O2.

Lab O2. Variation of Atmospheric Pressure with Altitude

Introduction
In this lab you will investigate the ideal gas law. By moving a plunger in a sealed
container of air, you will find the variation of pressure with volume. By warming the air with
your hands, you will investigate the variation of pressure with temperature. Then, by riding up
and down in the elevator with your apparatus, you will investigate the variation of pressure with
altitude.

We all live near the bottom of an ocean of air. At sea level, the weight of the air
overhead presses on us with a pressure of ≅ 105 N 2 = 14.7 lbs 2 (12.5 psi in Boulder). We are
m in
not aware of the this great weight because the air presses on us from all sides, even from our
insides (due to the air in our lungs).
At higher altitudes, there is less air and less weight overhead, and the pressure is less.
Also, because air is readily compressible, the lower layers of air are compressed by the weight of
the air above. Thus, the pressure and density of air decrease at higher altitudes. That’s why a
helium balloon rises: the pressure on the underside of the balloon is greater than the pressure on
the top.
Air pressure falls with altitude exponentially,
p
⎛ mgh ⎞ p
−⎜ ⎟ o
⎝ kT ⎠
(1) p = po e
p
o
e h
ho
where h is the height above a level where the pressure is po, m is the average mass of an air
. × 10 −23 J / K ) and T is the temperature in degree
molecule, k is the Boltzmann constant ( k = 138
Kelvin (K). Note that both mgh and kT have the units of energy, so the exponent is
dimensionless. It is a remarkable characteristic of the exponential function that eq’n(1) is true
regardless of where we set the zero of h, so long as the pressure at h = 0 is po. This equation
(derived in the Appendix A) is not quite correct, because its derivation assumes that the
atmosphere is isothermal when in fact, the temperature of the air varies considerably with
altitude.
Eq’n (1) can be rewritten
− ⎛⎜ h h ⎞⎟
(2) p= poe ⎝ 0 ⎠ ,

where h o = kT mg is a characteristic height , called the scale height, of the atmosphere. The
scale height is the height increase which reduces the air pressure by a factor of
1 1
= ≅ 0.368. If you started at sea level (p = 1atm) and climbed a mountain with a
e 2.718...
height ho, the pressure at the peak would be 1 atm / e ≅ 0.37 atm . At an altitude of 2ho , the
pressure would be 1atm / e2 ≅ 0.135 atm.

Fall 2004
 O2.2

If the pressure change ∆p can be measured over some small height change ∆h, then the
scale height ho can be determined.

∆p dp ⎛ 1⎞ p ∆h
≅ = po ⎜ − ⎟ e− h ho = − o, ⇒ ho = −po . (3)
∆h dh h = 0 ⎝ ho ⎠ ho ∆p
h=0

In this experiment, you will use a differential pressure transducer to measure the change
in air pressure during a elevator ride to the 10th floor of Gamow Tower ( ∆h = 32.2 m , according
to CU records). You can feel the pressure change in you ears as you ride the elevator, so it is not
surprising that it can be measured with some precision.
The principle of operation of the pressure transducer is simple. Two chambers within the
device are separated by a flexible metal membrane. If there is a pressure difference between the
two chambers, the membrane will flex slightly, bowing out from the high-pressure side. Small
movements of the membrane are very accurately determined by measuring the capacitance
between the metal membrane and a fixed metal surface nearby. The transducer output is a DC
voltage ( 0 → ±1V ) which is proportional to the pressure difference between the two ports; when
the pressures are the same, the output is 0V.

metal
membrane

lower p higher p

fixed metal
surface

The two side chambers are labeled A and B on the instrument. When A has the higher pressure,
the voltage output is positive. When A has the lower pressure, the voltage output is negative.

There is an unfortunate abundance of units for pressure. One standard atmosphere (at sea
level) in various units is:

1 atm = 760 torr = 406.8 inches H 2 O = 1.013 × 105 Pa = 1.013 bar = 1013 mbar.

A torr is a millimeter of mercury (mm Hg). A pascal (Pa) is 1 newton/meter2 , the SI unit of
pressure. Here in Boulder, at an altitude of 5400 ft ≅ 1650 m, the pressure is about 840 mbar ≅
630 torr, but it varies by ± 20 mbar, depending on the weather. You can read the current air
pressure (barometric pressure) in the 1140 Lab, in cm Hg, with the mercury barometer on the
wall (Ask the TA where it is).

Your first task in this lab will be to calibrate the pressure transducer by applying a
known pressure difference and measuring the resulting voltage output. To do this, you will
connect one side of the transducer to a sealed flask of known volume V, which can be varied

Fall 2004
 O2.3

with a piston over a range ∆V . As the volume is changed, the pressure p will change according
to the ideal gas law

(4) pV = NkT ,

where N is the number of molecules in the volume, k is the Boltzmann constant, and T is the
temperature in oK. (Air is not an ideal gas, but it is close enough for our purposes.)
For a fixed number of N of molecules, the pressure change ∆p resulting from a volume
change ∆V depends on whether the change is isothermal ( ∆T = 0 ) or adiabatic (no heat added
or lost, ∆Q = 0 ). The relationships between ∆p and ∆V , for the two cases, are:

∆p ∆V
(5) =− (isothermal, ∆T=0)
p V

∆p 4 ∆V
(6) =− (adiabatic, ∆Q=0)
p 3 V

The first equation (5) can be quickly derived from the ideal gas law by assuming that T
and N are constant. The second equation (6) is not true for a monatomic gas, but is true for a
diatomic gas, such as air; equation(6) is derived in the Appendix B.
In Part I of this experiment, the pressure change will be nearly isothermal, so we will use
equation (5).

Procedure

Part 1. Calibration of the pressure transducer

In this part, you will determine the calibration constant C of the pressure transducer,
which is the pressure difference per voltage output C = ∆p ∆volts . One side of the transducer
will be left open to air and the other side connected to a glass Erlenmeyer flask and a plastic
syringe, as shown below.

Fall 2004
 O2.4

valve (shut position)

0.045V

50 cc syringe

pressure 1 volt out max

flask transducer
~ 1300cc

port to room air

The valve on the flask can be opened to allow the pressures to equilibrate or closed to vary the
pressure in the flask. The volume of the (flask + connecting hoses+ syringe with plunger at 30cc
mark) is approximately 1300 cc. The precise volume is marked on the flask.
Turn the transducer on by plugging in the wires to its batteries. Open the system to room
air with the valve, and check that the voltage output is close to 0.000V. Record any offset from
zero. With the flask open to room air, set the syringe plunger at some point near the middle of
its range, then isolate the system (flask+hose+syringe) from the room by closing the valve.
When you change the volume of the enclosed air by pulling or pushing on the syringe, the piston
does work on the enclosed air, resulting in a temperature change of the air. However, after
about 30 seconds, the temperature of the air inside the flask will equilibrate with outside
temperature, due to thermal conduction through the glass. Consequently, the final pressure
change is isothermal, although the initial pressure change is somewhat adiabatic. Because of
these temperature changes, you will observe that, immediately after you change the volume, the
magnitude of the output voltage will reach a maximum, but then it will fall as the temperature
equilibrates, reaching a stable value after less than a minute.
Pushing the syringe plunger in decreases the volume (∆V < 0); pulling it out increases the
volume. Change the volume over the range -15 cc to +15 cc in ≈5 cc steps, and record the
equilibrium voltage output for each volume. Using eq’n (5), convert the volume changes
(∆volume) to pressure changes (∆p), and plot the transducer output in volts vs. the pressure
change ∆p in torr. To use eq’n(5), you will need the volume V (which is marked on the flask)
and the room pressure p (from the mercury barometer – remember to convert from cm Hg to
torr). Your measured ∆p's for positive ∆V's should be nearly the same magnitude as those for
the negative ∆V's. If they are not, it is probably due to temperature drift during your
measurements. Repeat until you get consistent results.
From your data, determine the calibration factor C = ∆pressure / ∆volts (torr/volt) of the
transducer. Do this by computing C = ∆p / ∆volts (the change from zero) for each of your 6
data points and averaging. Also, compute the standard deviation of the mean of your values.

∑ ( Ci − C)
2
1 σ
(7) C= ∑ Ci ,
N i
σ= i
N −1
, σ mean =
N
.

Fall 2004
 O2.5

[In your report, be careful not to use the symbol ∆V , unless it is clear whether you mean
∆( volume) or ∆ (volts) . Check with your TA that you have approximately the right value for
the calibration constant C, before proceeding to Part 2.]

Part 2. Change of pressure with temperature.

According to the ideal gas law (4), if the volume V and number of molecules N are fixed,
then the pressure p is proportional to the temperature T, and the fractional change in temperature
is equal to the fractional change in pressure

∆T ∆p
(8) = , (T in Kelvin).
T p

After opening the valve on the flask to equilibrate the pressures inside and out, close the
valve to seal the system. Then warm the flask by placing your hands around it for about 20
seconds. You will see the output voltage start to change after only a few seconds. Record how
long you had your hands on the flask, and record the total change in output voltage. From these
data, compute the rise in pressure and the rise in temperature of the air in the flask.

Part 3. Change of pressure with altitude

To prepare for an elevator ride, disconnect the hose from the glass flask, transfer the
transducer, batteries, voltmeter, etc. to the cart , and attach the hose to the thermos bottle on the
cart. The thermos bottle has excellent thermal insulation, insuring that the temperature of the air
inside will not vary significantly as you transport the apparatus around the building. With the
valve open, depress the plunger on the syringe completely to empty the syringe of air. Leave the
valve open so that the pressures on both sides of the transducer are equal. Go to the Gamow
Tower elevator. (To get there, exit the 1140 lab, turn left and go down the corridor, through the
double doors. The elevator is by the vending machines.)

Fall 2004
 O2.6

valve (shut)

0.045V

50 cc syringe

pressure 1 volt out max

transducer

thermos port to room air

Just before getting on the elevator with your apparatus, seal the system and note the
voltage output (it should be very close to 0V). Ride to the 10th floor, record the voltage there,
then return to the basement and record the final voltage. When you return to the basement, the
final voltage should return to close to the initial value. If your final voltage and initial voltage in
the basement differ by more that about 0.020 V, something went wrong − try the trip again. The
quicker you can make the trip, the better.
Take the average voltage change from your trip up and trip down and compute the
pressure change between the basement and the 10th floor. The height change is ∆h = 32.2 m .
With your data and eq’n (3), compute the scale height ho of the atmosphere. Calculate the
uncertainty ∆h o .

⎛ mgh ⎞
−⎜ ⎟
⎝ kT ⎠
Appendix A: Derivation of p = p o e

A layer of air has weight; it is held up because the pressure below is greater than the
pressure above. Consider a layer of air of thickness ∆h , area A, and density ρ; its volume is
A ⋅ ∆h and its weight is

(A.1) w = mg = ρ ⋅ A ⋅ ∆h ⋅ g

Fall 2004
 O2.7

p(h + ∆ h)

∆h
h
area A

p(h)

This weight is supported by the net force on the layer due to both the pressure above and the
pressure below:

(A.2) p ( h ) ⋅ A − p ( h + ∆h ) ⋅ A = ρ ⋅ A ⋅ ∆h ⋅ g ,

or,

dp
(A.3) p ( h + ∆h ) − p ( h ) = − ρ ⋅ ∆h ⋅ g ⇒ = −ρg.
dh

If the density is constant, we have

(A.4) ∆p = ρ g ∆h .

vacuum This equation is the principle of operation of barometers. The pressure due
to the weight of a column of fluid (usually mercury) with density ρ and
height ∆h, is balanced by atmospheric pressure. From (A.4), one can show
that the height of a column of mercury in a mercury barometer at STP
∆h (standard temperature and pressure) is 760 mm.

The density of air is not constant, however; it varies with pressure.

From the ideal gas law,

N Nm kT kT
(A.5) p = kT = = ρ ,
V V m m

where m is the average mass of an air molecule and ρ is the mass density of air. Hence,

⎛ m⎞
(A.6) ρ = ⎜ ⎟p .
⎝ kT ⎠

Fall 2004
 O2.8

Inserting this into (A.3) yields

dp ⎛ mg ⎞
(A.7) = -⎜ ⎟p
dh ⎝ kT ⎠
The solution of this differential equation is

⎛ mgh ⎞
−⎜ ⎟
⎝ kT ⎠
(A.8) p = po e .

This equation is only approximately true because we have made two assumptions which are not
strictly correct: air is an ideal gas and the atmosphere is isothermal.

Appendix B: Derivation of pressure change during an adiabatic volume change.

The pressure change due to an adiabatic volume change depends on whether the gas in
question is monatomic, like helium or neon, or diatomic, like the main components or air,
oxygen and nitrogen. Assuming that the gas container is thermally isolated from the outside
world, then no heat is exchanged during the volume change. However, the piston does work on
the gas as the volume changes, so energy, in the form of work, is exchanged with the outside and
the temperature of the gas varies.

We first derive the temperature change ∆T . We will need the First Law of
Thermodynamics:

(B.1) ∆
N U = − p ∆V + ∆
N Q


energy added work done on the system heat added

and an expression for the total internal energy of a gas of N atoms or molecules. For the case of
a monoatomic ideal gas:

(B.2) U = N × KE per atom = 3 N kT .

2

While for a diatomic gas,

(B.3) U = N × ( KE + rotational energy) per molecule = 3N kT .

Equations (B.2) and (B.3) can both be written as U = b N kT , where b is either 3/2 or 3.
Assuming ∆Q = 0 , (B.1) , (B.2), and (B.3) yields

p ∆V
(B.4) b N k ∆T = − p ∆V ⇒ ∆T = - 1 ∆V = - 1 T .
b Nk b V

Fall 2004
 O2.9

So, for an adiabatic change of volume, we have

∆T ∆V
(B.5) = -1 .
T b V

NkT
Now we can compute the pressure change. Starting from the ideal gas law, p = , and
V
differentiating, we have

Nk NkT p p
(B.6) ∆p = ∆T - ∆V = ∆T - ∆V .
V V2 T V

Substituting (B.5) into (B.6) gives

(B.7) ∆p =
p ⎛ 1 ∆V ⎞
T⎝
⎜− b T ⎟ -
V⎠
p
V
∆V = - 1 + 1 p
b (
∆V
V
) 3 (
= − 5 or 4 p
3 )∆V
V
,

The coefficient is 5/3 for a monatomic gas and 4/3 for a diatomic gas.

PreLab Questions:

1. The ideal gas law states pV = NkT . If N and T are fixed, then pressure is inversely
C
proportional to volume: p = , where C is a constant. Consider a large volume V with fixed N
V
and T. If V changes by a small amount ∆V , what is ∆p , the change in p? (Hint: remember
df
calculus: if f = f ( x), then δf = δx .) Use this result to prove eq’n (5).
dx

2. On planet X (not earth), the pressure at sea level is 300 torr. The pressure falls by 1.0 torr
when the height is increased by 60 m. What is the scale height ho on planet X?

3. What is the average air pressure (barometric pressure) in Boulder in units of mm Hg?

4. Compute the scale height of the atmosphere, ho . Assume that the temperature of the
atmosphere is 300K and the average mass of an air molecule is m = 4.78 × 10 −23 gram.
Give the answer in both meters and miles. (Hint: Be sure to use a consistent set of units in your
calculation.)

5. Assume the scale height of the atmosphere is ho = 4.00 km (it is actually larger than this). By
what factor does the pressure fall as your height is increased from 2 km to 6 km? From 24 km to
32 km? [Hint: you can answer this question with a very simple calculation (simple enough to do

Fall 2004
 O2.10

in your head). If you answer this question with a very complicated calculation, you will receive
only partial credit.]

6. In this problem, you can assume that the scale height of the earth’s atmosphere is about 8 km
.You return from a hike and discover that your empty plastic 1 liter water bottle has a dent on
one side. When you take off the lid, air rushes in and the dent disappears. If you hiked down
from a pressure of 560 torr to a pressure of 620 torr, by approximately how much did the volume
of the 1 liter (1000 cc) bottle decrease? (Assume no temperature change and that the plastic does
not resist the volume change.)

7. In this experiment, when you use eq’n (3) to compute ho from your measured values of ∆h
and ∆V , how do you know the value of po?

8. Use eq’n (A.4) to show that the height of a column of mercury in a barometer is 760mm at
STP. The density of mercury is 13.6 grams/cm3. (Hint: In a barometer, ∆p is 1 atmosphere, but
you have to use a consistent system of units.)

9. Prove eq’n (8). Start with the ideal gas law and the assumption of constant N and V.

10. In part 1, you are to plot the voltage output of the pressure transducer vs. the pressure
change. Sketch what this graph should look like. How is this graph related to the calibration
constant C of the pressure transducer?

Fall 2004