The Second Law of Thermodynamics

A Quick Look by Albert Hines

A good understanding of this topic requires a thorough knowledge of statistical analysis, quantum theory, and of thermodynamics. However, we may look at a few simplified concepts which will help to illustrate the property called "entropy" with which the second law of thermodynamics deals. The second law of thermodynamics in essence states this: "There is property called entropy which, in a closed environment can only increase. Equivalently, in an open system, this property, entropy, can decrease only as little or less than the increase of entropy of the surrroundings of this system."

So what is this stuff called entropy? Entropy was developed from statisical analysis of molecules in order to put a value upon the probability of a substance being in a particular state. This is the crucial point of entropy. It measures the likelihood that a group of molecules, each of which may have a different speed, rotation, energy level, etc., will have their overall average character (temperature, pressure, enthalpy, specific volume, viscosity, conductivity, etc.) at a certain value. The second law of thermodynamics dictates that the average properties (pressure, temperature, etc.) of a "lump" of matter will be at the most probable state. Time for an example:

Example 1:
A casino operator has one hundred slot machines. The probability of a payoff for each slot machine is about one to one thousand. The machine costs a quarter to play and pays off $100. The machine may pay off at random times and each machine may have a slightly different payoff, but the average property, the profit of the operator, is in the long run sure. The operator may lose money one day, but in the long run, his profit is sure.

The concept of entropy is very similar. Every molecule is flying, spinning, vibrating differently, but the overall properties of the billions of molecules appear, as far as we can measure, constant. It is statistically possible that all the molecules of bowl of soup bump into one of your chunks of potato and that it flies out into space! But the probability is so remote, that it has never been observed experimentally. Entropy is a game of chance. But with matter consisting of so vastly many molecules, the odds win out almost every time. When the odds don't win out, we call it a "miracle," - something which is not supposed to happen scientifically. But in the course of natural science, the odds being immeasurable, that which happens is the most likely statistically.

"Yes, but what is entropy? What does it feel like? Explain it to me in everyday terminology, give me fifty cents worth of entropy!" says the reader. Well, unfortunately no one can answer that question exactly. But if there is any consolation, no one can tell you what "energy" is either. Since we use the term so frequently to describe our personal understanding of energy, we are familiar with it. We see a fast car or a hot stove and say "that sure has energy!" But what is energy? We see some of its results, but cannot describe the stuff itself. The same is true of entropy. If we were to begin to use the word "entropy" frequently (every time a firecracker explodes or we turn on an electical heater, we learned to say "Well, that surely created a bucket load of entropy!") then we would be familiar also with entropy. But the accurate way this author knows to describe it is that it is a measure of the chance (probability) of given molecules to exist in a certain state.

While from a thermodynamic point of view, we can study entropy without considering probability or statistics at all, much insight is gained by looking into the true nature of the beast. When we do consider the second law without the background behind it, we tend to always use one of the basic results that the entropy must increase, therefore, such and such will happen. And this is not totally invalid, but can be used to generate false deductions. Many great men have helped to build upon the knowledge of the concept of entropy - Einstein, Boltzmann, Planck, Bose, Clausius, Kelvin, Carnot, Maxwell, de Broglie, Tribus, Shannon, literally too many contributors to name. But in the first third of the twentieth century, the probability of molecular states was firmly turned into a mathematical function describing this stuff called entropy. When this new property - entropy - was analyzed, the following were among the results (some were theorized earlier, but put on a solid theoretical basis after a second law analysis)

-> The property called 'entropy' in any system can, under ideal conditions, remain constant. In practice, any change made in the system results in an increase in the entropy. Alternately, it can decrease by an amount smaller than that of something in contact with the system (i.e. entropy is 'squirted' into the surroundings). Once entropy is generated, no known process can cause it to be destroyed. This is known as the concept of 'irreversibility', since the overall 'entropy level' can never be restored to its original condition (this has been described colloquially, saying "you can't unscramble an egg.").

-> Without work input, thermal energy transfers (heat goes) in only from a 'hot' region to a 'cold' region, never from cold to hot.

-> Any cyclic device (one which returns to the same state after performing a task) which either produces or consumes work must operate from a temperature difference.

Example:
an internal combustion engine must have a combustion temperature above that of the immediate surroundings in order to produce work. A steam generator must have a heater and/or cooler to produce work.

-> Due to the huge number of molecules in a macroscopic world, whenever the opportunity arises, matter will always go from a state of less probability to a state of greater probability (not 'order', but 'probability'). Any exception would be considered a "miracle". It is important to stress the difference between probability and order, because we normally hear entropy referred to as that "curse" that makes things go from a state of order to a state of dis- order. This is dead wrong, because 'order' cannot be measured and hence, can be used to "prove" falsehood based upon a flawed definition of order.

-> There is an upper limit to the efficiency of any device. This limit is equal to the efficiency which results from a device that accomplishes the same task without increasing entropy. Any actual device will be less efficient (require more energy) in accomplishing the task. This excess energy expended will always eventually be converted into a rise in temperature of something (the surroundings).

-> Work energy is more 'precious' than heat energy. Work is harder to obtain, easier to loose, and more productive than heat. Work may easily be converted into heat, but heat is difficult to convert into work.

-> There is a limit to the amount of work obtainable from any given situation (less than the total energy e=mc2). This limit is equal to the amount of work done in a manner which does not generate (create) entropy. Once some of this 'available work' is lost or destroyed, it is gone forever. This is called 'lost work' or 'irreversibility.'

-> Some of the factors which contribute to lost work are: friction, unrestrained expansions (such as releasing a high pressure gas into the air), mixing of unequal substances, hysteresis (losses due to materials not recovering perfectly), electrical resistance heating, contact of two materials at different temperatures, combustion, shock waves, non-equilibrium, and permanent bending or deformation. Any time one or more of these effects are involved in a process, entropy is produced, work is lost, and efficiency drops.

-> The amount of lost work caused by any process is directly proportional to the rate of entropy creation. The more entropy is produced, the more work is lost. While the terms cannot be used interchangeably, they are inseparably coupled.

Example:
We may create some entropy by mixing water into a bowl of syrup. But when we do, we forever lose the opportunity to gain some work by allowing a semi- permeable membrane to raise the level of the syrup by osmotic diffusion of the water into the sugar solution. We can only regain some of the work by input of more work into the system (centrifugal separation for example). The amount of work input required is always at least that of the amount gained for such an attempt.

-> There is a temperature below which it is impossible to achieve. This temperature is referred to as 'absolute zero.' This absolute zero has some very interesting and strange characteristics that I won't go into. If this temperature could ever be obtained, it could not be sustained.

-> The potential for extracting work from a given situation (let us call it a given 'system') can either decrease or remain constant, but never increase without energy input. From observations, it is found that the more the system is disturbed (changed in any way from its original state) the more likely it is to lose its work potential (and thus generate entropy). There are a few processes which do not lose measurable work potential, however, these are the exception to the norm. (The term 'work potential' should not be confused with the term 'potential energy' which is a gravitational work storage and has little to do with the present discussion)

Example:
If we consider a pile of rocks, we could lower every rock carefully to the ground using a lever and thereby raising a weight (thus storing energy). But if we kick over the rocks, they will fall without utilizing this work, and this lost work can never be restored (i.e. we will have to spend at least as much work to restore it to its original condition).

Example:
It is easy to arrange rocks so that they appear in a very orderly geometric pattern on the floor. However, a jumbled, apparently disorderly pile of rocks would have less entropy, because work could be extracted by lowering them to the ground. Thus, orderliness is a subjective and fallible measure of entropy. * Light My Path Publications *
Back