~vair relative plane + ~vplane relative ground
Far away from the plane, ~vair relative ground = 0 (no wind) so then the airplane must be travelling at a
speed of u0 . Therefore, the speed of the air at P relative to the ground is:

h0
u0 − u0 = 23 m/s
hP
11
b) Using definition 10, condensation into water vapor is favored when the dynamic pressure 21 ρv 2 is high.
This occurs when v is at a maximum, or when the streamline lines are close together. From the diagram,
they are closest together at the top left corner of the wing.
c) From problem 17, we find that

1 2
v + cp T = 0.
2
We find from part a that,
! s
1 2 1 2 h20 2cp ∆T
∆v = vcrit 2 − 1 = cp ∆T =⇒ vcrit = .
2 2 hQ h0 /h2Q − 1
2
To finally solve this problem, we have to find ∆T . By relating the ratios, we find that
psa − pw psb − psa psa − pw
= =⇒ T = (Tb − Ta ) − Ta
Ta − T Tb − Ta psb − psa
where pw = rpsa . We note that ∆T = Ta − T , thus by substituting this into our final expression of vcrit ,
we find that s
2cp psa (1 − r)
vcrit = (Tb − Ta ) .
h20 /h2Q − 1 psb − psa
pr 19. Consider a horizontal displacement of a gas. The impulse the gas experiences is:
J = Fnet dt = ∆pSdt = (p1 − p2 )Sdt
where S is the cross sectional area. Take p1 > p2 such that the gas is moving towards the right (which
we arbitrarily set as the positive direction). This gives the change in momentum to be:
∆p = (ρSx2 )v2 − (ρSx1 )v1 = (ρS)v22 dt − (ρSx1 )v12 dt
Setting these two expressions equal gives:
p1 − p2 = ρv22 − ρv12
12
Since this is true for any two intervals, then the quantity
p + ρv 2 = constant
must be preserved.
pr 20. Let us consider a sound wave which propagates in the direction of x-axis; then, the air density
ρ = ρ(x − cs t). Following the idea 16, we consider a frame which moves with speed cs , with coordinate
axis x0 = x − cs t. In that frame, the density perturbation remains constant in time, ρ = ρ(x0 ). This
means that we can use idea 14, so that we obtain two equations:
(ρ + ∆ρ)(v + cs ) = ρcs
1 1
(v + cs )2 + cp T = c2s + cp T
2 2
where v cs is the speed of the gas in the laboratory frame. Note that we also have our momentum
equation to be
P + (ρ0 + ∆ρ)(v + cs )2 + cp T = P0 + ρ0 c2s .
From the problem statement, we have that ∆ρ0 = ρ0 which means that from our first equation we have
cs
2ρ0 (v + cs ) = ρ0 cs =⇒ v = − .
2
Going into our momentum equation, we have that
cs 2 ρ0 c2s
P + 2ρ0 cs − = P0 + ρ0 c2s =⇒ P = P0 + .
2 2
Also, note that by the ideal gas law, we have:
Pµ
P V = nRT =⇒ P µ = ρRT =⇒ T = .
ρR
We can now use these results to go into our second equation (energy) to find that
P0 + ρ0 c2s /2

1 cs 2 1
cs − + cp = c2s + cp T.
2 2 2Rρ 2
We can simplify further to get
c2 µ

1 2 P0 µ 1
c + cp + s = c2s + cp T
8 s 2Rρ0 4R 2
then expanding tells us

c2s cp T0 cp c2s µ 1
+ + = c2s + cp T.
8 2 4R 2
From the problem, we see that cp = 5R/2µ and solving gives us the equation
s s
√

5RT0 5 γRT0
cs = = =⇒ the speed of the wave becomes 2.5 times faster.
2µ 2 µ
13
pr 21. a) Let us assume that the temperature stays roughly constant. This means that the sublimation
rate is also constant and exerts some pressure p on the vapour. We know that the saturation vapour
pressure p0 is defined such that the rate of sublimation = rate of deposition. This means that the pressure
exerted by the sublimation is also p0 . Therefore the force is:
M
p0 A = M a =⇒ a =
p0 A
b) Both evaporation and condensation apply the same pressure at saturation (p0 /2, to be exact), but since
the particles escape never to come back (because λ the length of the vessel), there is no condensation
and thus only half the pressure is applied. Therefore,
p0 p0 A
A = M a =⇒ a = .
2 2M
pr 22. a) First off, we find the pressure at 20◦ C on the graph. At this point, the pressure is approx-
imately given by 2.3 kPa. We are told that the relative humidity is 90% which means that the relative
pressure is given by
2.3 kPa · 0.9 = 2.07 kPa.
The temperature on the graph when it is approximately 2.07 kPa is around 18.5◦ C. This then tells us
that the temperature difference is
20◦ C − 18.5◦ C = 1.5◦ C
b) We are given the equations
Qc = a(T0 − T )
Qe = b[ps (T ) − pa ]
Dividing these two equations through gives us

Qc a T0 − T
=
Qe b ps (T ) − pa
from here, we know imediately that a/b = 65 Pa/K and T0 = 20◦ C. Because r = 0, then pa = 0, and
because r = 0, then Qc = Qe . Therefore, our new equation is
20 − T
1 = 65 =⇒ ps (T ) = 65(20 − T ).
ps (T )
From here, we find the intersection point with this line is (6.5, 0.87), which implies the temperature is
6.5◦ C .
c) In steady state, we have that

a
Qc = Qe =⇒ (T0 − T ) = ps (T ) − pa .
b
Substituting a/b = 65 Pa/K and T0 = 20◦ C and T ≈ 2300r kPa where r is the relative humidity gives
us
65(20 − T ) = ps (T ) − 2300r.
14
• When r = 1, we have the equation the line as
ps = 3600 − 65T.
The intersection of this line with the given curve is at T = 20◦ C and ps = 2300 Pa.
• When r = 0.8, we have the equation of the line as
ps = 3140 − 65T.
The intersection of this line with the given curve is at T = 18.75◦ C and ps = 2000 Pa.
Since p ∝ r and we have the values of pressure at two different values of r, we can find a linear relation
between pressure and relative humidity to get the equation
ps = 1500r + 800 =⇒ 65∆T = 800(1 − r) =⇒ ∆T = 12.3(1 − r) .
d) For the boundary condition, heat dissipated through evaporation. Therefore,
dT
k = b(ps − 2300r) = 800(1 − r) =⇒ Ṫ ∝ 1 − r.
dt
We then see that
Ṫ80 1 − 0.8
= = 4.
Ṫ35 1 − 0.35
pr 23. The water in the reservoir will boil if the saturation pressure exceeds the excess pressure:
ps > patm + ρgh
This threshold is ps > 0.982 MPa, which using the graph corresponds to a boiling point of 183◦ C. Once
the water has reached this temperature, it will start boiling, sending steam up and pushing up the water
in the channel such that the pressure suddenly drops down to 0.1 MPa. This sudden drop in temperature
will cause the superheated water to boil violently until it settles down to 100◦ C at which point cool water
will slowly fill up within the channel.
The sudden drop in temperature of ∆T = 83◦ C released an energy of
E1 = M c∆T
where M is the mass of the entire reservoir. This energy went into turning water into steam, which takes
up an energy:
m c∆T
mλ = M c∆T =⇒ =
M λ
where m is the mass of the water that got turned into steam. Therefore m/M is the ratio we seek, which
we can determine to be m/M = 15.4% .
15
pr 24. a) We are given that

pi ai
ln = + bi
p0 T
Substittuing the values from the table for A we get,
aA
ln 0.284 = + bA
313
and
aA
ln 1.476 =+ bA .
363
Solving these we get aA = −3748.49K and bA = 10.72. The boiling temperature is the temperature at
which the saturated vapour pressure equals the atmospheric pressure, i.e. ppA0 = 1 This gives
aA aA
ln 1 = + bA =⇒ TA = ≈ 350K
TA bA
Similarly solving for B gives,

aB = −5121.64
bB = 13.735
TB ≈ 373K
pA pB
b) Firstly, evaporation will start at the interface. We use fact 15 to conclude that p0 + p0 = 1, at the
time t1 aA aB
+b +b
=⇒ e t1 A + e t1 B − 1 = 0.
Bisection method (or you could just randomly put some values less than 370 K, to zero in on the root) can
be used to find the root of this equation, which gives t1 ≈ 340K = 67◦ The saturated vapour pressures
for the two liquids at his temperature are
pA ≈ 0.734p0
pB ≈ 0.267p0
.
mA mB
Now let mA and mB be the mass of liquid A and B that escape in a bubble. We have ρA = ρB =⇒
mA p A MA mA
mB = pB MB =⇒ mB = 22. Here ρA and ρB are the densities of the vapours of A and B. Thus the rate
at which A evaporates is 22 times that of B. Therefore the temperature starts increasing again when A
is completely evaporated. The amount of B that has evaporated during this time is 100 22 = 4.5g. Thus
at τ1 , there will be no A left, while 95.56g of B will be left. Also, the temperature t2 is the boiling point
of B.
t2 = 373K = 100◦ C
pr 25. We consider the changes in energy along the three interfaces. Consider moving the edge inwards
by a distance a L where L is the length of the boundary. Between the solid and the gas, the initial
contact area is (1 − r)aL and the final contact area is aL so the change in energy is:
∆E1 = σsg rAL
Between the solid and the liquid, the initial contact area is raL and the final contact area is 0, so the
change in energy is:
∆E2 = −σs` rAL
16
Between the liquid and the gas, the initial contact area is aL cos α + (1 − r)aL, and the final contact area
is 0, so the change in energy is:
∆E3 = −σg` aL cos α − σg` (1 − r)aL

dU
At equilibrium da = 0, or:
∆E1 + ∆E2 + ∆E3 = 0 =⇒ r (σsg − σs` ) = σg` cos α + σg` (1 − r)
Letting σsg − σs` = σg` cos α0 gives:
rσg` cos α0 = σg` cos α + σg` (1 − r) =⇒ cos α = r cos α0 − (1 − r).
Letting α0 = 110◦ and r = 0.006 gives: α = 174.9◦ ≈ 175◦ .
pr 26. We have a meniscus as shown in figure, and separate a fraction of water (depicted in grey)
by a fictitious horizontal plane passing the flat bottom of the meniscus, and consider the force balance
for the grey volume. At the separation plane inside the liquid, the hydrostatic pressure equals to the
atmospheric one. Indeed, at the bottom of the meniscus, through the flat water-air interface, there is no
capillary pressure, hence the hydrostatic gauge pressure must be zero; inside the liquid, the hydrostatic
pressure is a function of height only, so the pressure remains equal to the atmospheric one through the
horizontal plane. So, the volume depicted by grey in the figure is surrounded by atmospheric pressure,
i.e. there is no extra net force acting on it due to pressure.
By idea 17, the gauge pressure due to capillary forces across a curved interface is ρgh = σr in cylindrical
geometry where H = V /(πr2 ) gives the radius of the cylinder. We can determine the volume through a
force balance:
ρ∆V g = σ(2πr)
Plugging in our expression for r gives:
r
2σ Vπ
∆V = ≈ 0.59 mL
ρg H
It is believed that the symbolic answer to this problem in the handout is incorrect as it is not dimensionally correct.
pr 27. Consider a circular frustum with radii r1 and r2 . Let θ1 and θ2 be the angles the tangents at
these two circular sections make with the vertical. Therefore, balancing forces in the horizontal direction,
we must have:
σ(2πr1 ) cos θ1 = σ(2πr2 ) cos θ2
Since this applies for any arbitrary r and θ, we can say that:
r cos θ = constant
17
An alternative way of phrasing this is:

d dr dθ dr
r cos θ = 0 =⇒ cos θ − r sin θ = 0 =⇒ = r tan θ
dx dx dx dθ
Separating variables:
Z Z
1
dr = tan θdθ
r
ln(r) = ln(sec θ) + C
r = r0 sec θ
p
r = r0 1 + tan2 θ
s 2
dr
r = r0 1 +
dx
dr
Solving for dx , we can separate variables again and integrating:

−1 r x
x = r0 cosh =⇒ r = r0 cosh
r0 r0
If the separation is L, we get:
R = r0 cosh(L/2r0 ) =⇒ L = 2r0 cosh−1 (R/r0 )
where R is the radius of the ends. Finding the maximum value of L by using a calculator gives us 13.3 cm
Solution 2. We want to minimize the energy of the soap film E = Sγ. This is achieved when the area
S is at a minimum. The area is: Z L/2 p
S= 2πr 1 + r02 dr
−L/2
√
We want to minimize r 1 + r02 which we can take to be our Lagrangian. We can apply the Euler
Lagrange equations, except instead of r being dependent on time, it’s dependent on x. This equivalent
is:

d ∂L ∂L
0
=
dx ∂r ∂r
0

d r p
2πr √ = 2π 1 + r02
dx 1 + r02
which can be simplified to:
1 + r02 = Ar2
where A = 1/r02 . The solution is well known. Motivated by the fact that 1 + sinh2 x = cosh2 x, we can
guess the solution:
r(x) = r0 cosh(x/r0 )
pr 28. (a) We can apply idea 18 by breaking the droplet up into a fictitious surface by imagining a
circular cut at the tip of the needle. The length is πd where d is the diameter so the upwards force is:
18
πσd. This force supports the weight of the droplet mg so we have:

πσd
m=
g
(b) If the droplet is approximated to be a sphere of radius R, then by fact 17 the gauge pressure inside
the droplet will be ∆p = 2σR . The droplet is surrounded by air except for at the tip of the syringe, an
d 2

area of π 2 , where the droplet is in contact with water with gauge pressure ∆p. Thus the net force
2
on the droplet due to pressure from outside is π d2 ∆p in the downwards direction. Then our adjusted
equilibrium equation becomes
2
d
mg + π ∆p = πdσ
2
In order to calculate ∆p we must first find R using our first order approximation from part (a) and our
assumption that the droplet is a sphere:
4 πdσ
ρ πR3 = m =
3 g
1 1
3dσ 3 4ρg 3
R= =⇒ ∆p = 2σ
4ρg 3dσ
Substituting this into (1) we get

2 1
d 4ρg 3
mg = πdσ − π 2σ
2 3dσ
1
πd2 σ(4ρg) 3
= πdσ − 1
2(3dσ) 3
Hence,
1
πdσ πd2 σ(4ρg) 3
g= − 1
g 2g(3dσ) 3
pr 29. If the water was displaced a small distance s from equilibrium (distance |AB|), the surface
tension forces would have to cancel out by the principal of virtual work.
19
This means that

σs + σs0 = 0
where s0 is the distance the water has expanded (distance |AD|) after moving away a distance s from
equilibrium. We express the total energy as
h
E = σS + σ 0 S + mg
2
where S is the surface area, m is the mass of the puddle, and h = V /S is the thickness of the puddle.
We then substitute these relations to get
ρgV 2
E = σS(1 − cos θ) + .
2S
Minimizing this function for E gives us
s
dE ρgV 2 ρgV 2
= σ(1 − cos θ) − = 0 =⇒ S = .
dS 2S 2 2σ(1 − cos θ)
Note that h = V /S, therefore,

r
V 2σ
h= = (1 − cos θ) .
S ρg
Solution 2. We consider equilibrium of the surface defined by the two endpoints B and C in the figure
of the solution before. There will be two surface tension forces directed at both endpoints of magnitude
σx where x is the total length of the surface. Taking forces into horizontal equilibrium gives us
Z Z
σx(1 − cos α) = P x · ds sin α = P xdy
where P is the external pressure. Taking this result into account tells us that
Z h
σx(1 − cos α) = ρgyxdy
0
r
1 2σ
σ(1 − cos α) = ρgy 2 =⇒ y = (1 − cos α)
2 ρg
It is believed that the symbolic answer to this problem is incorrect as the same answer derived here is given in Wikipedia
pr 30. From idea 19 we conclude that the shape of the cross-section of the meniscus is identical to the
cross-section of the pool of liquid laying on the desk (cf. problem 29). This is because on each point of
the surface, we have the surface tension and gauge pressure cancelling out each other to result into the
same type of the surface except inverted as shown in the picture below (the result is also rather intuitive
if you think about it). From fact 17 we conclude that we only have to look at half of the height of the
pool of liquid since in cylindrical geometry the pressure becomes halved.
20
From problem 29, we can conclude that the height is given by

r
2σ
h= (1 − cos θ).
ρg
q
Since the contact angle is 180◦ , we have that cos 180◦ = −1 =⇒ h = 2 ρg
σ
we can then find that the
r
σ
height of the meniscus is half of this or in other words hm = .
ρg
pr 31. Since the potential energy in a sphere is lower, let us assume that the potential energy stored
in the surface is going to completely transferred to kinetic energy instantaneously. Let us determine the
radius of this sphere. Since the volume is equal, we have:
1/3
3d2 h

4 1
2
π(d/2) h = πr3 =⇒ r =
3 2 2
Therefore, the change in energy is:

1/3 !2
3d2 h √ √

1 3 3
p
σ∆S = σπdh − σ4π = σπh h h − 3 9d/4
2 2
Let us assume this change in energy causes half of the liquid to move at a speed of v. Then:
s√
√ √ 3
p
h − 3 9d/4
r
1 2 3 3
p
3
σ
(0.5m)v = σπh h h − 9d/4 =⇒ v = 4 √
3
2 ρd h
using the fact that m = ρπ(d/2)2 h. If we take v to be the average velocity, then the characteristic time
would be given by:
r s √3
h 1 ρd h7
t= = √3
p
v 4 σ h − 3 9d/4
We see that the characteristic time depends on the height h of the original cylinder. Since we want
the time for the most unstable perturbations, we want to maximize t and we can do this by taking the
derivative. Doing so gives us:
3
7 9
h= d ≈ 3.57d
6 4
21
Therefore, the characteristic time is:

r s p
1 ρd 3
(3.57d)7
t= √
3
p ≈ 0.0088 s
4 σ 3.57d − 3 9d/4
Solution 2. Dimensional analysis tells us that the characteristic time is:

r
ρd3
t= = 0.003727 s
σ
pr 32. For the sake of contradiction, let there be 2 reversible engines A and B with different efficiencies
ηA and ηB where ηA > ηB . Although they will be working with the same hot and cold baths, suppose
this different efficiency is achieved through using a different design or using a different gas. We will prove
that this cannot be the case and that the efficiencies must be the same no matter how it is designed.
Let engine A takes a heat QA from the hot bath and delivers a heat qA to the cold bath and engine B
takes a heat QB and delivers a heat qB to the cold bath. Since the Carnot Cycle is reversible, we will
use engine A to drive engine B backwards so that B acts like a heat pump, as depicted below.
Hot Bath
QA QB
ηA QA
A B
qA = (1 − ηA )QA qB
Cold Bath
The efficiency of engine A is:

qA
ηA = 1 −
QA
so the heat delivered to the cold bath is qA = (1 − ηA )QA . Let us assume the leftover work this produces
W = ηA QA goes into driving engine B backwards. Due to conservation of energy:
qB + ηA QA = QB
Coupled with the efficiency equation for engine B we get:

ηA
QB = QA
ηB
Therefore, since ηA > ηB , the hot bath is gaining heat and the cold bath is getting colder. This clearly
violates the second law of thermodynamics which states that for a reversible process ∆S = 0. The only
22
way for this analysis to be true is if our assumption that the efficiencies are different is incorrect and
that QB = QA . Therefore, no matter how the engines are constructed as long as they are reversible,
their efficiencies can only depend on the temperatures of the hot and cold baths.
pr 33. Assuming that idea 2 is correct, it may be proved that between the quantity of heat Q, which
in a cyclical process-of the kind described above is transformed into work (or, where the process is in
the reverse order, generated by work), and the quantity of heat Q2 which is transferred at the same time
from a hotter to a colder body Cor vice versa), there exists a relation independent of the nature of the
variable body which acts as the medium of the transformation and transfer; and thus that, if several
cyclical processes are performed, with the same reservoirs of heat K1 and K2 , but with different variable
bodies, the ratio 3. will be the same for all. If we suppose the processes so arranged, according to their
magnitude, that the quantity of heat Q, which is transformed into work, has in all of them a constant
value, then we have only to consider the magnitude of the quantity of heat Q2 . which is transferred, and
the principle which is to be proved takes the following form:
If where two different variable bodies are used, the quantity of heat Q transformed into work is the same,
then the quantity of heat Q2 which is transferred, will also be the same.
Let there, if possible, be two bodies A and A0 (e.g. the perfect gas and the combined mass of liquid and
vapour, described above) for which the values of Q are equal, but those of the transferred quantities of
heat are different, and let these different values be called Q2 , and Q02 , respectively: Q02 , being the greater
of the two. Now let us in the first place subject the body a to a cyclical process, such that the quantity
of heat Q is transformed into work, and the quantity Q is transferred from K2 to K1 . Next let us subject
A0 to a cyclical process of the reverse description, so that the quantity of heat Q is generated out of
work, and the quantity Q02 . is transferred from K2 to K1 . Then the above two changes, from heat into
work, and work into heat, will cancel each other since we may suppose that when in the first process
the heat Q has been taken from the body K1 and transformed into work, this same work is expended in
the second process in producing the heat Q, which is then returned to the same body K1 . In all other
respects also the bodies will have returned, at the end of the two operations, to their original condition,
with one exception only. The quantity of heat Q02 transferred from K1 , to K2 has been assumed to be
greater than the quantity Q2 transferred from K1 to K2 . Hence, these two do not cancel each other, but
there remains at the end a quantity of heat, represented by the difference ∆Q2 , which has passed over
from K1 to K2 . Hence a passage of heat will have taken place from a colder to a warmer body without
any other compensating change. But this contradicts the fundamental principle. Hence the assumption
that Q02 is greater than Q2 , must be false.
Again, if we make the opposite assumption, that Q02 , is less than Q2 we may suppose the body A0 to
undergo the cyclical process in the first, and a in the reverse direction. We then arrive similarly at the
result that a quantity of heat Q2 − Q02 , has passed from the colder body K2 to the hotter K1 which is
again contrary to the principle.
Since then Q02 , can be neitlier greater nor less than Q2 , it must be equal to Q2 ; which was to be proved.
We will now give to the result thus obtained the mathematical form most convenient for our subsequent
reasoning. Since the quotient Q/Q2 is independent of the nature of the variable body (fact 2), it can
only depend on the temperature of the two bodies K1 and K2 which act as heat reservoirs. The same
will of course be true of the sum
Q Q + Q2 Q1
1+ = = .
Q2 Q2 Q2
This last ratio, which is that between the whole heat received and the heat transferred, we shall select
for further consideration; and shall express the result obtained in this section as follows:
23
The ratio Q1 /Q2 can only depend on the temperatures T1 and T2 .
This leads to the equation:

Q1
= f (T1 , T2 ).
Q2
Since the process is isothermal, we can obtain the equation
V
Q1 nRT1 ln Vfi T1
= V
= .
Q2 f
nRT2 ln V1 T2
Using this definition, the Carnot’s Cycle efficiency can then be rewritten as
Q2 T2
ηC = 1 − =1− .
Q1 T1

Part of this solution is taken from Rudolf Clausius’ original work ”The Mechanical Theory of Heat”.
pr 34. According to Prevost’s theory of exchange, in order to maintain thermal equilibrium, any object
must emit the same energy as it receives. Thus the absorption and emission of light must be the same
at all frequencies.
A consequence of this problem is that radiation and absorption properties of a material must be identical
throughout the entire spectrum. It can be similarly shown that a partially reflecting material must have
equal transmittance from both sides. It may seem that dark window glasses are more transparent when
looking from inside of a darkly lit room, but this is a mere illusion: when looking from outside, a small
fraction of reflected abundant outside light can easily dominate over the transmitted part of the light
coming from inside, but the opposite is not true. The effect can be enhanced by overlaying an absorbing
and reflecting layers and turning the reflecting layer outside. Then, while total transmittance is equal
from both sides, the reflectance from outside is larger because from outside, reflected light does not pass
through the absorbing layer.
pr 35.
Proof. We analyze 4 processes on a Carnot cycle working in phase:

• We have our substance undergo an isothermal expansion at a constant temperature T . The sub-
stance undergoes isothermal expansion since it becomes in contact with a reservoir at a temperature
T . A bit of the mass of the liquid evaporates during this process and as a result, the total volume
that the substance works on changes from V1 to V2 . We can approximate this change in volume by
a straight line.
• We have our substance undergo adiabatic expansion where the temperature and pressure increase
to T + dT and p + dp.
• The substance is compressed isothermally at temperature T + dT when it is put in thermal contact
with a reservoir of temperature T + dT . In this process, the vapour mass condenses and therefore,
the pressure of this process is constant at p + dp.
• The substance is compressed adiabatically and it’s pressure and temperature goes to (p + dp, T +
dT ) → (p, T ).
24
The graph that this cycle creates is a parallelogram with height −dp and edges of lengths m∆V . The
work done by this Carnot cycle is then –m∆V dp. Furthermore, this substance receives heat during the
first process which is of value of mL. We then can write the efficiency η of this thermodynamic process
as:
W −m∆V dp ∆V dp
η= = =− .
Q mL L
Assuming all reversible processes means that this in turn must equal to the efficiency of a Carnot cycle
or
Th T + dT dT
ηCarnot = = =− .
Th − T` dT T
Therefore, equating both of these together gives us
dT ∆V dp dp L λµ
− =− =⇒ = = ps .
T L dT T ∆V RT 2
Solving the differential equation, we have:

p λµ 1 1
ln = −
p0 R Ti Tf
We can assume that ∆T = Tf − Ti T , allowing us to simplify our relationship to:

λµ∆T λµ∆T
P = P0 exp =⇒ ∆P = P0 exp −1
RT 2 RT 2
Using the first order expansion, we get:

P0 λµ∆T
∆P = ≈ 350 Pa.
RT 2
Let’s consider one mole of some material at the phase boundary so that it is equally stable as a liquid
and gas alike. Therefore, its Gibbs free energy must be the same at the phase boundary. Let us denote
dGl by the Gibb’s free energy of the liquid state and ∆Gg by the Gibb’s free energy of the gas. We
require that Gl = Gg . Consider raiseing the temperature by dT and the pressure by dP such that both
phases remain equally stable therefore we require dGl = dGg . Also, since
dG = −SdT + V dp
We have that
−Sl dT + Vl dP = −Sg dT + Vg dP.
Where (V1 , S1 ) is the volume and entropy of the liquid and (V2 , S2 ) is the volume and entropy of the gas.
We then have
dp Sg − Sl
dp(V2 − V1 ) = dT (S2 − S1 ) =⇒ = .
dT Vg − Vl
RT
It is easier to write the difference in entropies as Sg − Sl = λµ/T and V = p from the ideal gas law.
Therefore,
dps λµ
= ps .
dT RT 2
25
Let us expand on entropy S = S(V, T ) as a function of state with the multivariable chain rule:

∂S ∂S
dS = dV + dT.
∂V T ∂T V
∂S

In a closed system, temperature and pressure remain constant and therefore, ∂T V
dT = 0. We are
then left with
∂S
dS = dV.
∂V T
We then use Maxwell’s relations:

∂S ∂p ∂p
= =⇒ dS = dV.
∂V T ∂T V ∂T V
Next, note that by using the multivariable chain rule on P = P (V, T ) we have:

∂p ∂p
dP = dT + dV
∂T V ∂V T

∂p
and ∂V = 0 as a change in phase transition does not change the pressure with constant parameters.
T
Therefore,
∂p dp ∂p
dP = dT =⇒ = .
∂T V dT ∂T V
This means that - after going back to the expression of entropy - we have
dp dp ∆S ∆H
dS = dV =⇒ = = .
dT dT ∆V TV
RT
Note that ∆H = λµ and V = p from the ideal gas law. Therefore,
dps λµ
= ps .
dT RT 2
pr 36. The Clausius-Clapeyron equation for water is:

dps Lµ
= ps .
dT RT 2
Separating variables gives:
1 dps Lµ
= .
ps dT RT 2
Integrating this gives us
Z p s2 Z T2
dps dT ps Lmv 1 1
= Lµ =⇒ ln 2 = − .
p s1 ps T1 RT 2 p s1 R T1 T2
Simplifying this result finally gives us

Lmv 1 1
ps2 = ps1 exp − .
R T1 T2
where ps1 and T1 are the initial points of integration. We are given, ps = p0 e−U/kB T and therefore,
ps2 = p0 e−U/kB T2 , ps1 = p0 e−U/kB T1 .
26
This mean that

U 1 1
ps2 = ps1 exp − .
kB T1 T2
Comparing the 2 equations gives, U = Lµ/NA as kB = R/NA .
To interpret this equation, note that we can rewrite it as the latent heat of evaporation per mole. We
can imagine the gas phase to have an energy of zero and thus the latent heat of evaporation represents
the energy ”stored” inside the bonds in the water molecules. Thus, it represents the work that needs to
be done to evaporate one mole of water
pr 37. First, let us convert the temperatures from celsius to kelvin.
T1 = 20 + 273 = 293 K
T2 = 0 + 273 = 273 K
Furthermore, converting the time from hours to seconds gives us t = 36000 s Note that the heat flux as
given in Kalda’s introduction to thermoelectricity is
P
Φ=
ηC
where ηC = 1 − TThc . For a small increment of time dt the mass gained is dm and therefore the heat flux
is dm
dt λ. Equating these two expressions together gives us
Z m Z t
dm Th Th
λ = I 2R =⇒ dm = I 2R .
dt Th − Tc 0 0 Th − Tc
Integrating through and dividing both sides by λ tells us that
I 2 RtTh
m= ≈ 1.5 g .
(Th − Tc )λ
pr 38. Heat leaving the room in stable state is equal to heat produced by heater by Fourier’s law.
P (T2 ) = k(T2 − T1 )
In the second case,

P (T4 ) = k(T4 − T3 )
Therefore, point (T4 , P (T4 )) must be on the line which passes through point (T3 , 0) with the same slope
as the line passing through (T2 , P (T2 )) and (T1 , 0) This gives T4 = 1.4T3
27
28
4 Solutions to Revision Problems
This section will contain problem 38-86 of the handout. Revision problems take concepts and ideas from earlier
problems and places them in a new context. As a result, many of the problems in this section will seem
familiar. This however, does not mean that all the problems in this section are easy. Some of the hardest
problems originate in this section.
pr 39. At nominal voltage, we have two relations of P = I0 V0 and R = VI00 . As resistivity is proportional
to temperature, we then have the temperature at which the filament is supposed to emit light to be
R T RT0 V0
= =⇒ T = = T0 .
R0 T0 R0 I0 R0
The power P radiated according to Stefan-Boltzmann law is

P V0 I0
= kσT 4 =⇒ P = kσT 4 · πld =⇒ ld = .
A πkσT 4
The initial resistance is given by
ρ0 l ρ0 l πR0 d2
R= = =⇒ l = .
A πd2 /4 4ρ0
Therefore,
r
πR0 3 V 0 I0 3 4V0 I0 ρ0
d = =⇒ d = .
4ρ0 πkσT 4 kπ 2 R0 σT 4
pr 40. The rate at which the temperature rises or falls is proportional to total heat output. Until
t = 1100 min heater is not switched on as temperature decreases continuously. After t = 1100 min power
of heater is added to heat lost i.e.
Ptotal = Pheater + Pheat lost
where Pheat lost is the power lost to surroundings. The total Power is proportional to rate of change in
temperature which is the slope of given graph. For the heater, it is is change in slope at t = 1100 min.
Now applying calculations gives us:

2 2
Ṫheater = Ṫtotal − Ṫheat lost =⇒ Ṫheater = − − = 0.038 ◦ C/min.
69 218.18
Therefore, equilibrium temperature is when power of heat losses is equal to power of heater(in magnitude)
i.e. the temperature at which the the Ṫheat lost = 0.038 ◦ C/min. This is can be found in the graph before
t = 1100 min between T = 20 ◦ C to T = 24 ◦ C.
pr 41. Before tackling this problem, we have to think of what the setup tells us. It may be tempting to
use the adiabatic pV γ = const relation to help us solve this problem but note that there is heat exchange
being done with the surrounding and furthermore, the process is being done at constant pressure so we
cannot take this to be an adiabatic process. This process is instead isobaric since it is being done at
constant pressure. Also, to make our naming shorter, we denote:
29
1. Process I: Helium receives a certain amount of heat, because of which the piston moves up by
d1 = 5 cm.
2. Process II: After waiting for some longer period of time, an additional displacement d2 of the piston
was observed.
Since both processes are isobaric, note that the heat added to the surroundings can be accurately
predicted with the formula
Q1 = Cp n∆T.
In process I, the heat added to the surroundings is given by
Q1 = CpHe nHe ∆T
since only Helium recieves the amount of heat and not Hydrogen.The change in temperature can be
accurately predicted by the ideal gas law:
p∆V pSd1
p∆V = nR∆T =⇒ ∆T = =
nHe R nHe R
where S is the cross-sectional area of the container. Therefore,
pSd1 Cp pSd1
Q1 = CpHe nHe = He .
nR R
In process II, the piston goes to an equilibrium temperature Teq . Both Hydrogen and Helium are changing
volume and temperature to achieve this equilibrium state so therefore, we denote the heat added by
Q2 = (CpHe nHe + CpH2 nH2 )∆Teq .
By ideal gas law, we can write
pV0 = nH RT
p(3V0 ) = nHe RT
This implies that nHe = 3nH2 . Furthermore, once again by ideal gas law note that
p∆V pS(d1 + d2 )
p∆V = (nHe + nH )R∆Teq =⇒ ∆Teq = = 4 .
(nHe + nH )R 3 nHe R
Next, note that the heat added in process I must be the same as the heat added in process II by the
second law of thermodynamics. Therefore,
CpHe pSd1
= (CpHe nHe + CpH2 nH2 )∆Teq .
R
Substituting the relations we found from the ideal gas law tells us that
CpHe pSd1 1 pS(d1 + d2 )
= nHe (CpHe + CpH2 ) 4 .
R 3 3 nHe R
Simplifying gives us the expression

4 1
CpHe d1 = (CpHe + CpH2 )(d1 + d2 ).
3 3
Substituting CpH2 = 7/2R and CpHe = 5/2R tells us that
4
3 CpHe − CpHe − 13 CpH2 45 5 17
3 2R − 2R − 3 2R 1
d2 = 1 d1 = 5 17 d1 =− d1 .
CpHe + 3 Cp H 2 2R + 3 2R
11
1
This tells us that the piston moves down by 11 d1 = 0.45 cm.
30
pr 42. (a)
Proof. We note that the total change in energy is given by
∆U = nCV ∆T
= CV n1 (T − T1 ) + CV n2 (T − T2 )
By the ideal gas law, pV = nRT , we have that

CV p0 (V − V1 − V2 )
∆U =
R
we also see that internal energy change must be equal to the work produced by external pressure.
Therefore,
CV p0 (V − V1 − V2 )
= p0 (V − V1 − V2 ).
R
Since CV /R is non-zero we find that
V − V1 − V2 = 0 =⇒ V = V1 + V2 .
(b) We note that by the ideal gas law

pV
pV = nRT =⇒ n = .
RT
This tells us that the moles of gas in our scenario is given as

p0 V1 V2 p0 V1 + V2
n= + =
R T1 T2 R T0
therefore, we find that
V1 + V2
T0 = ≈ 16.5◦ C
V1 /T1 + V2 /T2
(c) Note that relative humidity is denoted as the mass of water vapor divided by the mass of water
vapor in saturation. In this problem, we have to calculate the relative humidity of the mixed air. First,
note that y-axis of the graph that is given in part (c) represents density. We have to convert our mass
relation to density which is fairly easy as the density is directly proportional to mass which means that
r = ρρvsv .
Next, we need to determine these two densities. First, note that there were two individual mixtures of
air, one, inside the room at temperature t1 = 25◦ C and the other outside of the room at temperature
t2 = 1◦ C. We mark these two points on our graph as point 1 and 2 respectively. We then connect these
two points with a line. The temperature of the mixed air as shown in part (b) is 16.5◦ C. The density
of the water vapor while be the intersection of the line x = 16.5 and the line connecting the two points
1 and 2. By finding the intersection of these two points, we find that ρv is at point 3 with a density of
1.65 g/m3 . The mass of the water vapor in saturation is represented by the intersection of the graph and
the line x = 16.5 (this is because the graph shows the dependence of saturated vapor density of water
as a function of temperature). This gives us the point 4 with a density of ρvs = 1.37 g/cm3 . We now
finally obtain the relative humidity to be
ρv 1.65 g/cm3
r= = = 1.20 = 120%.
ρvs 1.37 g/cm3
31
(d) The mass will have to undergo a small temperature change to condensate. This means that it will
have to transfer heat to its surroundings and with that information, we can attempt to balance the
heat going out into the system and the heat going into the system. The heat going out of the system
moves to the surrounding and will go to the outside air that has a given density ρ0 . With a temperature
change denoted by ∆t, we write Qout = mc∆T = ρ0 V c∆t. The heat going into the surroundings changes
the temperature of the mixed air. If t0 = 16.5◦ C is the initial temperature of the mixed air, we write
Q = q(ρv (t0 − ρvs (t0 + ∆t)). We conserve heat to get the equation
ρ0 c∆t = q(ρv (t0 − ρvs (t0 + ∆t)).
We then find by rearranging variables that

ρ0 cp ∆t
ρvs (t0 + ∆t) = ρv (t0 ) − .
q
Let us denote T = t0 + ∆t. This means that ∆t = T − t0 . Our equation then looks like
ρvs (T ) = 1.65 − 0.47(T − 16.5).
This is an equation of a line, and the intersection of this line with the graph of the saturated vapor
density gives us the new condensated mass. As shown in the picture below, we have our new point 5 to
be the density of the new mass. Remember that we have to find the new mass of the condensated mass
which will be given by m = ∆ρ(V1 + V2 ). We see that ∆ρ ≈ 0.25 g/m3 from the graph so m ≈ 7.5 g.
32
pr 43. (a) Note that the total force on a cross-sectional area of the cylinder is given by
mg
F = p0 πr2 + mg =⇒ p = p0 + .
πr2
By ideal gas law we note that
p∆V p0 (πr2 ∆s) (p0 πr2 + mg)∆s

p∆V = nR∆T =⇒ ∆T = = = .
nR nR nR
The total temperature after collision will then be given as
(p0 πr2 + mg)∆s

T = T0 + ∆T = T0 + .
nR
(b) Note that work is given by p∆V . The change in volume is πr2 ∆S so all in all, we have W =
(p0 πr2 + mg)∆s.
(c) The process is isobaric since it occurs at a constant pressure which means that the energy is given
by the heat converted to the system or Q = cp n∆T . We know from part (a) that
(p0 πr2 + mg)∆s

∆T =
nR
which means that
(p0 πr2 + mg)∆s cV + R (p0 πr2 + mg)∆s c
V

Q = cp n = n = ∆(p0 πr2 + mg) +1 .
nR R n R
33
dQ
(d) The power is given by P = dt which means that for a unit time ∆t it is
c ∆s
V
P = +1 (p0 πr2 + mg).
R ∆t
The number of molecules for a unit time is given by power over energy (this is justified as it is dimen-
sionally correct). The energy of a single photon will be Ep = hν = hc
λ which means
P λ
ṅ = =P .
Ep hc
W
(e) The efficiency of the process is given by η = Q. The change in mechanical energy of the gas mg∆s
serves as the change in work so therefore,
mg∆s 1
η= cV

2
= 2
.
1+ R (p0 πr + mg)∆s p0 πr
+ 1 1+ cV
mg R
(f ) When the cylinder is rotated, the gas undergoes adiabatic expansion to a new pressure p0 due to the
lack of pressure from external surroundings. Noting that T γ P 1−γ = const the new temperature of the
gas is
γ−1
p0 γ
Tf = T0 .
p
pr 44. The tire does not exchange any heat with its surroundings which means that it undergoes an
adiabatic expansion/compression. Since we are relating two state variables p and T , we use the relation
p1−γ T γ = const.
Substituting both final and initial variables shows us

γ−1 γ−1
pf γ p0 + p1 γ γ−1
p1−γ
0 T0
γ
= p1−γ
f Tfγ =⇒ Tf = T0 = T0 =3 γ T0 ≈ 400 K.
p0 p0
pr 45. (a) Let us analyze each step in the cycle:
I. This is an isobaric process (or in other words, constant pressure). This is because no thermodynamic
parameter is changing in the piston apart from the change in volume from 5 → 1. Fresh air comes into
the valve and fills the cylinder.
II. For process 1 → 2, we have a reversible adiabatic process. Both parameters of pressure and volume
are changing and the air is being compressed adiabatically.
III. For process 2 → 3, we have an isothermal process. This is because the combustion spark into the
valve is only for a very short time so we essentially have a constant volume. We then have heat absorbed
from a series of reservoirs of temperatures T2 and T3 .
34
IV. Now the gas is disposed of and the pressure and volume again change adiabatically from 3 → 4.
This process then repeats once again periodically through an isothermal process of 4 → 1.
We can now draw the thermodynamic cycle.

P
Q1
2
4
Q2
1
5
V2 = V3 V1 = V4 V
(b) The heat out of the isotherm on 2 → 3 is given by
Q1 = CV (T3 − T2 )
while the heat out of the isotherm on 4 → 1 is given by
Q2 = CV (T3 − T1 ).
Since efficiency is given by

W Q1
η= =1−
Q2 Q2
we find that
T4 − T1
η =1− .
T3 − T2
On an adiabat, T V γ−1 is constant or in other words,
T1 V1γ−1 = T2 V2γ−1 , T4 V1γ−1 = T3 V2γ−1
which implies that

γ−1
T4 − T1 V2
(T4 − T1 )V1γ−1 = (T3 − T2 )V2γ−1 =⇒ = .
T3 − T2 V1
Hence:
1 1−γ
η =1− γ−1 = 1 − k .
(V2 /V1 )
Remark: This problem is essentially resemblant of the Otto cycle, a four-stroke cycle in internal com-
bustine engines that is found in cars, electrical generators, etc.
35
pr 46. First, let us figure out how to calculate the energy cost. It is given in the problem that the
electrical energy cost is c = 0.1 EUR/kWh so if we find the amount of energy dissipated, we can then
convert this to monetary value with using the electricity cost constant via dimensional analysis.
Let us make use of idea 4 in Kalda’s handout. If a system is described byP a parameter x (which can be
time, coordinate, velocity, etc.) and a quantity A can be expressed as A = i Fi ∆x, where ∆x is a small
interval of the parameter x, the sum is taken over all the small intervals, and Fi is a function of x. In
this case, our parameter is temperature. We divide the graph into small chunks where the temperature
at each endpoint of a singular chunk is Ti where i ∈ Z+ . For the cooling period, during the number of
days ∆Ni for which the temperature stayed in the (small) range between Ti and Ti + ∆Ti , the heat loss
is Qc = (Ti − T0 )C∆Ni · 3600 s/h. Note that (Ti − T0 )∆Ni is a horizontal narrow rectangular region
between the graph and the vertical line T = T0 . Therefore, the net energy can be represented as
 
X X
Qtot = QA + QB = C  (T0 − Ti )∆Ni + (Ti − T0 )∆Ni  = C(AA + AB )
Ti <T0 Ti >T0
where AA and AB are the respective areas shown in the graph below. Next, we need to make use of
the efficiencies as the total power is not just all we need. Note that when the device is used as an air
conditioner (i.e Ti < T0 ) we use the COP ηc and when the device is used to heat the room (i.e Ti > T0 )
we use the efficiency ηh . Therefore, our effective heating power is given by

QA QB AA AB
Qeff = + =C + .
ηc ηh ηc ηh
By measuring areas, we can approximate the effective energy to be Qeff ≈ 2900 kWh and multiplying by
c gives us the answer of 290 EUR.
pr 47. (a) The power dissipated by the wire is
U02 U 2S
P = = 0 .
R ρel l
36
By idea 1, we can write

dQ dT dT
P ≡ =C = mc
dT dt dt
which means that
U 2S
Z Z
dT
mc = 0 =⇒ mcρel ldT = U02 Sdt.
dt ρel l
By integrating and solving this equation for t, we find
U02 At ρ20 ρel cl2 (T1 − T0 )

= Alρ20 c(T1 − T2 ) =⇒ t = = 25 ms.
ρel l U02
(b) As U1 is large, ρel changes significantly with temperature. Applying the same initial equations as in
part (a), we find that
ρ20 cl2
Z Z
mcρel ldT = U12 Sdt =⇒ t = × (Area under ρel − T graph).
U12
The area under the graph is approximately 11.6+0.4

2 × 3400 · 10−7 and therefore t ≈ 1.6 ms.
pr 48. (a) The wording of the temperatures T1 , T2 , and T0 may be slightly confusing so let us clarify
this before solving the problem. T2 is the temperature of the cold air going into the room while T0 is the
warm air from the room and T1 is the cold air that is already inside the room.
Let us assume that there is a constant difference of temperature across the opposite sides of the plates
given by ∆T ≡ T0 − T2 . By fact 6, we note that for small tempera difference ∆T ≡ T0 − T2 , the heat
flux is proportional to ∆T . In other words,
Q̇ ∝ T0 − T2 = const.
The heat flux is also equal to

Q̇ = mcp Ṫ = ρV cp Ṫ = ρshcp Ṫ
where s is the cross sectional area for an air element of volume V . We remember that the thermal
conductance of the metal is σ (the heat flux through a unit area of the plate per unit time, assuming
that the temperature drops by one degree per unit thickness of the plate). This means that we can write
σs(T0 − T2 )
Q̇ = .
d
Since the heat flux is proportional to the difference in temeperature which is constant, this means that
the temperature gradient is linear with respect to position. If the velocity of the air is v, we write with
dimensional arguments that
v(T2 − T1 )
Ṫ =
x
we write ∆T ≡ T2 − T1 here since we are looking at the temperature difference horizontally from the cold
air going into the room and the cold air that is already in the room. Substitituting Ṫ into our initial
expression of Q̇ and equating that to our other expression with thermal conductance, we result in the
equation
v(T2 − T1 ) σs(T0 − T2 )
ρshcp = .
x d
37
To solve this equation for T2 , we can cross multiply to get
ρshcp (T2 − T1 ) = σsx(T0 − T2 ) =⇒ T2 (ρhcp v + σx) = ρshcp dT1 + σsxT0 .
Dividing over gives

xσT0 + ρhcp dT1
T2 = .
xσ + ρhcp vd
(b) Since the tempera difference is very large, the temperature gradient is not linear by fact 6. This is
because (a) heat conductivity of the materials may depend on the temperature, (b) the heat flux due to
heat radiation is a non-linear function of T1 and T2 (however, it can be still linearized for small values
of ∆T ); (c) large temperature differences may cause convection of air and fluids which will enhance heat
flux in a nonlinear way. Therefore, we have to rely on the graph to carry out calculations. Note that by
idea 1:
dQ dT
P ≡ =C .
dT dt
By integrating, we find that
Z t Z T1
C
dt = dT =⇒ t = 12C = 120 s.
0 T2 P
cp
pr 49. (a) The adiabatic index is defined as the ratio of cp and cV so that γ = cV . Note that
cp = cv + R =⇒ cv = cp − R
which means that upon substitution

cp γ
γ= =⇒ γ(cp − R) = cp =⇒ cp = R.
cp − R γ−1
m
(b) Remember that from the ideal gas law, we have that pV = nRT where n = M. Therefore,
m m p0 M
p0 V = RT =⇒ ρ = = .
M V RT
(c) We can conserve momentum of a cross-sectional area of air S for a small interval of time dt assuming
that the air moves at a velocity v. Momentum is p = mv = (ρ × V )v. So we can write for a small interval
of time, velocity remains approximately constant and that p = (ρ × Sv · dt)v. We can also write that
p = F t = S∆p · dt where ∆p is the difference in pressure. Therefore,
S∆p · dt = (ρ × Sv · dt)v =⇒ ∆p = ρv 2 .
The difference in pressure ∆p in terms of density inside the pipe of length L can be written as ∆p =
∆ρgL = (ρ0 − ρ)gL. This means that

2 p0 M
(ρ0 − ρ)gL = ρv = − ρ gL = ρv 2 .
RT
38
dQ
(d) From idea 1, we can write P ≡ dt . The process is isobaric so
d dn
P = (CV n∆T ) = CV ∆T .
dt dt
From the ideal gas law:
ρV dn ρS · dx ρSv
n= =⇒ = = .
M dt M · dt M
Therefore, from part (a) we can substitute to write
ρSv γ ρSv
P = CV ∆T = R(T − T0 ) .
M γ−1 M
(e) From part (b), we know that ρ = pRT

0M
which means that

p0 M 1 1 p0 M T − T0 ∆ρ ∆T
(ρ0 − ρ) = − = =⇒ = .
R T0 T R T0 T ρ T0
From part (c), we write

∆ρ v2 v2
= =⇒ ∆T = T0
ρ gL gL
which means that when substituting into our equation in (d), we have that
γ v 3 ρS
P = R T0 .
γ − 1 gL M
Remember that from the ideal gas law:
m m p0 M
p0 V = RT =⇒ ρ = =
M V RT
wich means that
γ gL P
v3 = .
γ − 1 S p0
Remember that
" #
v2 v2 γ gL P 2/3

1
∆T = T0 =⇒ T = T0 1 + = T0 1 + .
gL gL gL γ − 1 S p0
pr 50.
1. The temperature dependence of the bubble volume is w = w0 − α(T − T0 )V . Therefore, by ideal

gas law,
pw p 0 w0 p0 T w0
= =⇒ p = .
T T0 T0 w0 − V α(T − T0 )
2. The equality no longer holds when the volume is negative or
w0
w0 − V α(T − T0 ) ≤ 0 =⇒ + T0 ≤ T.
Vα
As T approaches Tmax pressure increases. Due to high pressure compressibility of glycerin and
elasticity of walls must be considered.
39
pr 51. The saturated vapor pressure at 80◦ = 50 kPa while the saturated vapor pressure at 20◦ is
2.4 kPa. When cap is open, the inside pressure is equal to the outside pressure which implies that the
partial pressure of air in bottle is equal to (101 − 50) kPa = 51 kPa. When the cap is closed, the volume
of gaseous mixture is constant. Also, note that the number of moles of air trapped in bottle are constant.
Let final partial pressure of air in bottle be x. By ideal gas equation we yield:
P0 V0 P1 V 1 51 · V x·V
= ⇒ = =⇒ x = 42.33 kPa.
T0 T1 273 + 80 273 + 20
Therefore, the total pressure is the sum of both pressures at 20◦ or (42.33 + 2.4) kPa = 44.73 kPa.
pr 52. (a) Since the process is adiabatic, we have:

1−γ
P0 γ
T0γ P01−γ = T1γ P11−γ =⇒ T1 = T0
P1
Solving, we get T1 = 279.2 K.
(b) We know from the ideal gas law that:

M P1 P0
∆ρ = −
R T1 T0
Since the change is linear, we have ρ1 = ρ0 − αh =⇒ ∆ρ = −αh. The change in pressure is given by:
Z h
1 2∆P
−∆P/g = (ρ0 − αh)dh = ρ0 h − αh2 =⇒ −αh = − − 2ρ0
0 2 gh
Therefore, we have:
2∆P M P1 P0
− − 2ρ = −
gh R T1 T0
Solving this, we get h = 1440 m. Alternatively, we can write down the first line as:
P
= constant
ρT
and go from there.
(c) If there was no rain, the temperature would be given by:
1−γ
P0 γ
T20 = T0 = 264.4 K.
P2
However, there is also condensation, which is an exothermic process and the condensation of the water
vapor releases heat into the atmosphere. This is given by:
mair mwater M LV
cV ∆T = mwater LV =⇒ ∆T = = 6.1 K
M mair R γ
γ−1
Therefore,
T2 = T20 + ∆T = 270.7 K.
(d) Given a certain unit area, 2000 kg of moist air can travel up the mountain ridge in 1500 s, so the
rate at which water condenses is:
2.45
2000 · 1000
r= = 0.00327 kg/(s · m2 )
1500
40
The height of the water column after t = 3 hours is hus:

rt
= 3.5 cm
ρ
(e) We use the same relationship:

1−γ
P3 γ
T3 = T2 = 300 K
P2
pr 53. Lemma. The molecular flux of molecules through the surface area A of ice is given by
1
Φ = n hvi .
4
Proof. Let f (v) be the Maxwell-Boltzmann distribution of velocities of the particles. What this means
is that the probability that a particle has velocity in [vx , vx + dvx ] × [vy , vy + dvy ] × [vz , vz + dvz ] is
q
f vx2 + vy2 + vz2 dvx dvy dvz .
Set up spherical coordinates with origin at the hole. We will now count the number of particles that hit
the hole in a time dt using a funny double counting argument, where we start by counting the number
of particles that hit the hole with a certain velocity and then integrate over all velocities.
We will start by counting the number of particles that move with speed v (technically speed in [v, v + dv],
but from now on we’ll be lazy about this) and spherical coordinate angles (θ, φ). Here θ = 0 means
pointing toward the hole, and θ = π/2 is parallel to the plane of the hole (the spherical coordinates for
the velocity are flipped compared to those for space, since the θ = 0 rays are anti-parallel). In a given
volume dV , the number of particles with this velocity is just
(ndV ) · f (v) · v 2 sin θ dv dθ dφ.
For this given velocity, the volume in space that will allow such particles to hit the hole is a tilted cone
object with base A, slant θ, slant height vdt, and aligned in the proper φ direction. In particular, its
volume is A(vdt) cos θ, so the number of particles with velocity (v, θ, φ) hitting the hole in time dt is
(nAdt) · f (v) · v 3 sin θ cos θ dv dθ dφ.
Thus, the rate of particles leaving is

Z ∞ Z π/2 Z 2π Z ∞
3
α = nA v f (v) dv sin θ cos θ dθ dφ = πnA v 3 f (v) dv.
0 0 0 0
Note that Z ∞Z π Z 2π Z ∞
2
hvi = v · f (v) · v sin θ dv dθ dφ = 4π v 3 f (v) dv,
0 0 0 0
which tells us that
nA
hvi, α=
4
as desired. Note that proof didn’t depend on the particular form of the Maxwell-Boltzmann speed
distributiona for the proof b .
41
This means that the mass flow rate of the molecules leaving the teacup is given byc
ρ0 hvi A
w= .
4
The time t for the complete evaporation ice will then be given by
m 4m
t≈ =
w ρ hvi A
where m is the mass of the ice in the teacup. As we are to estimate the dimensions of the cup ourselves,
let us do just that. Most convential teacups hold up to 150 mL of water and since the density of ice is
0.91 g/cm3 , we can estimate the mass of the ice in the water to be m ≈ 136 g. The area of a cup of tea
is A ≈ π(4.5)2 ≈ 63 cm3 .
Measurements were done with this teacup:
The average velocity is given by s

Z ∞
8RT
hvi = vf (v)dv =
0 πµ
which means that r r
4m 4m πµ m 2πµ
t= = = .
ρ hvi A ρ0 A 8RT ρ0 A RT
Pµ
By the ideal gas law, the vapor pressure can be represented as ρ0 = RT which means that
r r s
m 2πµ mRT 2πµ m 2πRT
t= = = .
ρ0 A RT P µA RT PA µ
While the ice is in motion, the astronaut experiences a force given from problem 21 which implies that
the astronauts acceleration is (given that the astronauts mass is M ≈ 130 kg):
PA PA
F = =⇒ a =
2 2M
42
This means that

s !2
1 1 PA m 2πRT
Lacceleration = at2 =
2 2 2M PA µ
2

1 PA 2πm RT
=
2 2M P 2 A2 µ
m2 πRT
=
2M P Aµ
(0.136 kg)2 π · 8.3 J · mol−1 · K−1 · 272 K
= ≈ 88 m
2 · 130 kg 550 Pa · 0.0063 m2 · 0.018 kg/mol
After the astronaut has stopped accelerating, it will be moving at a constant velocity of
s ! s
PA m 2πRT m 2πRT
v = at = = ≈ 0.46 m/s.
2M PA µ 2M µ
The velocity is constant since after sublimation, there are no other external forces acting on the astronaut.
Since this is happening in a vacuum, the astronaut could potentially move at a constant velocity until
infinity. However, the aim of this problem is to estimate how realistic the method is. This means that we
should aim to look at a realistic time that the astronaut travels the extra distance to the space station
a distance of L = 100 m away. The extra distance travelled will be
s ! s
m 2πRT m 2πRT
Lvelocity = v(τ − t) = τ −
PA µ 2M µ
where τ is the total time of travel. The total length travelled is then
s ! s
m2 πRT m 2πRT m 2πRT
L = Lacceleration + Lvelocity = + τ− .
2M P Aµ PA µ 2M µ
We now solve for τ . By substituting values, we yield
100 ≈ 88 + 0.46(τ − 35) =⇒ τ ≈ 252 s.
This is a realistic amount of time for the total travel, so yes, this method is realistic to use.
a
Credits to here for the proof of this alternate thermodynamic system.
b
Note that the flux here is only approximate as there will be an extra potential energy contribution to the vapors right
after evaporation and since Maxwell’s distribution is based on Boltzmann factors, we can’t really calculate the extra energy
that is added to the partition function. However, we can neglect these contributions as they are very small.
c
This method is only approximate of the actual value of L because some of the molecules that escape the teacup will
actually go back. However, the number of molecules that bounce back is very slim so we can neglect them.
pr 54. We use a model, according to which the already broken part of the soap film gathers into a
single front and moves all together towards the still preserved part of the film. Let the surface area of
the broken part of the soap film be S. It then follows that the surface energy of the soap film is
U = 2Sσ
43
Using the facts that the surface energy will turn into all of kinetic energy and that the mass that was
occupying the broken part of the soap film is 2Sρh, we can set an equation
1
2Sσ = mv 2
2
1
2Sσ = Sρhv 2
2
r
4σ
v=
ρh
Now that we have the velocity of the soap film, we can approximately find the distance the soap film
travels by using v = dt,
D D
t= =⇒ t = q .
v 4σ
ρh
pr 55. Bubble A has a radius r and the pressure difference between the inside and the air is ∆pA,air =
2 · 2γ
r . The extra factor of two is because a bubble film essentially has two surfaces. Bubble B has a
radius 2r and thus the pressure difference between the inside and the air is ∆pB,air = 4γ
2r . We now look
at the pressure difference between the two bubbles, and we get:

1 1
∆pB,A = 4γ −
r 2r
Therefore, the radius of curvature of the surface in the middle is:

1 1
= =⇒ R = 2r
R 2r
We can assume that the middle surface is in the form of a spherical cap so we can use the formula:
A = 2πR2 (1 − cos θ)
where θ is the angle that the middle surface subtends. We can determine θ by drawing a free body
diagram. We look at the intersection of the middle film, the film from bubble A and the film from bubble
B. Since all three films have the same surface tension, the three films must form an angle of 120◦ with
each other. With a little bit of geometry, we can show that the angle the middle film subtends is also
θ = 120◦ . Therefore, the surface area is:

2 3
A = 2πR = 12πr2
2
d 2

pr 56. (a) The power received by plate is equal to the power received by lens or π 2 I
44
With small angle approximations, the radius of the image is approximately f α/2. Therefore, the power
2
radiated by image π f2α σT 4 . In thermodynamic equilibrium the power received is equal to the power
radiated. This then implies that
s r
πd2 πf 2 α2 4 d I
I= σT =⇒ T = .
4 4 αf σ
(b) At the sun’s surface the intensity of the sun is given by

4 2 2
Isun = σTsun =⇒ 4ISun πRSun = 4IπROrbit
where ROrbit is the orbital distance of Earth from Sun. This means that
4 R2
σTSun Sun
I= 2
ROrbit
By Second Law of Thermodynamics temperature of plate can’t be more than temperature of Sun. There-
fore,
r
d I 2 d T 2 RSun 2 d α
T ≤ TSun =⇒ ≤ TSun =⇒ × Sun ≤ TSun =⇒ ≤ 1 =⇒ d ≤ 2f .
αf σ αf ROrbit αf 2
pr 57. The smoke rises until it’s density becomes the same as the density of the air around it. From
problem 69 and 70, note that
dT dp ∆p
= (1 − γ −1 ) =⇒ ∆T = T (1 − γ −1 ) .
T p p
The change in pressure is given by ∆p = ρgh and since ρ = pµ/RT from the ideal gas law, we get
pµ
∆p = gh
RT
45
also note that

cV cV
γ −1 = =
cp cV + R
which means that
R µgh
∆T = .
cV + R R
Rearranging to solve for h tells us that
∆T R(cV + R)
h= .
µg
pr 58. (a) Let us start out with the definition of the molar heat capacity and apply the first law of
thermodynamics:
∂Q ∂V
C= = CV + P .
∂T ∂T P
Since the process takes placed in a bubble, the thermodynamic process is polytropic. What this means
is that
p3 V = const.
Let us apply the ideal gas equation for one mole:
RT
pV = RT =⇒ p = .
V
Substituting this relation of pressure into our polytropic equation gives us
RT 3

· V = R3 T 3 V 2 = const.
V
We can apply implicit differentiation (note that the derivative of a constant will become zero) to this
relationship to yield
dV 3V
3R3 T 2 V 2 dT + 2R3 T 3 V dV = 0 =⇒ = .
dT 2T
The heat capacity of a diatomic gas at constant volume is CV = 52 R so
5 3P V 5 3
C = R+ = R + R = 4R .
2 2T 2 2
(b) Consider a small expansion of the bubble. We can relate pressure and volume by:
dp dV
=− .
p V
4
The infintesimal change in volume is dV = 4πr2 dr and the volume is V = 3
3 πr . Also note that by
Laplace’s law, the pressure is given by p = 4σ
r . This means that
4σ
pdV · 4πr2 dr 12σdr
dpout =− =− r 4 3 =− 2 .
V 3 πr
r
Note that there is also a decrease in the pressure due to surface tension

4σ 4σdr
dpin = ∆ = 2 .
r r
46
This means that

4σdr 12σdr 8σdr
dptot = 2
− 2
=− 2 .
r r r
Consider a small area element S that moves outward a small distance x. The mass of the bubble is given
by m = ρSh which means that the force is
r
8 8σ
ρhẍ = − 2 x =⇒ ω = .
r ρhr2
pr 59. (a) From E → A, the gas is undergoing an isothermal expansion. Since P V = constant and V
is increasing, then P must be decreasing. Therefore, pE > pA . From B → D it is again an isothermal
process. However, there is both a heating component and a cooling component to it. We need to
determine what is the overall effect. This can be easily done by assuming that in one cycle, the net
heat gain is zero (or else it would continuously cool down/warm up). Therefore, since energy is already
gained from E → A, it must be lost from B → D so pD > pB . Finally, we can rank them based off of
their relative heights:
pE > pA > pD > pB > pC
(b) Note that we will use a different notation than the problem text, and let the subscript represent the
lettered state. From D → E, it is an isothermal process so:
PD1−γ TDγ = PE1−γ TEγ
where
CV 1 7
CP = CV + R =⇒ 1 = + κ =⇒ γ = =
CP 1−κ 5
Solving for TD gives:
(1−γ)/γ
PE
TD = TE
PD
(1−γ)/γ
PA + 20
= TE
PD
= 194.8 K
(c) Again, the process from A → B is adiabatic and we have TB = TD . Therefore:
PB1−γ TBγ = PA1−γ TAγ

γ/(1−γ)
TA
PB = PA
TB
= 220.6 hPa
Alternate Solution: Writing out the adiabatic compression and expansion equations, we get:
PD1−γ TDγ = PE1−γ TEγ

PB1−γ TBγ = PA1−γ TAγ
47
By noting that TA = TE and TB = TD , we can divide the two equations and get:
PD PE
=
PB PA
PD PA
PB =
PE
= 220.6 hPa
(d) (i) There are four processes, an isothermal expansion, an adiabatic expansion, an isothermal cooling,
and an adiabatic compression. Let us first deal with the adiabatic processes. For an adiabatic process,
we have Q = 0 so the net work done by the gas is equal to the change in energy:
W = −nCV ∆T
The net work due to the two adiabatic processes will be zero since ∆T1 + ∆T2 = 0. Therefore, we only
need to focus on the isothermal processes. The work done at the top is:
Z
WB→D = P dV
Z PD
1
= −nRT dP
PB P

PB
= nRTB ln
PD
At the bottom, the work done is:

PE
WE→A = nRTE ln
PA
so the net work per mole is:

PB PE
Wnet /n = RTB ln + RTE ln
PD PA

PE PE
= −RTB ln + RTE ln
PA PA

PE
= R(TE − TB ) ln
PA
where we have used:

PD PE
=
PB PA
from earlier.
(ii) From the first law of thermodynamics:
U =Q−W
Since U = 0, the heat loss must equal the work done. The heat loss at the top must be equal to the
work done at the top, or:
PD
Qloss /n = −WC→D = RTB ln
PC
48
(e) The Hadley Circulation is just a modified Carnot Cycle so the efficiency is just:
194.8
η =1− = 35.1%
300
(f ) The efficiency is:

Wnet Wnet
η= =
Qgained Wnet + Qloss
Substituting in the expressions from part (D) gives:

R(TE − TB ) ln PPEA
η=
R(TE − TB ) ln PPEA + RTB ln PPDC

(TE − TB ) ln PPEA
=
TE ln PPEA − TB ln PPEA + TB ln PPBA PPCE

(TE − TB ) ln PPEA
=
TE ln PPEA − TB ln PPB C
This is smaller than ηideal if:

(TE − TB ) ln PPEA T − TB
< E
TE ln PPEA − TB ln PPB C TE

PE PE PC
TE ln < TE ln − TB ln
PA PA PB

PC
ln >0
PB
PC > PB
Since from part (A) we have stated that PC > PB , then that means even under ideal conditions, the
system cannot reach Carnot efficiency. It can only reach Carnot efficiency if PC = PB , which will simply
not be the case.
(g) (I) cannot be true. In an isothermal process W 6= 0. In fact, friction is even necessary to keep it at
a constant temperature as the gas will be absorbing energy.
(II) By having condensation only occur at a higher temperature, this implies the process is irreversible.
See (III) for why this lowers the efficiency.
(III) is true. An engine can only achieve Carnot efficiency when ∆S = 0, that is the process is reversible.
The evaporation is irreversible as the condensation only occurs at a much higher altitude. This means
that the efficiency is lowered.
(IV) is not necessarily true. Although generally phase shifts involve a change in energy, if the transition
is kept at constant volume, entropy can be kept constant.
Therefore the answer is (II)&(III)
49
pr 60. (a) In analogy to electrical circuits, the resistance of a single wire is given by:
L
R=
κS
Since we have four wires in parallel, the effective resistance is:
L
R=
4κS
(b) For a small period of time dt, the temperature is changed by ∆T degrees. This means that
dQ
P ≡ = C∆Ṫ .
dt
Let us look at the electrical analogy between heat conduction and electric circuits. In this case, Q̇ acts
as the current while the temperature difference ∆T ≡ T − T0 acts as the voltage difference. From this
perspective, we can also define
∆T
Q̇ = .
R
This means that from adding these two individual expressions (as there is an extra Q̇ due to our ”voltage
difference”) we have that
∆T
P cos ωt = C∆Ṫ + .
R
As power varies with time, we attempt to seek the solution in the form of
T = T0 + ∆T sin(ωt + φ).
Taking the derivative of our sought solution and substituting into our differential equation gives us
∆T π
P cos ωt = C∆T ω cos(ωt + φ) + cos ωt + φ −
R 2
which can be seen as the sum of multiple rotating vectors or:
P~ = P~C + P~R .
Each of these vectors have an individual magnitude of

∆T
PC = C∆T ω, PR =
R
50
Therefore, from pythagorean theorem, the amplitude of oscillations is

2
∆T P0
P02 = (Cω∆T ) +2
=⇒ ∆T = √ .
R + R−2
C 2ω2
Our solution is in the form of
P0 cos(ωt + φ)
T = T0 + ∆T sin(ωt + φ) = T0 + √ .
C 2 ω 2 + R−2
Let us estimate the phase difference. We can go back to our phase diagram and see that

PC Cω
φ = arcsin = arcsin √ .
P0 C 2 ω 2 + R−2
This means that our final solution is
√
P0 cos ωt + arcsin Cω/ C 2 ω 2 + R−2
T = T0 + √ .
C 2 ω 2 + R−2
(c) Remember from part b that our amplitude of oscillations of ∆T is given by ∆T = √C 2 ωP20+R−2 . We
want there to be as large a change as possible with a small change in C. This is equivalent to maximizing
the derivative ie setting the double derivative to 0. Upon taking two derivatives, we get the function
3P0 C 2 ω 4 P0 ω 2
− = 0.
(C 2 ω 2 + R−2 )5/2 (C 2 ω 2 + R−2 )3/2
√
Seperating and solving for ω yields ω = 1/ 2CR.
(d) As the questions asks to estimate, we ignore all numerical prefactors. From part c), we know that
C ≈ 1/ωR. This should be of the same order as the heat capacity of the bridges. Therefore, we have
1 κ
= ρlSc =⇒ ωc ≈ .
l/κS · ω cρL2
pr 61. (a) Let the exitance and luminosities be M1 and J1 and M2 and J2 for the inner and outer
sphere respectively. Assume that the spheres have different emmistivities ε1 and ε2 respectively. We
then can write that
J1 = M1 + (1 − ε1 )F11 J1 + (1 − ε1 )F12 J2
and
J2 = M2 + (1 − ε2 )F21 J2 + (1 − ε2 )F22 J1
as the equations of heat radiation through each sphere respectively (note that F is the view factor). All
the radiation emmited by one sphere will reach the other sphere so F11 = 0 and F12 = 1. We also note
that 2
R1
A1 = F21 A2 =⇒ F21 = .
R2
This also means that 2
R1
F22 = 1 − .
R2
51
Substituting these values of the view factor into our expressions of radiosities tells us:
J1 = M1 + (1 − ε1 )J2
2 2 !
R1 R1
J2 = M2 + (1 − ε2 ) J2 + (1 − ε2 ) 1 − J1
R2 R2
Using the fact that M2 = ε2 σT24 we can solve for J1 in terms of J2 in the second expression:
2
R1
J2 − ε2 σT24 − (1 − ε2 ) R 2
J2
J1 = 2
(1 − ε2 ) 1 − R R2
1
We can solve for J2 in terms of J1 in the first expression to get

2
J1 −ε1 σT14 R1 J1 −ε1 σT14
1−ε1 − ε2 σT24 − (1 − ε2 ) R 2 1−ε1
J1 = 2
(1 − ε2 ) 1 − R R2
1
Isolating for only J1 finally gives us

2 2
R1
ε1 ε2 − ε1 R 2 − ε1 ε2 R
R2
1
σT14 + (1 − ε1 )ε2 σT24
J1 = 2 2 .
R1 R1
ε2 + ε R 2 − ε1 ε2 R 2
The net heat flux between the cylinders is the heat flux from the inner cylinder or
A 1 ε1 4πR12 σ(T14 − T24 ) εσ4πR12 (T24 − T14 )

Q̇ = (σT14 − J1 ) = 2 = .
1 − ε1 1 1−ε2 R1 1 + (1 − ε)R12 /R22
ε1 + ε2 R2
The last step was done by noting ε = ε1 = ε2 .
(b) Note that τ P = λm and m = 43 πρR3 . Therefore,
4
τ = πρR3 λµ/P.
3
pr 62. a) After each pumping cycle, the amount of gas inside, reduces by a factor of 1 − α. This means
that after N pumping cycles, the pressure inside the bulb decreases by a factor of β = (1 − α)N . When
N → ∞, the factor β can be approximated by β = e−N α . This means that
1
β = e−N α =⇒ N = ln β .
α
b) We know that work is generally defined by

W = −pV.
However, since the volume in the cylinder is given by αV , then after N cycles, the work is defined by
W = −N αp0 V =⇒ W = p0 V ln β .
52
c) We know that the adiabatic law gives us the proportion pV γ ∝ T γ and the ideal gas law tells us
pV ∝ T =⇒ (pV )γ ∝ T γ . Substituting these in, gives us pγ−1 ∝ T γ . When the pressure in the bulb
becomes βp0 , the pressure in the cylinder, will in proportion, be increased by a factor of 1/β. Therefore,
T = T0 β 1/γ−1 .
d) The trick to this part is that the only time energy is lost is in the heating up of the released air since
everywhere else the work is ”reused”. There is two places work is done. Actually creating the vacuum
and then heating the air. To actually create the vacuum takes P V energy and to heat up the air takes
the sum of all the temperature heating. We calculate the second part by sum of
β (1/γ−1) − 1
Cv · β · P V α .
1+α
Since β = 1/(1 + α)(n/γ) where n is the pump number we just use a geometric series to sum these up
and replace Cv = 1/γ − 1. Thus, multiplying out everything cancels and you just end up with another
P V so the answer is 2P V .
pr 63. (a) By idea 1, note that

dQ dT
P ≡ =C .
dt dt
Therefore, by rearranging, we have that
Z T1 Z τ
P Pτ
T = dT = dt =⇒ C = = 360 J/◦ C.
T0 0 C T0 − T1
(b) The delay in the change of temperature reading and the inital rise in temperature is due to the
sensor’s surroundings not initially heating up. By the definition of heat capacity, an infintesimal amount
of heat that is transfered to the system is
dQ = −CdT̄ .
After the temperature of the sensor’s surroundings and the plate become equal, the heat dissipated by
the plate is given by
dT κt
κ(T − Tamb ) = −C =⇒ ln(T − Tamb ) = − + m
dt C
where m is an arbritary constant. Before the sink recieves heat Q, it’s temperature must be be the
ambient temperature as the electrical component has been switched off for a long time. Theoretically,
we assume that the temperature of the whole plate and sensor is the same right after t = 0 for our
equation to be valid for the entire duration (although realistically it is valid only after some time). Let
this temperature be T2 . Then we have that
Q = C(T2 − Tamb )
and therefore our goal is to determine the value of m from the graph given. Our solution of ∆T ≡
T2 − Tamb will be exponential or em . From the table, it is clear that Tamb = 20◦ C. Therefore, we plot
our graph with y ≡ ln(T − Tamb ) and x ≡ t. From the graph drawn, we find m = 4.73 and therefore,
Q = 41 kJ.
53
pr 64. Radiation occurs between the celestial body and its environment while conduction occurs inside
the celestial body. When the body is in thermal equilibrium heat production is equal to heat emission.
Let the surface temperature be TS and the center’s temperature be TC . Then at the surface
r
4 4 ρP R
4πR2 σTS4 = πR3 ρ =⇒ TS =

3 3σ
Inside the celestial body consider a thin shell of radius x where x ≤ R of thickness dx at a temperature
T with a temperature difference dT . Then,
Z TC
−k(4πR2 )dT ρP 0 ρP R2
Z
4 3
πx ρ P = =⇒ dT = − xdx =⇒ TC = TS + = 4689 K
3 dx TS 3k R 6k
pr 65. An ideal-gas cycle consisting of two isotherms at temperatures T1 and T2 , and two isochores
joining them will look like:
P
Q1 Q2
4
T2
2
Q3
Q4
1 T1
V1 V2 V
Since there are no adiabatic processes, heat will be lost on every step. For isochoric processes on 2 → 3
and 1 → 4, the heat released will be
Q1 = CV R(T2 − T1 )
Q3 = CV R(T1 − T2 )
For isothermal processes on 1 → 2 and 3 → 4, the heat released will be

V2
Q2 = RT2 ln
V1
V1
Q4 = −RT1 ln
V2
The heat released during cooling feeds the heating isochore. Therefore, the work done is the change in
heat along both isotherms or:
W = Q2 − Q4 .
54
The efficiency of the cycle will be given by
W Q2 − Q4 T1 ln VV12 T1
η= = =1+ V2
= 1− .
Q2 Q2 T2 ln V1 T2
pr 66. (a) The total heat energy will be given by Q = nq where n is the number of moles and q is the
heat. Note that for a heat differential, dq = Cv dT holds true. By integrating over we find that
Z
q = Cv dT ≈ 550R.
The number of moles will be given by the mass over the molar mass or
a3 ρ
n= ≈ 0.1 mol.
M
Multiplying these two values together gives us approximately 520 J.
(b) Consider a photon radiated from the cube. It caries a momentum p = E⊥ /c where E = hf . Let us
assume that the photon initially imparts at an angle θ. The parallel momentum is given by
E⊥
pk = cos θ.
c
Note that the intensity of light is proportional to cos θ by Lambert’s law. Since intensity is proportional
to frequency, the momentum averaged over all directions is
Z π
E⊥
cos2 θdΩ

pk =
0 c
where dΩ = 2π sin θdθ is the solid angle. Integrating this tells us that

E⊥ Q
pk = = .
3c 3c
The mass of the cube is a3 ρ so, the velocity is given by
Q
v= 3 .
3a ρc
(c) By Stefan-Boltzmann Law, we have that

P
= σT 4
A
and by idea 1:
dQ dT
P ≡ =C .
dT dt
This means that
CdT = σT 4 dt =⇒ T 3 dT = σT 4 dt.
Separating the differential equation tells us
dT
= −Bt =⇒ T = A · e−Bt .
T
(d) We can estimate the momentum imparted upon the cube will be equal to the heat energy over the
average velocity of the hydrogen molecules. In other words,
r r
3 Q MH Q MH
a ρv = =Q =⇒ v = 3 .
vH RT a ρ RT
55
pr 67. The total light emission power of the LED is P0 = 1µW .
Q0
W
LED
Since we have Z ν2
P0
= I(ν, T )dν
A ν1
We substitute the well known expression for I(ν, T ) from Planck’s radiation law:
λ2 ν2
2πhc2 dλ ν3
Z Z
P0 2πh
= hc = 2 hν dν
A λ1 λ5 (e λkT − 1) c ν1 e kT0 − 1
We can use the well known integral
∞
η 3 dη π4
Z
= 6ξ(4) =
0 eη − 1 15
and substitute the known values and using the problem condition. We find from evaluating the integral
that T1 ≥ 1116K. Now, the maximum possible efficiency of the LED is just
T1
ηmax = − 1 = 1 + 0.354 = 1.354
T1 − T0
pr 68. Solution 1: Note that the internal energy is given by cV T and when this goes to the rocket
nozzel, part of this turns into kinetic energy µv 2 /2. Since the gas is monoatomic, part of it transfers into
work pV = RT . Therefore, we have our conservation of energy equation to be
1 2
cV T0 + RT0 = cV T1 + RT1 + µvexit .
2
Using the fact that cv = 52 R, we can simplify this to solve for the final exit velocity
s
5 5 1 2 2 7(T0 − T1 )
RT0 + RT0 = RT1 + RT1 + µvexit =⇒ vexit = .
2 2 2 µ
56
The force is then given by the exit velocity or
2 7(T0 − T1 )
F = mv = ρ1 Avexit = ρ1 A .
µ
We can attempt to use the approximation that T0 T1 to simplify this problem:

7AρT1 T0 7Ap1 T0
F = −1 ≈ .
µ T1 µ T1
Solution 2: Applying Bernoulli’s principle for the gas between the nozzle and the chamber exit,
v 2 Cp T Cp T0 v2 Cp T1
+ = const. =⇒ 0 + = exit +
2 M M 2 M
Now, the thrust applied equals
2
F = ρSvexit
P1 M 2 2Cp (T0 −T1 )
and ρ = RT1 , vexit = M . This means that
2Cp (T0 − T1 )P1 S

F =
RT1
7R
Given that: T0 T1 =⇒ T0 − T1 ≈ T0 and Cp = Cv + R = 2
7T0 P1 S
∴F ≈
T1
pr 69. Let us analyze a small section dz in the atmosphere with density ρ. The relation between
pressure and density is then given by
dp = −ρgdz.
We know that p = nkB T and ρ = nm where m is the mass of a single molecule and therefore,
dp mgp dp mg
=− =⇒ T =− dz.
dz kB T p kB
Note that
and therefore,
dp dT
(1 − γ) +γ = 0.
p T
Now substituting gives us

dT γ−1 mg µg
=− =⇒ dT = −(1 − γ −1 ) dz.
dz γ kB R
Integrating from T0 to T gives us

Z T Z z
µg µgz
dT = −(1 − γ −1 ) dz =⇒ T = T0 − (1 − γ −1 ) .
T0 0 R R
57
pr 70. (a) We essentially apply the same process as problem 69. Note that
and therefore,
dp dT
(1 − γ) +γ = 0.
p T
This means that
dT dp
= (1 − γ −1 ) .
T p
(b) Let us analyze a small section dz in the atmosphere with density ρ. The relation between pressure
and density is then given by
dp = −ρgdz.
We know that p = nkB T and ρ = nm where m is the mass of a single molecule and therefore,
dp mgp dp mg
=− =⇒ T =− dz.
dz kB T p kB
(c) From parts a and b we can substitute to give us

dT γ − 1 mg µg
=− =⇒ dT = −(1 − γ −1 ) dz.
dz γ kB R
Integrating from T0 to T gives us

Z T Z z
µg µgz
dT = −(1 − γ −1 ) dz =⇒ T = T0 − (1 − γ −1 ) = 20.6 C◦ .
T0 0 R R
pr 71. (a) The internal energy will be given by

4 3
U = µCV ∆T = µC πR (Tc − T0 ) ≈ 16768 J.
3
(b) By Fourier’s Law,

∆T κ (T1 − T0 )
J = Q̇ = κ = ≈ 2458 W/m2
R R
(c) Heat is transferred to the egg via its surface
P = J × 4πR2 = [4πKR (T1 − T0 )] ≈ 19.3 W
(d) Power is equal to the change in internal energy over time or:
µc 43 πR3 (Tc − T0 )

dU ∆U ∆U
P = ≈ ⇒ ∆t = = .
dt ∆t P 4πκR (T1 − T0 )
58
This means that

µCR2

Tc − T0
τ= ≈ 869 s.
3K T1 − T0
pr 72.
1. The temperature dependence of the bubble volume is w = w0 − α(T − T0 )V . Therefore, by ideal

gas law,
pw p 0 w0 p0 T w0
= =⇒ p = .
T T0 T0 w0 − V α(T − T0 )
2. The equality no longer holds when the volume is negative or

w0
w0 − V α(T − T0 ) ≤ 0 =⇒ + T0 ≤ T.
Vα
As T approaches Tmax pressure increases. Due to high pressure compressibility of glycerin and
elasticity of walls must be considered.
pr 73. A-i) First, note that by the ideal gas law pV = nRT0 which means that ρ = PRT Mair
0
. The differ-
ential change in pressure for a differential change in height dz is dP = −ρgdz. Then, after substituting
the above expression for the density of air as a function of pressure, we get the following expression:
Z P
Mair g Z
Z
dP
=− dz.
P0 P RT0 0
Mair g
Evaluating, we find that α = RT0 .
A-ii) The density of air, ρ, can be taken as a constant; thus,

Z P Z Z
dP = −ρg dz,
P0 0
and P (z) = P0 − ρgh.
A-iii) After substituting the given values, PB ≈ 88.24 kPa.
B-i) Consider an infinitesimally thin rectangular of prism of air with an area A and a thickness dr. If
the pressure goes from P to P + dP from one side of the piece of air to the other, the net force on it is
2
AdP . This net force is the centripetal force, so AdP = mvr . Noting that m = ρV = ρAdr, and solving
v2
algebraically, dP
dr = ρair r .
vG rG
B-ii) Since angular momentum is conserved, mvG rG = mvr; thus, v = r .
B-iii) Point G is on the isobar boundary layer, and so is Point B; therefore, it can be said that PG = PB .
Then, using the Bernoulli equation, approximating vA ≈ 0, and substituting PG = PB , PA = 21 ρair vG 2 +
q
−PB ) √
PB . Some algebraic manipulation yields vG = 2(PAρair = 2gh ≈ 141 m/s.
B-iv) Points C and G are also on the isobar boundary layer, meaning that PC = PG . Again using
Bernoulli’s equation, PG + 12 ρair vG
2 = P + ρ gz + 1 ρ v 2 . Cancelling out P and P and ρ
C air 2 air C C G air as
2 (1 − rG 2 ) = 2gz. As a result,
well as substituting the expression v = vGrrG which was derived before, vG rC
2
vG √ v2
z = 2g (1 − rC ). However, through a simple rearrangement of v = 2gh, h = 2gG . Thus, hz = 1 − rrG 2 ,
rG 2
which enables us to graph hz vs. rrG , as shown.
59
B-v) Since the term 2gz

2 will decrease as vG increases, the radius will be less dependent on the height
vG
for high-speed tornadoes. As a result, tornadoes with relatively uniform diameter tend to have higher
ground rotation speeds.
C-i) The angular velocity inside the core is constant since it behaves like a rigid body. Therefore,
v = ωr = vG rrG .
2
C-ii) From before, dP v
dr = ρair r . Multiplying both sides by dr and integrating the expression from the
center to a far distance away, we have:
Z rG 2 Z ∞ 2
vG r vG r
PD − P0 = ρ 2 dr + 2 dr .
0 rG rG rG
2 ≈ 76.48 kPa.
Evaluating and solving algebraically for PD , we get PD = P0 − ρvG
1
C-iii) Assuming adiabatic behavior, P 1−γ ∝ Tγ . Therefore,
γ−1
PG γ
TG = T0 ≈ 4.89 ◦ C
P0
and
γ−1
PD γ
TD = T0 ≈ −6.25 ◦ C.
P0
C-iv) The low temperature causes condensation of moisture that gets sucked into the core, which releases
a lot of latent heat, a source of the tornado’s energy.
D-i) We assume that the house is tightly enclosed with pressure P0 inside. By substituting v = vG rrG
dP 2
into dr = ρair vr , the following integral can be set up:
Z ∞ 2 r2
vG G
∆P = ρair dr.
2rG r3
Evaluating, ∆P = 18 ρair vG 2 . Since the lift force is F = ∆P A, the ratio of the lift force to weight for the
L
roof is FFGL = ρroof
∆P A
Atg = 3.75.
D-ii) The lift force is not much larger than the weight of the roof, most roof are mounted firmly to
withstand forces many times it’s weight. Therefore, chances are the pressure difference would not cause
the house to explode so soon (unless the roof is very poorly mounted). Opening windows during a tornado
isn’t a good idea for another reason. Flying debris is responsible for most twister-related injuries, so
standing next to an opening that could potentially blast you with shards of glass and other projectiles
isn’t a great idea.
60
pr 74. A-i) The buoyant force is going to be:
F = ρair Vballoon g
P MA
where ρair = RT from the ideal gas law. Applying the ideal gas law to inside the balloon gives:
(P + ∆P )V
RT =
n
Therefore, the buoyant force is:
P
nMA g
P + ∆P
P MA P MA z0
A-ii) We know that ρ = RT =⇒ ρ = RT0 (z0 −z) . The differential change in pressure for a differential
change in height dz is
P MA z0 g
dP = −ρgdz =⇒ dP = − dz.
RT0 (z0 − z)
This means that by integrating,
Z P
MA z0 g z dz

z0 − z
Z
dP P M A z0 g
=− =⇒ ln = ln
P0 P RT0 0 z0 − z P0 RT0 z0
and
MA z0 g
z RT0
P = P0 1 − .
z0
P0 MA MA ρ0
Also, note that ρ0 = RT0 =⇒ RT0 = P0 and therefore:
ρ 0 z0 g
η=
P0
B-i) We can apply the method of virtual work. The work needed to change the radius by dr is:
∆P (4πr2 )dr
which causes a change in energy of
dU = 4πr02 κRT (4λ − 4λ−5 )dr/r0
Equating gives:
4κRT
∆P (4πr2 )dr = 4πr0 κRT (4λ − 4λ−5 ) =⇒ ∆P = (2λ−1 + λ−7 )
r0
The graph pretty much looks like:
8κRT
∆P =
r0 λ
except for small values, at which it increases to infinity quickly.
We can alternatively treat the energy as:
U = 4πr2 γ
where γ is a varying surface tension. Solving for γ and dropping the constant factor, we can apply
Laplace’s pressure ∆P = 4γ/R and solve for ∆P .
61
B-ii) Initially P0 V0 = n0 RT0 where V0 is the unstretched volume. Finally
(P0 + ∆P )V0 λ3 = nRT0
as V ∝ r3 using the result from part B(i) we get ∆P = 4κRT −1 − λ−7 ). This gives us
r0 (λ

n 1
n0 λ3 − 1
P0 V0 λ3 (1 + a(λ−1 − λ−7 )) = nRT0 =⇒ a = = 0.11
(λ−1 − λ−7 )
pr 75. We follow the same sign notation as in problem 6, since they are parts of the same problem.
1200 K
Since we clearly have the constant corresponding to the cutoff frequency η0 = 6000 K , thus
R η0η 3 dη
0 eη −1
γ= R ∞ η3 dη
0 eη −1
follows from Planck’s radiation law. To simplify this result, we use the approximation for small η
eη ≈ 1 + η ∀ η 1
Also since we have

∞
η 3 dη π4
Z
= 6ξ(4) =
0 eη − 1 15
γ is simplified to
R η0 η 3 dη
0 eη −1
γ= ≈ 4.106 × 10−4
π4
15
Now by definition of γ, we have the relation Pin = γPout , which gives us
Poutput = P γeff
2
2 R⊕
× 4πR⊕ σT 4 =a× σT4 (4πR
2
)
2L
2
2 4 R⊕
× 4πR⊕ σT = a × σT4 (4πR
2
)
2L
σT4 (4πR2 )

2
πR × × γ = 4πR⊕ 2 (σTS 4 )
4πL2
1 r√
γT 4 R 2 4

γR
TS = = T
4L2 2L
Note that the temperature of the satellite is independent of the size of the satellite, as long as the size
is not large. Thus, substituting the value of γ, we get
r√
γR
TS = T ≈ 41 K
2L
62
pr 76. Since an ideal heat engine has practically zero heat and friction losses, it takes some amount
of heat from one object and delivers heat to another object having the same thermal capacity. Say QA
amount of heat was initially withdrawn from A and QB heat given to B, and let TB < TC < TA , where
TC is the common temperature reached. By second law of thermodynamics, entropy is equal or:
QA QB
=
T1 T2
Also, by definition
dQA = −CdT1
dQB = CdT2
From these we have

Z TC Z TC
dT1 dT2 dT1 dT2 TC TA p
−C =C ⇒ −C =C ⇒ ln = ln ⇒ TC = TA TB
T1 T2 TA T1 TB T2 TB TC
We can also apply relations for differential quantities of QA and QB as we wrote earlier in the solution.
For dQA , we have:
Z TC p
dQA = −CdT1 ⇒ QA = C dT1 = C(TA − TC ) = C(TA − TA TB ).
TA
For dQB : Z TB p
dQB = −CdT2 ⇒ QB = C dT2 = C(TC − TB ) = C( TA TB − TB )
TC
Now clearly by the first law of thermodynamics, we have

p p p 2
∆W = QA − QB = C(TA − TB − 2 TA TB ) = C( TA − TB ) .
pr 77. Define T0 to be the air temperature, T1 to be temperature supplied to the colder joint, and T2
to be the temperature supplied to the warmer joint.
Let temperature of colder joint be T0 − T1 , and warmer joint be T0 + T2 . The thermocouple acts as a
reversible carnot engine in reverse, with electrical energy supplying the necessary work; therefore we can
relate work done to heat in/heat out. The Carnot Cycle usually involves work/heat, but we can also
look at the work per second (or power) and obtain an equivalent result. Note that the amount of power
done is
P = IV = Iα(T1 + T2 ).
The amount of heat that is supplied to the cold side of the thermocouple is κT1 . The amount of work
done is then
(T0 + T2 ) − (T0 − T1 )
W = κT1
T0 − T1
Similarly the amount of heat supplied to the warm part of the thermocouple is κT2 . This tells us the
amount of work done is
(T0 + T2 ) − (T0 − T1 )
W = κT2 .
T0 + T2
63
We can now equate these together to get the equality of
(T0 + T2 ) − (T0 − T1 ) (T0 + T2 ) − (T0 − T1 )

Iα(T1 + T2 ) = κT1 = κT2
T0 − T1 T0 + T2
simplifying gives us
κT1 κT2
Iα = = .
T0 − T1 T0 + T2
We can now easily solve for T1 to get
Iα(T0 − T1 ) = κT1
IαT0 = T1 (κ + Iα)
Iα
T1 = T0
κ + Iα
pr 78. The friction in the disk disappears because the CO2 evaporates. The created pressure due to
the force applied to the disk is given by
F F
P = = 2.
A πr
Dry ice begins to sublime only when vapor pressure exceeds ambient pressure. Therefore, the minimum
vapour pressure needed is given by
F
Pvap = Pair + = 419.3 Pa.
πr2
From the graph, we can look at the points to see that T ∼ 213 K
64
pr 79. The heat exchange through walls is equal to P tt−t 0

1 −t0
while the heat exchanged with incoming
v
air = V Cp (t − t0 ) where V is molar volume. Therefore
t − t0 v
P =P + Cp (t − t0 )
t1 − t0 V
which implies that
P
t = t0 + vCp
⇒ t = 13.16◦ C.
P
t1 −t0 + V
pr 80. The density of some matter of mass m can be given by
m
ρ=N
V
where N is the number density of the substance. This means that for dry and humid air on each
respective air can be given as
md mh
ρd = Nd , Nh .
V V
The number density of the dry air is given by
M M
Nd = =
Ma 28.8
65
while the number density of the humid air will be
M0 M0 M0 M0
Nh = 0.02 · + 0.98 · = 0.02 · + 0.98 · .
Mw Ma 28.8 18
We can now compare the ratio of densities since we require the number density to be the same or in
other words,
M 0
ρd Nd md 28.8 · M 1
= = M0 M0
= 0.02 0.98
= 0.9881.
ρh Nh mh M 0 · 0.02 · 28.8 + 0.98 · 18 M 28.8 18 + 18
This means that the moist air is then
ρm = ρh = 0.9881ρd = 1.2352 kg/m3 .
pr 81. (a) Since the process happens in a pipe, we apply bernoulli’s law:
1
P + ρv 2 + ρgh = const.
2
The two places we apply Bernoulli’s law is at the bottom of the furnace and at a height h. Using the
assumption that smoke temperature can be assumed to be constant throughout the entire length of the
chimney we find that
1
Ph + ρsmoke v 2 + ρsmoke gh = Patm
2
from taking the reference point to be at the bottom of the furnace. We write that
Ph = Patm + Pair =⇒ Patm − Ph = −Pair = ρair gh.
This then means that

s
1 ρair
ρsmoke v 2 = gh(ρair − ρsmoke ) =⇒ v = 2gh −1 .
2 ρsmoke
If we want efficient withdrawal of all the gas, we then need for v ≥ B

A so all the gas can leave. In other
word, we can create an inequality:
s −1
B2

ρair B ρair
2gh −1 ≥ =⇒ h ≥ −1
ρsmoke A 2gA2 ρsmoke
and by the ideal gas law

ρair Tsmoke
=
ρsmoke Tair
which gives us
−1
B2

Tair
h≥ −1
2gA2 Tsmoke
B 2 1 Tair
≥ 2 .
A 2g ∆T
66
(b) We simply analyze the ratios of h(30) and h(−30):

T (30)
h(30) Tsmoke −T (30) h(30)
= T (−30)
=⇒ = 1.45 =⇒ h(30) = 145 m.
h(−30) 100
Tsmoke −T (−30)
(c) Note that s s r

ρair Tair ∆T
v= 2gh − 1 = 2gh − 1 = 2gh .
ρsmoke Tsmoke Tair
This means that the velocity is constant.
(d) Consider the gas at a height z. From the bernoulli equation in part (a), we find that the pressure is
given by
1
Pz + ρsmoke v 2 + ρsmoke gz = Patm .
2
Using the fact that
1
ρsmoke v 2 = gh(ρair − ρsmoke ),
2
we can then say that
Pz = Patm − gh(ρair − ρsmoke ) − ρsmoke gz.
(e) In a time ∆T , the mass of air that is released is given by m = ρ · Av∆t which means the kinetic
energy of the air is
1 1
T = mv 2 = (ρ · Av∆t)v 2 .
2 2
Substituting our expression of v 2 from before yields,
∆T
T = ρav∆tgh .
Tair
Note that P = ∆E/∆t which means that
∆T
Pair = ρAvgh .
Tair
The power produced by the sun is equal to (note the mass here is per unit time)
Psun = mc∆T = ρAvc∆T
which means that we can then divide both expresions to result in the efficiency of the system or
Pair gh
η= = .
Psun cTair
(f ) Since η ∝ h, the relationship is linear.
(g) By plugging in numbers to our efficiency in (e), we find that η = 0.64%.
(h) The power produced can be given by
Psun = ηGS ≈ 45 kW.
67
(i) If there are 8 sunny hours per each day, we simply multiply our result from (h) to get 360 kW.
(j, k) The mass flow rate will be given by
r
∆T
w = ρAv = ρA 2gh .
Tair
We also know that
GS
P = wc∆T = GS =⇒ w = .
c∆T
Therefore, by equating both expressions we have the equation
r 1/3
G2 S 2 Tatm

∆T GS
ρA 2gh = =⇒ ∆T = .
Tair c∆T A2 c2 ρ2 2gh
pr 82. At a temperature of 77.4 K (i.e. at the boiling point of nitrogen), the pressure of saturated
nitrogen vapour is 1 atm ,while the saturated pressure of oxygen becomes 1 atm at a higher temperature
of 90.2 K.
The molar ratio of oxygen and nitrogen is 21 : 79. The ratio of the partial pressures of the two components
will also be very close to molar ratio, because, until the start of liquefaction, the behaviour of each gas
constituent is very close to that of an ideal gas. When the partial pressure of oxygen is 0.2226 atm,
liquefaction of oxygen starts. The partial pressure of oxygen does not increase further as temperature is
constant.The partial pressure of N2 at the onset of oxygen liquefaction is 0.2226 × 79 21 atm = 0.8374 atm.
This is less than the saturated vapour pressure of nitrogen at this temperature, which, since 77.4 K is
nitrogen’s boiling point, has a value of 1 atm.Consequently, nitrogen does not liquefy at this pressure.
Therefore total pressure at this point P1 is 0.8374 + 0.2226 = 1.06 atm. As the compression is isothermal,
volume at this point V1 is 1.001 × 0.500
1.06 l = 0.4721 l. During the subsequent compression, the partial
pressure of the oxygen, already in two phases, does not change, while the nitrogen pressure increases
from 0.8374 atm to 1.00 atm. This latter pressure will be reached when the volume has been reduced by
a factor of (0.8374/1.00) = 0.8374. Therefore volume at this point V2 is 0.8374 × 0.4721 l = 0.3953 l.After
that, the total pressure remains constant (at 0.2226 + 1.00 = 1.22 atm) until the liquefaction is complete,
just the volume is decreased now.
pr 83. (a) Consider a cross-section with width dh. The pressure is then given by
dP = −ρgdh.
From the ideal gas law, we have that

PM
ρ=
RT
which means that Z H Z P/e
Mg dP M gH 1
− dh = =⇒ − = ln = −1.
0 RT P P RT e
This means that
RT
M gH = RT =⇒ M̄ = = 14.5 g/mol.
gH
68
(b) By the ideal gas law

n NA P0
RT = P0 =⇒ n = .
NA RT
The mean free path λ is given by
RT
λ = (σn)−1 =⇒ λ = = 3286.67 m.
NA σP0
(c) We are given in part (a) that

1 1
λEB = H =⇒ nEB = = .
λEB σ σH
This means that from the ideal gas law:
nEB RT
PEB = .
NA
Now, once again integrating and doing the same steps as in part (a), we have that
Z hEB Z PEB
M gdh dP Mg PEB
− = =⇒ −(hEB − h) = ln =⇒ hEB = 425 km.
h0 RT P0 P RT P0
(d) Note that Maxwell’s distribution in spherical form is given by

3/2
mv 2

2 m
f (v)dv = v exp − sin θdθdϕ.
2πkB T 2kB T
The
R ∞ probability of the particle having a velocity more than the escape velocity will then be given by
vesc f (v)dv or in other words, integrating over all spaces gives us
3/2 Z ∞ Z π/2 Z 2π
mv 2

m 2
P (v) = v exp − sin θdθ dϕ
2πkB T vesc 2kB T 0 0
s Z ∞
2m3 mv 2

2
= 3 T3 v exp −
πkB vesc 2kB T
We need to calculate the escape velocity at the exobase which can be done with the equation
r
2GM
vesc = = 1.08 × 104 m/s
R+h
and therefore, the integral gives us
s Z ∞
2m3 mv 2

P (v) = 3 T3
2
v exp − = 2.6 · 10−3 .
πkB 1.08×104 2kB T
dN
(e) Note that flux is given by Φ = dA·dt . We can calculate dN first using the spherical Maxwell’s
distrubution and using the fact that the number of molecules coming in at a certain time is given by
nv cos θdAdt where dA is a surface element. Therefore, expanding N gives us
X
dN = nH v cos θdAdt · f (v, θ, ϕ)dvdθdϕ
69
which tells us
3/2
mv 2

m
dN = nH exp − v 3 dv · sin θ cos θdθ · dϕ · dAdt.
2πkB T 2kB T
Integrating over θ and ϕ tells us

3/2
mv 2

m
dN = nH exp − v 3 dAdt
2πkB T 2kB T
and therefore,
3/2
mv 2

dN m
dΦ = = nH exp − v 3 dv.
dAdt 2πkB T 2kB T
The flux of the escaping atoms are then given as
∞ 3/2
mv 2
Z
m
Φ= nH exp − v 3 dv = 7.5 × 1011 1/m2 .
vesc 2πkB T 2kB T
2 P . Equating this force to the mass

(f ) Note that the atmosphere produces a force of F = P A = 4πR
of the atmosphere tells us that
4πR2P
2
mg = 4πR P =⇒ m = .
g
We then write
2P
4πNA R
NA m
N= = .
MAtm MAtm g
Then note that since Nitrogen is diatomic, we have NH = 2χH N = 1.2 × 1038 which gives us our answer.
(g) The number of molecules escaping after a unit time is
ṄH = Φ(R + hEB )2 = 4.35 × 1026 1/s.
The total time required for half the atmosphere to evaporate is then given by the characteristic time
interval:
NH /2
τ= ≈ 4500 years .
ṄH
(h) Redoing the calculations from the other parts for hydrogen gives us our new answer of 9.6×1011 years.
(i) One possible reason of why there is currently still some hydrogen in the Earth’s atmosphere is that
the hydrogen inside the Earth’s water gets cycled through the hydrologic cycle.
pr 84. (a) Since the temperature is constant, we have an isothermal compression where:
P0 V0 = P1 V1 =⇒ P0 r03 = P1 r13
Since r1 = 21 r0 , we have P1 = 8P0 .
70
(b) The heat generated is equal to the negative work done:

Z
Q = −W = − P dV
Using P V = nRT , we get:

V0 3mRT0 r0
Q = −nRT ln = ln
V3 µ r3
(c) We now have an adiabatic compression since not heat is exiting the system:
3γ−3
γ γ−1 r3
P V = constant =⇒ T V = constant =⇒ T3 = T0 .
r0
(d) The final radius refers to the point in which the gravitational potential energy is comparable to the
thermal energy:
3γ−3 1
GM µ r3 RT0 r3 4−3γ
= RT0 =⇒ r4 = r3
r4 r4 µGm
and using the relationship from part (c), we get:
3γ−3
RT0 r3 4−3γ
T4 = T0
µGm
pr 85. (a) First, we assume that the process is reversible (even though this is not very likely). Then,
the work done on the liquid is:
ρ 3 2
dW = (P − P0 )4πr2 dR = dEk =⇒ (P − P0 )R2 dR = d(r ṙ )
2
The initial pressure is given by:
P0 R03
Pi =
Ri3
so using P V γ , we get:
3γ
R03

Ri
P = P0
Ri3 r
Since γ = 5/3, this simplifies to:
5 !
R0 ρ 3 2
P0 49 − 1 r2 dr = d(r ṙ )
r 2
Integrating the left hand side, we can first make the substitution β = Rr0 to simplify it to:
Z α
73 α 3

3 49 2 3 1 49
P0 R0 − β dβ = P0 R0 − + −
7 β3 2 2α2 3 3
The right hand side evaluates to zero since it starts and ends off at rest. Thus, setting this to zero, we
get the equation:
6α5 + 147 − 689α2
Making the assumption that 6α5 1, we get a quadratic:
r
147
α= =⇒ Rmin = 0.462R0 = 2.31 µm
689
71
We also know from T V γ−1 that the maximum temperature is thus:
Tmax = 6.86 × 104 K.
(b) We can apply the same differential equation. The LHS stays the same, but the RHS no longer
becomes zero. The RHS can be evaluated to:
Z α3 β̇ 2
ρR05 ρR05 3 2
d(β 3 β̇ 2 ) = (α β̇ )
0 2 2
Setting it equal, we see that:
73 α3 ρR05 3 2

3 1 49 689 49 1
P0 R0 − 2+ − = (α β̇ ) =⇒ β̇ 2 ∝ 3
− 5−
2 2α 3 3 2 6α 2α 3
This is at a maximum when:

r
6 · 5 · 49
α= = 0.596 =⇒ Rf = 2.98R0
2 · 3 · 689
(c) We make the assumption that between these two times, the speed is roughly the same. The average
radius is:
hRi = 2.645 µm
and thus plugging in this into our expression for ṙ gives ṙ = 192.77 m/s such that the total time is:
∆t = 3.48 × 10−9 s
(d) By Stefan-Boltzmann Law, we have that
Q̇ = aσ · 4πr2 T 4 .
Substituting
2
Ri
T = T0 ,
r
we result in the expression of
Q̇ = 4πaσRi8 T04 /r6 .
We require that
1 1
Q ≤ U =⇒ Q̇ ≤ U̇

5 5
and therefore, we attempt to find U̇ as well. Note that
γ
Vi
U̇ = −Pi V̇ = −Pi V̇ = −4πPi Ri5 ṙ/r3 .
V
We now can set our expression to be
4πaσRi8 T04 1 4πPi Ri5 ṙ Pi r3 ṙ

≤ · =⇒ a ≤ ≈ 0.0107
r6 5 r3 5Ri3 σT04
72
pr 86. (a) We know that

dEavg
Cv =
dT
A free electron has 3 degrees of freedom. Therefore:
3 3
Eavg = kB T =⇒ Cv = kB
2 2
(b) The total energy is

Z S Z ∞
U= Ef (E)dS = CV E 3/2 f (E)dE.
0 0
From the graph 
1
 0 ≤ E ≤ EF − 2kB T
E EF +2kB T
f (E) = + EF − 2kB T ≤ E ≤ EF + 2kB T
 4kB T 4kB T
0 E ≥ EF + 2kB T

Therefore, by integrating from all cases we have:

∞ EF −2kB T EF +2kB T ∞
− 2kB T − E
Z Z Z Z
3/2 3/2 3/2 EF
E f (E)dE = E · 1 · dE + E dE + E 3/2 · 0 · dE
0 0 EF −2kB T 4kB T EF +2kB T
Simplifying and evaluating the integral tells us

Z ∞
2 (EF − 2kB T ) 5/2 2 (EF + 2kB T ) (EF + 2kB T )5/2 − (EF − 2kB T )5/2
E 3/2 f (E)dE = +
0 5 5 × 4kB T

7/2
2 (EF + 2kB T ) − (EF − 2kB T )7/2
− .
4kB T
As kB T EF
2 ! 5/2
∞ 2E5/2 kB T 2 5/2
Z
5kB T 15 kB T 5/2 2kB T 5/2 2EF
E 3/2 f (E)dE = 1− + +EF 1+ −EF = +3 EF .
0 5 EF 2 EF EF 5 EF
Therefore: 2 !
5/2 2 kB T
U = CV EF +3
5 EF
and
U
Eavg =
N
R EF0
E 1/2 dE as electrons are below energy below EF0 at T = 0K. It is given that
R
where N = dS = CV 0
3/2
2CV EF
fermi level is not dependent on temperature. Finally, we note that EF0 = EF =⇒ N = 3 which
means !
kB T 2

3 2
Eavg = EF +3 .
2 5 EF
The heat capacity is then:
dEavg 9k 2 T
Cv = = B .
dT EF
73
pr 87. (a) The energy for spin up dipole is given by µB B while the energy for spin down dipole −µB B.
Note that by Boltzmann’s law
Probability: p = Ae−E/kB T where A is a constant.
As there are only 2 possible states for the dipoles. The total sum of both states must add up to 1 by the
rule of probability.
1
pµB + p−µB = 1 =⇒ A = −µ B/k T
e B B + eµB B/kB T
e B −µ B B/k T e B B µ B/k T
Therefore, pµB = e−µB B/kB T +eµB B/kB T
and p−µB = e−µB B/kB T +eµB B/kB T
. The average energy will be the
weighted average of both states which is the multiplied by N or in other words:

µB BpµB + (−µB Bp−µB )
Es = N
pµB + p−µB
Therefore, !
e−µB b/kB T − eµB b/kB T

µB b
Es = N µB B = −N µB B tanh
e−µB b/kB T + eµB b/kB T kB T
’
µB B
(b) For 1 kB T , we can approximate the energy to be
−N (µB B)2
Es ≈
kB T
dEs N (µB B)2

We then find that C = dT = kB T 2
Here is a purely statistical analysis of this problem. Suppose there are N↑ states and N↓ states such that
N↑ + N↓ = N and the total energy is U = µB(N − 2N↑ ). Then, the multiplicity of having N↑ states is
given by:
N!
Ω=
N↑ !N↓ !
The entropy of this is thus:
S = k ln Ω = k [ln(N !) − ln(N↑ !) − ln ((N − N↑ )!)]
Applying Stirling’s approximation, we get:
S/k = N ln N − N↑ ln(N↑ ) − (N − N↑ ) ln(N − N↑ )
The definition of temperature is given by:

1 ∂S ∂N↑ ∂S 1 ∂S
= = =−
T ∂U ∂U ∂N↑ 2µB ∂N↑
where
∂S N↑ N − U/µB
= − ln = ln
∂N↑ N − N↑ N + U/µB
N U
where we have substituted N↑ = 2 − 2µB . Solving for U , we get the desired:

µB
U = −N µB tanh
kT
74
pr 88. We consider the vessel to be spherically symmetric. Let the mass of vessel be m and the radius
of vessel be r and thickness of steel be d with d r
m
m = 4πr2 dρ ⇒ 2πr2 d =
2ρ
For the hydrogen stored

4 3 νRT
P V = νRT ⇒ P πr = νRT ⇒ P = 4πr3
3
3
For the vessel to not break

νRT 3RT ν
P · πr2 ≤ σ · 2πrd ⇒ 4πr3
πr2 ≤ σ · 2πrd ⇒ ≤ σ · 2πr2 d.
4
3
This implies that

3RT ν m 3RT νρ 3RT νρ
≤σ· =⇒ ≤ m =⇒ mmin = .
4 2ρ 2σ 2σ
75

Solutions To Jaan Kalda's Problems in Thermodynamics: With Detailed Diagrams and Walkthroughs Edition 1.2.1

Uploaded by

Copyright:

Available Formats

Solutions To Jaan Kalda's Problems in Thermodynamics: With Detailed Diagrams and Walkthroughs Edition 1.2.1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Solutions To Jaan Kalda's Problems in Thermodynamics: With Detailed Diagrams and Walkthroughs Edition 1.2.1

Uploaded by

Copyright:

Available Formats

Solutions to Jaan Kalda’s Problems in

With detailed diagrams and walkthroughs

Rakshit, Ashmit Dutta, QiLin Xue, Kushal Thaman

Updated September 11, 2020

Jonathan Qin, Anant Lunia, Ameya Deshmukh and Hermab Podar.

Structure of The Solutions Manual

1 Solutions to Heat and Motion Problems

pr 1. Applying idea 1, we have:

period of voltage oscillations). This implies that

b) Again, we separate variables. This time however, our expression becomes:

By Stefan-Boltzmann Law, we have

Plugging in our given constants gives us ≈ 290 K .

Adding all the equations, we have

the power emmited is σ6s2 T 4 . We then have that

which then tells us √

This chapter will focus on problems 9-14 of the handout.

Substituting and solving gives us

Using the fact that mN = mNA nAL = M nAL, we can get

∴ 2E ≥ 3F with equality in case of only triangles

∴ j ≥ 6 with equality in case of only triangles

3 Solutions to Adiabatic processes

This chapter will focus on problems 15-24 of the handout.

pr 16. Using Idea 11, we have:

pr 17. From the prelude, we have the work as:

The compression work W is:

so putting everything together gives:

~vair relative ground = ~vair relative plane + ~vplane relative ground

c) From problem 17, we find that

J = Fnet dt = ∆pSdt = (p1 − p2 )Sdt

∆p = (ρSx2 )v2 − (ρSx1 )v1 = (ρS)v22 dt − (ρSx1 )v12 dt

Setting these two expressions equal gives:

We can simplify further to get

then expanding tells us

b) We are given the equations

Dividing these two equations through gives us

c) In steady state, we have that

• When r = 1, we have the equation the line as

• When r = 0.8, we have the equation of the line as

ps = 1500r + 800 =⇒ 65∆T = 800(1 − r) =⇒ ∆T = 12.3(1 − r) .

d) For the boundary condition, heat dissipated through evaporation. Therefore,

ps > patm + ρgh

The sudden drop in temperature of ∆T = 83◦ C released an energy of

pr 24. a) We are given that

Similarly solving for B gives,

∆E3 = −σg` aL cos α − σg` (1 − r)aL

∆E1 + ∆E2 + ∆E3 = 0 =⇒ r (σsg − σs` ) = σg` cos α + σg` (1 − r)

Letting σsg − σs` = σg` cos α0 gives:

rσg` cos α0 = σg` cos α + σg` (1 − r) =⇒ cos α = r cos α0 − (1 − r).

Letting α0 = 110◦ and r = 0.006 gives: α = 174.9◦ ≈ 175◦ .

An alternative way of phrasing this is:

If the separation is L, we get:

R = r0 cosh(L/2r0 ) =⇒ L = 2r0 cosh−1 (R/r0 )

πσd. This force supports the weight of the droplet mg so we have:

Substituting this into (1) we get

This means that

Note that h = V /S, therefore,

From problem 29, we can conclude that the height is given by

Therefore, the change in energy is:

Therefore, the characteristic time is:

We can assume that ∆T = Tf − Ti T , allowing us to simplify our relationship to:

This mean that

At the bottom, the work done is: