One of the fundamental properties of a gas is pressure. Formally, pressure is defined as the force applied by the gas over some area (such as the wall of a container). The common English unit of pounds per square inch reflects this definition.
Pressure and Temperature
The pressure of a gas arises from the collision of the gas particles (atoms or molecules) with the walls of the container. When one of these particles hits a wall of the container there is a change of momentum as the particle strikes the wall and bounces off. The momentum of our gas particles is mv, where v is the velocity. We know from our discussion of the Maxwell-Boltzmann distribution that the kinetic energy of a gas is directly proportional to the temperature. If we recall that kinetic energy is 1/2mv2, we see that the velocity of the gas and its momentum must be proportional to the temperature. We then expect the pressure of a gas to be directly proportional to temperature and this is exactly what is found.
Pressure and Volume
If pressure arises from the collisions of particles with the sides of a container, it is reasonable that the more particles striking the wall in some period of time the higher the pressure. If we put some fixed number of gas particles inside of a container, it is reasonable that the pressure must be inversely proportional to the volume of the container since the bigger the container the smaller the number of our gas molecules that are impacting with the wall in any given time.
Measuring Pressure
In 1643 Evangelista Torricelli, a mathemetician who studied with Galileo in Florence, carried out a remarkably simple experiment, but one that could not have been easy in his day, nor obvious in design. Torricelli took a long glass tube (about 1 m) sealed off at one end and filled the tube with mercury. Fabrication of glass tubes was not a trivial exercise at this time and while mercury was available (used mostly by alchemists and metallurgists for the extraction of gold) it was quite expensive. After filling the tube, Torricelli inverted the tube (with his finger over the open end) and placed the open end of the tube in a beaker filled with mercury. Rather than remaining up in the tube, the mercury dropped part way down the tube leaving the top part of the tube empty. Now this is not surprising to us today, but to Torricelli this was troubling since he realized that there was no air in the top of this tube. He presumed, correctly, that there was a vacuum at the top of the tube. This, of course, violated the classic Greek concept of Nature abhoring a vacuum. (Actually today we recognize that Nature really does abhor a vacuum, but makes up for this in ways that Torricelli couldn't possibly have imagined.) Torricelli measured the height of the mercury column in his glass tube and observed over a period of several months that the column of mercury rose and fell in a manner that reflected the weather. If it was a clear day the column would be relatively high, but before and during a storm the column would drop. Torricelli correctly realized that the height of the mercury column must be in some way connected with the atmosphere.
Although a full understanding of the relationship of weather, air pressure, and the mercury level was to take several centuries, the use of simple barometers as Torricelli's device came to be called to predict weather was quickly adopted. Thomas Jefferson kept a life-long record of barometeric readings and was the unofficial weatherman of the Continental Congress.
We now understand the process behind Torricelli's barometer. Consider the figure to the left. The pressure exerted by the column of mercury is identical to that being exerted by the atmosphere. Remember that the top of the barometer tube is a vacuum so there is no gas pressing down on the mercury from the top. (By the way, the mercury is not being held up by the vacuum. It's being pushed up by air pressure. When you use a straw with a drink, you are reducing the pressure in the straw permitting the atmosphere to force the liquid up the straw. I know it's counterintuitive, but that's how it really works.)
The pressure of the atmosphere is simply the total pressure (weight if you will) of all of the air above the barometer stretching out to the edge of the atmosphere. When you take a barometric reading, you are essentially weighing the atmosphere.
Before we go on, we must answer a question you should have at this point. Why does the barometric pressure drop when there is a storm? The storm suggests that there is a great deal of water in the air, so shouldn't that make the atmosphere heavier? One simple way to imagine this is to think of a volume of air. If the air is dry it consists of 20% O2 and 80% N2 and tiny amounts of other things. There will be a fixed number of gas molecules in our volume. If we add water to this volume we must remove some of the other molecules. Since water has a molecular mass of 18 while the molecular mass of O2 is 32 and N2 is 28, the volume of air actually has a lower total mass when the water molecules displace the air.
In the time since Torricelli experiments with gases required scientists to establish "standard" conditions. For pressure it was decided that a standard pressure would be typical atmospheric pressure at sea level. This value is 760 mm Hg and is called 1 atmosphere (or 1 atm). In English units this is 14.7 psi (pounds per square inch), and in metric units it is 1.01 x 10SUP>5 Pascals (or 101 kPa). A Pascal has units of Newtons/m2 where Newtons are the metric unit for force.
In addition to the barometer described above, there are a number of other methods for measuring pressure. One of these is called a manometer and is a first cousin of the barometer. Manometers are useful for measuring the pressure inside of a closed system. Let's look at a closed manometer of the type used for measuring low pressures hooked up to a flask. We start with mercury filling the closed arm of the manometer. If we use the vacuum pump to remove the air from the flask the air pressure pushing down on the mercury on the flask side decreases. When the pressure in the flask gets lower than the pressure of the mercury column, the column begins to drop away from the top of the tube. We can measure the pressure in the flask by measuring the difference in the heights of the mercury in the two sides of the manometer. This kind of pressure gauge is particularly useful for pressures down to about 1 mm which is about as small a difference that can be easily observed.
A second variant on the manometer design is illustrated below. Here one end of the manometer is open to the atmosphere so the pressure in the closed chamber is being compared to atmospheric pressure. The observed mercury pressure would have to be added or subtracted from atmospheric pressure to get the pressure inside of the chamber. In the case of this example, we started with the flask being evacuated so the pressure is the vapor pressure of the liquid.
Many other pressure measuring devices exist including various bellows and diaphram designs, hollow metal spirals that expand or contract with pressure, electronic guages that measure the conductivity of the gas remaining in a high vacuum, and pizeoelectric materials that change resistance when exposed to pressure. These many kinds of guages permit us to measure pressures from about 10-12mm Hg to many thousands of atmospheres.
Why Mercury?
Mercury has been known since prehistoric times and its use in metallurgy and gold extraction was discovered well before the Egyptian era. Mercury is the densest of all common liquids (liquid lead and uranium would be denser, but we don't encounter them very often) with a density of 13.6 g/ml. If we wished to substitute water for mercury in the barometer we would need a column of water 13.6 times that of mercury. One atmosphere would then be:
13.6 x 760 mm = 1.03 x 104mm or 10.3 meters! In English units that is about 33 ft.
Old fashioned water pumps which drew water to the surface by creating a vacuum above the water could only draw water up 33 ft. Actually, since the water has a small pressure and the seals were never really very good, these old pumps probably wouldn't have been much good beyond about 30 ft. In a populated area (particularly one with outhouses) this would put the water source dangerously close to sources of contamination. Most modern wells are deep wells and the water is pumped to the surface by placing a sealed electrical pump at the base of the well and pumping the water up. With such a well only the quality of the seals and the pump limit the height to which the water can be pumped.
In contrast to barometers for which mercury is the only reasonable liquid, manometers may be filled with other liquids. We usually select organic liquids with very low vapor pressures for these applications. For example, dibutyl phthalate is frequently used as a fluid in manometers such as those found on boiler condensors. Here the lower density of the organic liquid (butyl phthalate has a density of about 1.05 g/ml) is an advantage since a 1 mm difference in height in a mercury manometer corresponds to a 13 mm difference in height of dibutyl phthalate. It's much easier to read a difference of 13 mm than it is to read one of only 1 mm.