Could heat spontaneously flow out of these fried eggs, thus “unfrying” them and make ing them cooler. and flow back into the hot fiying pan? Such a process would conserve awed by the frst law of thermodynamics, But such a process isin ict impossible because it would volte the second law of thermoxiynamies: heat anot by iisell flow from one body into hotter boy. Is it ever possible for heat to flow from a cold body into a hot 754 THE SECOND LAW OF THERMODYNAMICS Mi certian roses cee vty in en dieton bet not the opposite. For example, heat always flows from a hot body to a cooler body, never the reverse. Heat flow from a cool body to a hot body would not vio- late the First law of thermody would be conserved. But it doesn’t happen in nature, Why not? As mechanical energy completely into heat; this happens every time we use a car's brakes to stop it, Going convert heat partially into mechanical ene example.) But e building a machine that converts heat completely into mechanical energy. Again why not? other example, note that itis easy to convert in the reverse direction, there are plenty of devices that y. (An automobile engine is an n the cleverest would-be inventors have never succeeded in The answer to both of these questions has to do with the directions of thermo~ ‘dynamic processes and is called the second law of thermodynamics, This law places fundamental limitations on the efficiency of an engine or a power plant. I also places limitations on the minimum ene input needed to operate are ator. So the second! law is directly relevant for many important practical problems. We catn also state the second law in terms of the concept of entropy, a quanti tative measure of the degree of disorder or randomness of a system. The idea of entropy helps explain why ink mixed with water never spontaneously unmixes and why we never observe a host of other seem possible processes.20.1 | Directions of Thermadynamie Processes 20.1 | Directions of Thermodynamic Processes ‘Thermodynamic processes that occur in nature are all irreversible processes. ‘These are processes that proceed spontaneously in one direction but not the other (Fig. 20.18). The flow of heat from 2 hot body to a cooler body is irreversible, as is the free expansion of a gas discussed in Sections 19.3 and 19.6, Sliding a book ‘across a table converts mechanical energy into heat by friction; this process i ire versible, for no one has ever observed the reverse process (in which a book ini- tially at rest on the table would spontancously start moving and the table and book ‘would cool down), Our main topic for this chapter is the second law of thermody namics, which determines the preferred direction for such processes, Despite this preferred direction for every natural process, we can think of class of idealized processes that would be reversible. A system that undergoes such ‘an idealized reversible process is allways very close to being in thermodynamic {equilibrium within itself and with its surroundings. Any change of state that takes place can then be reversed (made to go the other way) by making only an infini f in the conditions of the system. For example, heat low between ‘wo bodies whose temperatures differ only infinitesimally can be reversed by ly very small change in one temperature or the other (Fig. 20.1b). Reversible processes are thus equilibrium processes, with the system always in thermodynamic equilibrium, Of course, if a system were truly in thermody- namic equilibrium, no change of state would take place. Heat would not flow into ‘oF out of a system with truly uniform temperature throughout, and a system that is truly in mechanical equilibrium would not expand and do work a roundings, A reversible process is an idealization that can never be precisely attained in the real world. But by making the temperature gradients and the pre ‘sure differences in the substance very small, we can keep the system very close 10 equilibrium states and make the process nearly reversible, That's why we call a reversible process a quusi-equilibrium process. By contrast, heat flow with finite temp ‘28s, and conversion of work to heat by friction are all irreversible processes: no small change in conditions could make any of them go the other way. They are also all nonequilibrium provesses, in that the system is not in thermodynamic ‘equilibrium at any point until the end of the process ‘There is relationship between the direction of a process and the disorder or randomness of the resulting state, For example, imagine a tedious sorting job, such as alphabetizing a thousand book titles written on file cards. Throw the alphabetized stack of cans into the air. Do they come down in alphabetical order? ‘Alas, no: their tendency is to come down in a random or disordered state. In the free expansion of a gas discussed in Sections 19.3 and 19.6, the air is more disot- dered after it has expanded into the entire box than when it was confined in one side, just as your clothes are more disordered when scattered all over your floor than when confined 10 your closet Similarly, macroscopic kinetic energy is energy associated with organized, ‘coordinated motions of many molecules, but heat transfer involves changes in energy of random, disordered molecular motion. Therefore conversion of mechanical energy into heat involves an increase of randomness or disorder Inthe following sections we will introduce the second law of thermodynamics, by considering two broad classes of devices: heat engines, which are partly suc- cessful in converting heat into work, and refrigerators, which are partly success ful in transporting heat from cooler to hotter bodies. rature difference, free expansion of & Meta box ini t 10°C 755 Metal box at ri Ice we eet cy Mea hos st Metal box at we we guid Ligaid wc re Inoversble Reverie proses: process 20.4 (a) Ablock of ice melts irreversibly ‘when we pkace it in hot meal ox. Heat Flows from the bo int the ice and water, never the reverse (b) We ean melt ice reversibly ina metal box at O°C. By raisin ‘or lowering the temperature of the box by an infinitesimal amount, we ean make heat Flow cut of the bo int the ie (thus melt ing i) or into the box from the water (hus756 Activ prises 8.12. Cyclic Process—strategies 8.13 Cyclic Process—Problems 20.2 All motorized vehicles (except electric-powered vehicles) use heat engines for propulsion ‘CHAPTER 20 | The Second Law of Thermodynamics Your left and right hands are normally atthe same temperature, just like the metal box and ice in Fig. 20.16. Is rubbing your hands together to warm them a reversible process? Explain, 20.2 | Heat Engines ‘The essence of our technological society is th other than muscle power. Sometimes, mechanical energy is directly available: water power is an example. But most of our energy comes from the burning of fossil fuels (coal, oil, and gas) and from nuclear reactions. They supply ener, that is transferred as heat. This is directly useful for heating buildings, for cook: ing, and for chemical processing. but to operate a machine or propel a vehicle, we need mechanical energy. Thus it's important to know how to take heat from at source and convert as ‘much of it as possible into mechanical energy or work, This is what happens in gasoline engines in automobiles, jet engines in airplanes, steam turbines in elec: trie power plants, and many other systems. Closely related processes occur in the animal kingdom: food energy is “burned” (that is, carbohydrates combine with oxygen to yield water, carbon dioxide, and energy) and partly converted to mechanical energy as an animal's muscles do work on its surroundings. Any device that transforms heat partly into work or mechanical energy is called a heat engine. Usually, a quantity of matter inside the engine undergoes inflow and outflow of heat, expansion and compression, and sometimes change of phase. We call this matter the working substance of the engine. In internal- ‘combustion engines the working substance is a mixture of air and fuel: ina steam turbine itis water (Fig. 20.2), The simplest kind of engine to analyze is one in which the working substance undergoes a eyelie process, a sequence of provesses that eventually leaves the substance in the same state in which it started. Ina steam turbine the water is recy cled and used over and over. Internal-combustion engines do not use the same air over and over, but we can still analyze them in terms of eyelic processes that approximate their actual operation, Al heat engines absorb heat from a source at a relatively high temperature, perform some mechanical work, and discard of reject some heat at a lower lemperature. As far as the engine is concerned, the discarded heat is wasted. In internal-combustion engines the waste heat is that discarded in the hot exhaust gases and the cooling system; in a steam turbine it isthe heat that must flow out ‘of the used steam to condense and recycle the water. When a system is carried through a eyelie process, its initial and final internal energies are equal. For any cyclic process, the first law of thermodynamics requires that u ‘That is, the net heat flowing into the engine ina ‘done by the engine When we analyze heat engines, it helps to think of two bodies with which the working substance of the engine can interact. One of these, called the hor reser voir, represents the heat source: it can give the working Substance large amounts of heat at a constant temperature 7,, without appreciably changing its own tem: -U=0=9-W 0 @ W cyelic process equals the net work202 | Heat Engines perature, The other body, called the cold reservoir, can absorb large amounts of discarded heat from the engine at a constant lower temperature T.. In a steam: turbine system the flames and hot gases inthe boiler are the hot reservoir, and the ccold water and air used to condense and cool the used steam are the cold reser We denote the quantities of heat transferred from the hot and cold reservoirs as, Qj and Qc. respectively. A quantity of heat Q is positive when heat is transferred into the working substance and is negative when heat leaves the working sub- stance. Thus in a heat engine, Qy is positive but Qe is negative, representing heat leaving the working substance. This sign convention is consistent with the rules \we stated in Section 19.1; we will continue fo use those rules here. Frequently, it clarifies the relationships to state them in terms of the absolute values of the Q's and W’s because absolute values are always positive. When we do this, our nota tion will show it explicitly. We can represent the energy transformations in a heat engine by the enerey= flow diagram of Fig. 20.3. The engine itself is represented by the circle. The at Qj supplied to the engine by the hot reservoir is proportional to the width of the incoming “pipeline” at the top of the diagram. The width of the ‘outgoing pipeline at the bottom is proportional to the magnitude |Qc| ofthe heat rejected in the exhaust. The branch line to the right represents the portion of the heat supplied that the engine ecnverts to mechanical work, W. When an engine repeats the same eyele over and over, Qy and Qe represent the «quantities of heat absorhed and rejected by the engine during one cyele: Qy is pos itive, and Qe is negative, The net heat Q absorbed per eyele is = Oy + Oe = |Oul ~ Och 20.1) The useful output of the engine is the net work W done by the working substance. From the first kas, ‘amount of het W= Q=Qy + Oc = |ul ~ 10 (202) Ideally, se would ike fo convert al the heat Qy ito work: in tha ease we would have Oy = W and Qe ~ 0. Experience shows that this is impossible; there is lays some heat wasted, and Qe ivnever sero, We define the thermal efficiency 1. denoted by e, the quotient w © OW The thermal efficiency « represents the Fraction of Qy that is converted to work. To put it another way, is what you got divided by what you pay for. This isalvays fess than unity an al-¢oo-familiar experience! In terms ofthe Now dae tram of Fig. 20.3, the most efficient engine isthe one for which the branch Pipeline representing the work output is as wide as possible and the exhaust pipeline representing the eat thrown away sas narrow as possible When ke substitute the two expressions for W given by Eq, (20.2) into Eq, (20,3), we get the following equivalent expressions fore ofanen e (203) W_), Ge Qe Me oy 1 Qu Ou 1 | (ihermal efficiency of an engine) 20.4) Note that ¢ isa quotient of two energy quantities and thus isa pure number, with- ‘out units, OF course, we must always express W, Qj, and Qe in the same units 757 20.3 Schematic energy-flow diag heat engine.758 eee roret Heat Engines Problems involving heat engines are, First and foremost, prob: Jems in the first law of thermodynamies. Hence the Problem Solving Strategy in Section 19.4 is equally useful throughout the prevent chapter and we suggest thal you reread it IDENTIFY she velevanr concepts: A beat engine is any device that converts work partially to hea, as shown schematically in Fig. 20.3. We will seein Section 20.4 that a refrigerator is essen tally a heat engine running in reverse, so many of the same con= cepts aly. SET UP she problem as suggested in the Problem cay in Section 19.4, Equation (20.4) is useful in ‘whieh the therm efficiency ofthe engine is relevant Solving Stat- ations for EXECUTE she solution as follows: |. Be very careful withthe sign conventions for W and the Various Q's. W is positive when the system expands and Analyzing a heat engine A gaoline engine in large truck takes in 10,000 J of heat and delivers 2000 J of mechanical work per eyele, The heat is obained by burning gasoline with heat of combustion L, = 5.0 > HS. 8) What i the thermal efficieney ofthis engine? b) How much heat is discarded in each eycle? c) How much gasoline is burned in each cycle? d) Ifthe engine goes through 25 eycles per second, what is its power output in watts? In horsepower? e) How much gasoline is ured per second? Per hour? IDENTIFY and SET UP: The problem is about a heat engine, foe ‘whieh we are told the amount of work done pereyele (W = 20083) and the amount of heat taken in per eyele (Qy = 10,000 J). Hence sve use the fis orm of Eq (20.4) 10 Find the thermal efficiency. The first law of thermodynamics tllsus the amount of heat ejected por eyele, and the heat of combustion tells us how much gasoline must tbe bummed per eycle and hence per unit time. EXECUTE: a) From the fist expression in Eg. (20.4, the thermal Ww _ 20005 10,0003 = 020 208 ‘This is a fairy typical figure for ears and trucks if W includes only the work actually delivered to the wheels. b) From Eq. 20.2). W= Qu + Qoso Qc = W~ Oy = =8000 20003 ~ 10.0004 EMA PTER 20 | The Second Law of Thermodynamics docs work; Wis neptive when the stm compress Each Qs pase fitrpresents ett ening the sysem adi sogaive i epresens bet Lnving Ie ate ‘When you know iat a quantity is egive such as Qn the above discussion i sometimes helps to write I as = Ie. 2. Some probe deal wits power ther han ney qua ties Powers work er unit tie (P~ Wt). and eo teat camer Qieat ren xa tre per ei tine (H = Qft). In such problems it helps to ask, “What is W or Qin one seco roe hou) 3. Keeping steps (1) and in min, slve forthe te ‘arabe, EVALUATE your answer: Use the first law of thermodynamics to check your results, paying particular atention to algebraic sims ‘That i, 8000 J of heat leaves the engine during each ey ©) Let m be the mass of gasoline hurned during eaeh evel Oy = mt, Qn _ 10,0003 1. 50x 10's 10) The power P (rate of doing work) isthe work per eyele multi plied by the numberof eyeles per second: P= (2000 Leyele) (25 eyeless) = $0,000 W = 50 KW thn 5 uz 0.000 W) 5 Wy Then 0208 = 67 hp .©) The mass of gasoline burned per second isthe mass per eycle rultiplied by the number of cycles per second: (0.20 gleyete) (25 eyelevs) = 5.0 gis “The mass bumed per hours ; (3091) 3222 = magn = 19 EVALUATE: We can check our result in part (e) by conve more familiar quantity, dhe amount of fuel consumed per unit dis: tance:The density of gasoline is about 0.70 gem, so this is about 25.700 em. 25.7, 6 gallons of gasoline per hour. Ihe tuck Js traveling at $5 mia (88 kan), this represents fuel consumption Of 8.1 milevgallon (3.4 kav.) This is substantial sumption than a passenger ear bu fairly typical of203 | Intemnal-Combustion Engines How much heat is rejected into the environment during a 24-hour period by 1000 trucks like those in Example 20.1 driving around a city? 20.3 | Internal-Combustion Engines ‘The gasoline engine, used in automobiles and many other types of machinery, isa familiar example of a heat engine. Let’s look at its thermal efficiency. Figure 20.4 shows the operation of one type of gasoline engine. First a mixture of air and wsoline vapor flows into eylinder through an open intake valve while the piston imum of V (when the piston is all the way up) £0 4 maximum of r¥ (when it is all the way down). The {quantity ris called the compression ratio; for present-day automobile engines its value is typically 8 to 10. At the end of this intake stroke, the intake valve closes and the mixture is compressed, approximately adiabatically, to volume V during the compression stroke. The mixture is then ignited by the spark plug, and the heated gas expands, approximately adiabatically, back to volume rV, pushing on the piston and doing work: this is the power stroke. Finally, the exhaust valve ‘opens, and the combustion products are pushed out (during the exaust stroke), leaving the cylinder ready For the next intake stroke descends, inereasing the volume of the cylinder from a mi The Otto Cycle Figure 20.5 is a p¥-diagram for an idealized model of the thermody} processes in a gasoline engine, This model is called the Otto eyele. At point « the szasoline air mixture has entered the cylinder. The mixture is compressed adiabat- ically to point b and is then ignited. Hi (Qj is added to the system by the burn- Ike Exhaust Tomake Exhaust vie wae vale wave pen closed lowed open Piston Connecting >\rod > Crankshaft (a) take stroke (>) Compression stroke (c) Ignition (@) Powerstroke (@) Exhaust stroke 20.4 Cycle of four-stroke intemal-combustion engine (a) Intake stroke: Piston moves ‘down, causing a partial vacuum in eylinde: gasoline mixture lows through open intake valve into cylinder (b) Compression stroke: Intake valve closes, and mixture is ‘compressed as piston moves up. (e)Jeniion: Spark plug ignites misture. () Power sirnke: Ho buried mistute pushes piston dover, doing work, (e) Evhaus stoke: Exhaust ‘valve opens and piston moves up, pushing burned mixture out of eylinder. realy for next intake stroke, and he cycle eepeats 759760 9+: Compression aks lahat compression) se: nite ue heating at constant solume) ood Power ake {adisbai expansion) 1 Reject he 1 erie {soning constant volume} 2055 The p¥-diagram forthe Otto cycle ‘an idealized model ofthe thermodynamic Processes in a gasoline engine cure 20 1 The Second Law of Thermodynamics ing gasoline along line be, and the power stroke is the adiabatic expansion td. ‘The gas is cooled to the temperature of the outside air along line da: during this process, heat |Qr| is rejected. In practice tis gas eaves the engine as exhaust and does not enter the engine again. But since an equivalent amount of gasoline and air enters, we may consider the process to be cyclic We can calculate the efficiency ofthis idealized cycle. Processes be and da are constant-volume, so the heats Q, and Qe are elated simply to the temperatures: Qu Oc ‘The thermal efficiency is given by Eq, (20.4), Inserting the above expressions and cancelling out the common factor Cy, we find Qu+ Qc _h=T+T~ Qu T.-T, 205) To simplify this further, we use the temperature-volume relation for adiabatic processes for an ideal gas, Eq (19.22). For the two adiabatic processes ab and ed, TUV) = VT! and Tey! = Tv! We divide each of these equations by the common factor V7~" and substitute dhe resulting expressions For 7 and 7, back into Eq, (20.5). The result is Wee h (—Ber'—1) : Te Ty -T)e Dividing out the common factor (Ty ~ T,). we get (thermal efficiency in Oo cycle) (20.6) ‘The thermal efficiency given by Eq. (20.6) is always less than unity, even for this idealized model, With r= 8 and y = 1.4 (the value for ai) the theoretical efficiency is e = 0.56, or 56%. The efficiency can be increased by increasing r However, this also inereases the temperature atthe enel of the adiabatic compres- sion of the air-fuel mixture. I the temperature is too high, the mixture explodes spontaneously during compression instead of burning evenly after the spark plug ignites it. This is called pre-ignition or detonation; iteauses a knocking sound and can damage the engine. The octane rating of a gasoline is a measure of its ant knock qualities. The maximum practical compression ratio for high-octane, or mremium,” gasoline is about 10, Higher ratios ean be used with more exotic fuels The Owo eycle, which we have just described, is a highly idealized model. It assumes that the mixture behaves as an ideal gas; it neglects friction, turbulence, Joys of heat to eylinder walls, and many other effects that combine t reduce the cfliciency of a real engine, Another source of inefficiency is incomplete combus- tion, A mixture of gasoline vapor with just enough air for complete combustion of the hydrocarbons to H,O and CO; does not ignite readily. Reliable ignition requires a mixture that is wsoline. The resulting incomplete combus- I to CO and unburned hydrocarbons in the exhaust, The heat obtained from the gasoline is then Tess than the total heat of combustion; the difference is204 | Refrigerator. wasted, and the exhaust products contribute to wsoline engines are typically around 35% pollution, Etficiencies of real The Diesel Cycle ‘The Diesel engine is similar in operation to the gasoline engine. The most impor tant difference is that there is no fuel in the cylinder atthe beginning of the com- pression stroke. A litte before the beginning of the power stroke, the injectors Start to inject fuel directly into the cylinder, just fast enough to keep the pressure approximately constant during the first part of the power stroke. Because of the high temperature developed during the adiabatic compression, the fuel ignites spontaneously as iti injected: no spark plugs are needed. The idealized Diesel eyele is shown in Fig. 20.6. Starting at point a, air is com- pressed adiahatically 10 point b, heated at constant pressure to point ¢, expanded dighatically to point d, and cooled at constant volume to point a. Because the is no fuel in the cylinder during most of the compression stroke, pre-ignition can not occur, and the compression ratio r can be much higher than for a gasoline cengine. This improves eliciency and ensures reliable ignition when the fuel is injected (because of the high temperature reached during the adiabatic compres: sion). Values of rof 15 to 20 are typical: with these values and y = 1.4, the theo- retical efficiency of the idealized Diesel cycle is about 0,65 to 0.70. As with the (Otto cycle, the efficiency of any actual engine is substantially less than this. While Diesel engines are very efficient, they must be built to much tighter tolerances than gasoline engines and the fuel-injection system requires careful maintenance. Use Fig. 20.5, nor Eq. (20.6), to explain why the efficiency of an Otto-cyele ‘engine increases with inereasing compression ratio. 20.4 | Refrigerators We can think ofa refrigerator asa heat engine operating in reverse. Aheat engine takes heat from hot place and givesoff heat t ce. Arefrigeratordoes the “opposite: it takes heat fromacold place (the inside ofthe refigerator) nd givesitoff toawarmerplace usually theairintheroom where the reftigeratorislocated). Aheat ‘engine has a net ouput of mechanical work: he reffigerator requires net input of rmechanieal work, Using the sign conventions from Section 20.2, foraretrigerator Qc ispositive but both Wand Q, arenegatives hence |W] = —Wand |Qy| = ~Qy. ‘A flow diagram fora refrigerator is shown in Fig. 20.7. From the first law for acyclic process, Qu + Qe—W=0 oF —Qy= Qe~ W ‘or, because both Qy and W are negative [Onl = Qe + |W (20. ‘Thus, as the diagram shows, the heat |Qy leaving the working substance and given tothe hot reservoir is always greater than the heat Qc taken from the cold reservoir, Note that the absolute-value relation lQul = I@cl + [WI 208) is valid for both heat engines and refrigerators 761 4+: Compression sre {adibatic compression) osc lgite as (being constant pres) ‘cod: Powerstroke {adiabatic expansion) 20.6 The pViagram forthe idealized Diesel eye Refrigerator 20.7 Schematic energy-Hlow diagram of &762 Inside of CHAPTER 20 1 The Second Lave of Thermodynamics From an economic point of view, the best refrigeration cycle is one that removes the greatest amount of heat |Q¢| from the inside of the refrigerator for the least expenditure of mechanical work, |1W|. The elevant ratio is therefore |Qe|/|W| the larger this ratio, the better the refrigerator. We eall this ratio the coefficient of performance, denoted by K. From Eg, 20.8), |W) ~ |Qy] ~ [Qel.s0 Ie! I@cl - ee (209 wi [aul = [ee (coefficient of performance of a refrigerator) units; Kis then a As always, we measure Qy, Qc. and W all in the same ener “dimensionless number The principles ofthe common refrigeration eycle are shown schematically in Fig. 20,8a, The fluid “circuit” contains a refrigerant fluid (the working substance). ‘The left sie ofthe circuit ineluding the cooling coils inside the refrigerator) is at Jow temperature and low pressure; the right side including the condenser coils ‘outside the refrigerator) sat high temperature and high pressure, Ordinarily, both sides contain liquid und vapor in phase equilibrium. ‘The compressor takes in fluid, compresses it adiabatically, and delivers it to the condenser coil at high pressure. The fluid temperature is then higher than that of the air surrounding the condenser so the refrigerant gives off heat [Qy| and par- tially condenses to liquid. The fluid then expands adiabatically imo the evapors- tor ata rate controlled by the expansion valve. As the fluid expands, it cools considerably. enough thatthe uid in the evaporator coil is colder than its sur roundings. Itabsorbs heat |Qc| from its suerounl w them and partially ‘vaporizing The fluid then enters the compressor to begin another eyele. The com- pressor, usually driven by an electric motor (Fig, 20.8b). requires energy input and does work [1] on the work substance during each cycle = Condoms expan aa owt t at a= Condenser or > Compressor : Compecssor o 20.8 (a) Principle ofthe mechanical re arranged in practical relrigerator igration cycle. (b) How the key elements are204 | Refrigeratons vot ais onside Composer Aw air conditioner operates on exactly the same principle. In tis ease the refrigerator box becomes a room oF an entire building. The evaporator coils are inside, the condenser is outside, and fans circulate ar through these (Fig, 20.9). In large installations the condenser coils are often cooled by water. For air condi- tioners the quantities of greatest practial importance are the rate of heat remov% (the heat current # from the region being cooled) and the power input P = Wit to the compressor. IF heat |Qc| is removed in time t, then H = |Qe|/t. Then we ean express the coefficient of performance as [Gel _ me _ jw) BrP i ‘Typical room air conditioners have heat removal rates H of $000 to 10,000 Buu, ‘or about 1500-3000 W, and require electric power input of about 600 to 1200 W. ‘Typical coefficients of performance are about 2.5; the actual values depend on the inside and outside temperatures. Unfortunately, K is often expressed commercially in mixed unis, with in Btu per hour and P in watts. In these units, H/P is called the energy efficiency rating (EER); the units, customarily omitted, are Btwh)/W, Because 1 W = 3.413 Buw/h, the EER is numerically 3.413 times as large asthe dimensionless K. Room a eon- ditioners typically have an BER of 7 to 10. ‘A variation on this theme is the heat pump, used to heat buildings by cooling the outside air. Ik functions like a refrigerator turned inside out. The evaporator coils are outside, where they take heat from cold air, and the condenser coils are inside, where they give off heat to the warmer air. With proper design, the heat |:Qy| delivered tothe inside per eyele can be considerably greater than the work |) required to get it there Work is always needed to transfer heat from a colder to a hotter body. Heat flows spontaneously from hotter to colder, and to reverse this flow requires the :uddition of work from the outside. Experience shows that itis impossible to make «refrigerator that transports heat from a colder body to a hotter body without the _uddition of work. If no work were needed, the coefficient of performance would be infinite. We call such a deviee a workless refrigerator: itis a mythical beast, like the unicorn and the free lunch 20.9 Anair conditioner works onthe Same principle 763764 CHAPTER 20-1 The Second Law of Thermodynamics ‘Why can’t you cool your house by leaving the refrigerator door open? 20.5 | The Second Law of Thermodynamics Experimental evidence suggests strongly that itis impossible to build a heat {engine that converts heat completely to work, that is, an engine with 100% ther- ‘mal efficiency. This impossibility is the basis of one statement of the second law of thermodynamies, as follows: I is impossible for any system to undergo a process in which it absorbs heat from a reservoir ata single temperature and converts the heat com= pletely into mechanical work, with the system ending in the same stat which it began, We will call this the “engine” statement of the second law. (It is also known t0 physicists as the Kelvin-Planck statement of this n,) The basis of the second law of thermodynamics lies in the difference between the nature of internal energy and that of macroscopic mechanical energy. In & ‘moving body the molecules have random motion, but superimposed on this is a coordinated motion of every molecule in the direction of the body's velocity. The kinetic energy associated with this coordinated macroscopic motion is what we call the kinetic energy of the moving body. The kinetic and potential energies associated with the random motion constitute the internal energy. ‘When a body sliding om a surface comes to rest as a result of friction, the o nized motion of the body is converted to random motion of molecules inthe body and in the surface. Since we cannot control the motions of individual molecule: \We cannot convert this random motion completely back to organized motion, We ccan convert part of it, and this is what a heat engine does, Ifthe second law were nor true, we could power an automobile or run a power plant by cooling the surrounding air, Neither of these impossibilities violates the first law of thermodynamics, The second law, therefore, is not a deduetion from the first but stands by itself as a separate law of nature, The frst law denies the Possibility of creating or destroying energy: the second law limits the availability of energy and the ways in which it can be used and converted. ur analysis of refrigerators in Section 20.4 forms the basis for an alternative statement of the second law of thermodynamics. Heat flows spontaneously from hotter to colder bodies, never the reverse, A refrigerator does take heat from a ccolder toa hotter body. but its operation requires an input of mechanical e work. Generalizing this observation, we state: I is impossible for any process to have as its sole result the transfer of heat from a cooler to a hotter body. We'll call this the “refrigerator” statement of the second law (itis also known as the Clausius statement). Kemiay not seem to be very closely relate to th statement. In fact, though, the two statements are completely equivalent. For example, if we could build a workless refrigerator, violating the second or “ref erator” statement of the second lay, we could use it in conjunction with a heat engine, pumping the heat rejected by the engine back to the hot reservoir to be205 | The Second Law of Thermodynamics An \ suet > Sad o0v-fficen engine Workies refrigerator 05 b, so = 2V, = 2(8.31 «10 m') = 166 x 10m Also during the isothermal expansion a —» b p,V, = PxVin 80 py = 24 = 5.00 10° Pa ¥, For the adiabatic expansion bc, Ty! = TeV", 80 vent = asso fteh Using the idea-gas equation again for pointe, we find ‘nk (0.200 mol) (8.314 Hmal-K) (300K) vo 306% 10%m 0837 % 10° Pa a 19.8 10 ‘nko _ (0200 mol) (8.314 HmolK (300 K) my, 298 x 104m 67 10° Pa b) For the isothermal expansion « —* b, AU. (= Qu). we use Eq, (20,10): 0. To find Was Ms Woy = Qu = mikTy In" = (0.200 mol) (8.314 S/mol-K) (500 K)(In 2) = 5763 For the adiabatic expansion b > c, Que = 0. From the first law of thermodynamics, MU, = Qh, ~ Why ~ Wis hence the work Wi. «done by the gas inthis process equals the negative of the eange in internal energy of the gas, From Eq, (19.13) we have AU = nCyAT, where AT = Te ~ Ty (final temperature minus ink temperature). Using Cy = 20.8 Nimol-K for an ideal diatomic gas, we Find Wy = MU, = ney Te nC(Ty ~ Te) = (0.200 mol) (20.8 Jimol-K) (500 K ~ 300K) = $32 For the isothermal compression ¢ >, AU = 0; Eq, (20.11) gives My Way = Oc = nk In 298 x 10° (020 (8814 HK) (30%) 59.6 10m Mod For the aiahatic compression da, Qu, =O. and Way MU g, = —nCy(Ty ~ Te) = mey(To ~ = (0.200 mol) (20.8 1mol-K) (300 K ~ 500 K) = 801 ‘We cam tabulate the results as follows: PROCESS 2 w au ab 5763 5761 o be ° ax 9324 cod -46I 361 ° aoa 0 832) sx oral 2303 2301 0770 .€) From the table, Qy, = 576 J and the total work is 230 J. Thus w _ 2305 © Oy” sngg 7 OO = Ore We ean compare this w Tua Te Th the result from Bg, (20.14): = 0.0 = 40% CHAPTER 20 1 The Second Law of Thermodynamics expect in a complete eye, the net heat input is used to do work with zero net change in the internal energy othe system. Note also thatthe quantities of work in the two adiabatic processes arene lives of each other. Can you show from the analysis leading to Fg. (20.13) that this must alivays be the ease in a Carnot eycle? Note thatthe efficiency in this example is greater than that ‘oblained in Example 20.2. That's because the rato ofthe low temperatures is higher, (300 K)/(300 K) as compared to 500 KVSSO K EVALUATE: In the table of \ e cycle @ Jn part (b), note that for W and AU = 0. These results ate just what we woul The Carnot Refrigerator Because each step in the Carnot cycle is reversible, the entire cycle may be reversed, converting the engine into refrigerator, The coefficient of performance Of the Camot refrigerator is obtained by combining the general definition of K, Bq, (20.9), with Eg. 20.13) for the Carnot cycle, We first rewrite Eq, (20.9) as _ Ich lOcl/Oul l@ul ~ [Gel 1 = J@cl/lQul ‘Then we substitute Eg. (20.13), |Qcl/|Qu] = ToT. into this expression. The result is Kem (20.15) (coefficient of performance of a Carnot refrigerator) Whenthe temperature difference Ty ~ Teissmall, Kismuch kargerthan unity: inthis ‘case alot of heat can be pumped” fromthe lowerto the higher temperature with only. alittle expenditure of work. But the greater the temperature difference, the smaller the value of K and the more work is required to transfer a given quamtity of heat, the eyele described in Example 20.3 is run backward as erator, what i it eoeticient of performance? Mot 203 lol || Because the eyele is a Carnot cycle, we may also use Ei, (20.15) 130 IDENTIFY and SET UP: Equation (20.9) gives the coefficient of 0K performance of any refrigerator in terms ofthe heat extracted from sok 3K the cold reservoir per eyele and the work that must be done per eek, » EVALUATE: For a Carnot cycle, ¢ and K depend only on the tem- panes s shown by Es. 20-14) and (2015) and we don't aed to calculate Q and W. For eycles containing ireversible processes and more EXECUTE: In Example 20.3 we found that in one eyele the Carnot engine rejects heat Qe = ~6 J 10 the cold reservoir and does work W = 2304. Hence, when run in reverse asa refrigerator, the system extracts heat Qc = +346 J from the cold reservoir while requiring a Work input of W = ~230 J. From Eq, 20.9), however, these two equations are not valid, culations are necessary.206 | The Carnot Cycle ™ Fear oer \ Sopertticen 100%ticien| the highest possible efice efficient” engine (more eff The Carnot Cycle and the Second Law Carmot engine) combined with Carnot We can prove that no engine cam be more efficient than a Carnot engine oper- into work wih no nt het rans ote ating between the same two temperatures. The key tothe proofs the above ok exer, This woud oa he observation that since each step in the Carnot cycle is reversible, s08 bow of Geemedyenaie. maybe reversed. Run backward, the engine becomes a refrigerstor, Suppose we have an engine that is more ecient than a Cammot engine Fig, 20.12) Let the Carnot engine, run backward as a refrigerator by negative work —|W|, take in heat Qc from the cold reservoir and expel heat |Qy| to the hot reservoir. Th Supereticint engine expels eat [Qc|, bu to doth takes in gr. thet Qy + 3. lis work output i then W + 3, and the net effect ofthe two Machines together ist fake quamity of heat A and convert t completely into wtork. This violates the engine statement of the second law. We could construct a Similarargoment hit a superefficent engine could he used to volt the religer ator statement of the secon Ta. Note that we dont have to assume that the sper a similar way we can show that no refrigerator ator operat he entire eyele efficient engine is reversible. I can have a greater coefficient of performance shan a Carnot refri “Tus the statement that no engine canbe more eicen than Carnot engine is yet another equivalent statement ofthe second Law of thermodynamics also fol. | ————___ lows directly tha all Carnot engines operating between the same two temper- s atures have the same efficiency, irrespective of the nature of the working substance. Although we derived Eq, 20.14) fora Carnot engine using an ideal gas as is working substance isin fat valid for any Carnot engine, no mater what its working substance Equation (20.14), the expression forthe efficiency of a Camot engine, set an upper mitt the ffcioney af areal engine such asa steam turbine. To maximize this upper fii andthe actual eficiney of the eal engin maketh intake temperature Ty a high as posible an the exhaust temperature T a low as possible (Fig. 20.13) The exhaust terperature cant be Tower than the lowest temperature aval able for cooling the exhaust. For steam turbine at an eletric power plat, Te may te the temperature of river o lake water, then we want the boiler temperature Ty 20.Y8 To mainize liieney the tm to be as hh as posible. The vapor presses of al guid increase raply with Peres ade ce te a are limited by the mechanical strength of the boiler At 300°C N#h.ss posible, Psat ceramie materi the vapor pressure of water is about 240 X 10° Pa (235 atm); this is about the in excess of LUC without meling or maxiniin practical pressure in large present-day steam boilers. becoming wl the designer must temperature, $0.™m CHAPTER 20 1 The Second Law of Thermodynamics “The Kelvin Temperature Scale In Chapter 17 we expressed the need for a temperature seale that doesn’t depend fon the properties of any particular material, We can now use the Carnot eyele to define such a scale. The thermal efficiency of a Carnot engine operating between two heat reservoirs al temperatures 7;, and 7. is independent of the nature of the working substance and depends only on the temperatures, From Eq. (20.4), this thermal efficiency is frien ae ou x ‘Therefore the ratio Qc/Qy is the same for all Carnot engines operating between two given temperatures Ty and Te Kelvin proposed that we define the ratio of the temperatures, Te/Ty, to be equal to the magnitude of the ratio Qc/Qy of the quantities of heat absorbed and rejected: Te _1@cl _ _@e z = ~S© (definition of Kelvin temperature) (20.16) Tu Toul On a Equation (20.16) looks identical to Eg. (20.13), but there is a subtle and crucial difference, The temperatures in Eq, (20.13) are based on an ideal-gas thermome- ter, as defined in Section 173, while Eg, (20,16) defines a temperature scale based “onthe Camot eyele and the second law of thermodynamics and is independent of the behavior of any panicular substance. Thus the Kelvin temperature scale is truly absolute, To complete the definition of the Kelvin scale, we assign, as in Section 17.3, the arbitrary value of 273.16 K to te temperature ofthe triple point ‘of water. When a substanee is taken around a Camoteyele, the rato ofthe heats absorbed and rejected, |Qy//|Qc| is equal to the ratio of the temperatures ofthe reservoirs as expressed on the gas-thermometer scale defined in Section 17.3. Since the tiple point of water is chosen to be 273.16 K in bath scales, it follows that she Kelvin and ideal-gas scales are identical ‘The zero point on the Kelvin scale is called absolute zero, Absolute zero can be interpreted on a molecular level; at absolute zero the system has its minimum possible total internal energy (kinetic plus potential). Because of quantum effects, however itis or true that at = 0, ll molecular motion ceases, There are theo- retical reasons for believing that absolute zero cannot be attained experimentally, although temperatures below 10~" K have been achieved. The more closely we Approach absolute zero, the more dificult itis to get closer. One statement of the third law of thermodynamics is that it is impossible to reach absolute zero in a finite number of thermodynam An inventor looking for financial support comes to you with an idea for a gaso- line engine that runs on a novel type of thermodynamic cycle. His design is made entirely of copper and is air-cooled. He claims that the engine will be 85% efficient, Should you invest in this marvelous new engine? (Hint; See Table 74) steps.207 | Entropy 20.7 | Entropy ‘The second law of thermodynamics, as we have stated it, is rather different in form from many familiar physical laws. [tis not an equation or a quantitative relationship but rather a statement of impossibiliry: However, the second law ean be st jon with the concept of entropy, the subject of this section, We have talked about several processes that proceed naturally in the direction ‘of inereasing disorder. Irreversible heat flow increases disorder because the mol «cules are intially sorted into hotter and cooler regions; this sorting is lost when the system comes to thermal equilibrium. Adding heat to a body increases its dis order because it increases average molecular speeds and therefore the random: ness of molecular motion. Free expansion of a gas increases its disorder because the molecules have greater randomness of position after the expansion than betire Entropy and Disorder Entropy provides a quantitative measure of disorder, To introduce this concept, Je’s consider an infinitesimal isothermal expansion of an ideal gas. We add heat ‘dQ and let the gas expand just enough to keep the temperature constant, Because the internal energy of an ideal gas depends only on its temperature, the internal energy is also constant; thus from the first law, the work dW done by the gas is ‘equal to the heat dQ added, That is 1o- av pav-May oo iQ Vo onkT ‘The gas is in-a more disordered state after the expansion than before because the molecules are moving ina larger volume and have more randomness of position. ‘Thus the fractional volume change dV/V is a measure of the increase in disorder, tnd the above equation shows that it is proportional to the quantity dQ/T. We introduce the symbol S for the entropy of the system, and we define the infinites- imal entropy change dS during an infinitesimal reversible process at absolute tem- perature Tas dQ a= (in initesimal reversible process) (20.17) Ifa total amount of heat Q is added during a reversible isothermal process at absolute temperature 7, the total entropy change AS = S, ~ S, is given by = 2 (mroneietiems pres] 0: As = 5; — Entropy has units of energy divided by temperature; the SL unit of entropy is | WK. We can see how the quotient Q/T is related to the increase in disorder. Higher temperature means greater randomness of motion, If the substance is initially cold, with Title molecular motion, adding heat Q causes a substantial fractional increase in molecular motion and randomness. But ifthe substance is already hot, the same quantity of heat adds relatively litle to the greater molecular motion already present, So the quotient Q/T is an appropriate characterization of the increase in randomness or disorder when heat flows into a system. 72374 CHAPTER 20 | The Second Law of Thermodynamics ple Entropy change in melting One kilogram of ice at °C is melted and converted to water at °C. Compute its change in entopy. assuming thatthe melting is done reversibly, The heat of fusion of water is Ly = 3.34 % 10° tg, IDENTIFY: The melting occurs ata constant vemperature of O°C, 0 this is a reversible isothermal process. SET UP: We are given the amount of heat added (in tems of the hneat of fusion) and the temperature 7 = 273 K. (Note that in entropy calculations we must alvays use absolute, or Kelvin, te peratures.) We can then calculate the entropy change using Fg. 20.18), EXECUTE: The heat needed to melt the ice is Q = ml, = 3.34 X10" J, From Eg. (20.18) the increase in entropy of the ss 334 x 10° = OY 122 x 108K As=s, 23K EVALUATE: This increase corresponds to the increase in disorder od state of @ \whea the water molecules go from the highly or cxystalline solid to the much mote disordered state of a liguid (Fig. 20.14 Inany isothermal reversible proces, the entropy change equal the heat transferred divided by the absolute temperature. When we refroeze the water, @ has the opposite sign, and the entropy change of the water is AS = —1.22 10° 1K. The water molecules rearrange themselves into a erystal to form ice, so disorder and ‘entropy bath decrease, “ore Ty , RA Cayman be 50. ic 4 7 HW Je 20.14 Water molecules are arranged ina regular ordered way in fan ie crystal. When the ice melts, the hydrogen honds between ‘molecules are broken, increasing the waters disorder and is enteopy We can generalize the definition of entropy change to include any reversible process leading from one state to another, whether resent the process as a series of infinitesimal reversible steps. During ty 1 quantity of heat dQ is added to the system at absolute temperature an infinitesin is isothermal or not, We rep- cal step, T. Then we sum (integrate) the quotients 4Q/T for the entire process: that is, as= f 2 (Crecrrttaneata Waceaine pee) 0,10) ‘The limits 1 and 2 refer tothe initial and final states. Because entropy is a measure of the disorder of a system in any speci it must depend only on the current state of the syster ‘will show later that this is indeed the case. When a system proceeds from WoL on its past history, We tial state with entropy S; to a final state with entropy ,, the change in entropy AS = $; ~ S; defined by Eq. (20.19) does not depend on the path leading from the initial to the Final state but is the same for all possible processes leading from state I 10 state 2, Thus the entropy of a system must also have a definite value For given state of the system, We recall that insernal energy, introduced in Chap- ter 19, also has this property, although entropy and internal energy are very dif Ferent quantities. Since entropy is function only of the state of a system, we ean also compute entropy changes in irreversible (non-equilibrium) processes for which Egs. (20.17)207 1 Enopy 775 and (20.19) are not applicable. We simply invent a path connecting the given initial ‘and final states that does consist entirely of reversible, equilibrium processes and. ‘compute the total entropy change for that path, It is not the actual path, but the ‘entropy change must be the same as for the actual path, As with internal enengy, the above discussion does not tell us how to calculate centropy itself, but only the change in entropy in any given process. Just as with internal energy, we may arbitrarily assign a value to the entropy of a system in a specified reference state and then calculate the entropy of any other state with ref- ‘erence to this, 1 kilogram of water at °C is heated to 10°C. Compute its change in entropy. IDENTIFY and SET UP: In practice the process described would be ‘done iveversibly, perhaps by setting a pan of water on an electric range whose cooking surface is maintained at HK0°C. But the ‘entropy change of the water depends only on the initial and final slates of the system, ind is the same whether the process is fine thatthe temper ture ofthe wat wed reversibly in a series of infinitesimal ‘eps, in each of which the temperature is ised hy an infinitesimal amount d7. We thea use Ey, (20,19) wo integrate overall these steps sn calculate the entropy change forthe total process ence we ea in EXECUTE: From Eg. (17.14) the heat required to carry out each su infinitesimal step is dQ = med, Substituting this into Ey (20.19) and integrating, we find fe-f-2 yak’ 23k) (1.0044)(419018-K) 131 108K A reversible adiabatic process [A gas expands adiabatically and reversibly. What is its change in entropy? In an adiabatic process, no heat enters or leaves the system, Hence dQ = Oru here is no change in entropy inthis eversble process: Entropy change in a temperature change EVALUATE: The entropy change is positive, as it must be For a ‘process in which te system absorbs heat. Tn this ealelation we assumed that the specific heat c daesn't depend on temperature. That's pretty good approsimation, since ¢ for water inereases by only 1% between O°C and 100°C, | AUTIONE 19 soiving this problem you might be tempted 10 avoid doing an integral by using the simpler expression in Eg (20.18), 45 ~ OFF. This would be incorrect, however, because E9 (20.18) is applicable only to isothermal processes andthe initial and final temperatures in our example are not the same, I you persist and use AS = OIF anyway, you would know what to use for the ‘numerator [Q = (4190 0k3-K)(1.00k3)( 100K) = 4.19% 103] but you would be confused about whether to use T= 273 K (the initial temperature 7), 373 K (the final temperature T), or per haps 323 K (the average off) and Fit doesnt matter, because ane of these choices would give you the right answer (Try itand see) The only correct way to find the entropy change ina process ‘with different initial and final temperaturesis to use Eg. (20.19) AS = 0. Every reversible adiabatic process is 2 constant-entropy process. (For this reason, reversible adiabatic processes are also called isentropic processes.) The increase in disoeder resulting from the gas occupying a greater volume is exactly balanced by the decrease in disorder associated withthe lowered temperature and reduced molecular spe776 A thermally insulated box is divided by a panition into two com partments, each having volume V (Fig. 20.15), Initially, one com partment contains n moles ofan ideal gas at temperature T, and the bother compartment is evacuated, We then break the partition, and the gas expands o ill bh compartments. What is the entropy change in this free-expansion process? Eo IDENTIFY: For this process, O = 0, W = 0, AU = 0, and therefore ecause the systems an deal gas) AT’ = 0. We might think thatthe entropy change is zero because there is no heat exchange. But Eg. (20.19) can be used to calculate entropy changes for reversible ‘processes only this fee expansion is nor roversibl, and thet i a entropy change. The process is adiabatic because Q = 0, but its ‘ot isentropic because AS # 0. As we mentioned atthe heginning, of this seetion, entropy increases in a free expansion because the o » we 20.15 (a,b) Free expansion of an insulated ideal gas, (e) The Foe ‘expansion process doesn’t passthrough equilibrium states from a to. However, the entropy change S, — S, canbe calculated by using the isothermal path shown or any reversible path from a1 b For the Carnot engine in Example 0.2 (Section 20.6), find the total entropy change in the engine during one cycle, EO IDENTIFY and SET UP: The Carnot cycle is made up of four reversible steps. shown in Fig. 20.11. We find the entropy change CHAPTER 20 | The Second Law of Thermodynamics Entropy change in a free expansion positions ofthe molecules are more random than before the expan- SET UP: To calutate AS, we recall thatthe entropy change depends ‘only on the initial and final states. We ean devise a reversible process having the same endpoints, use Bq, (20.19) t caleulat its entropy change, and thus determine the entropy change in theo inal process. An appropriate reversible process in this ease isan isothermal expansion from V to 2V at temperature 7. The gas does ‘work W during this substitute expansion, oan equal amount of heat Q must be supplied to keep the internal energy constant. We find the entropy change for this reversible isothermal process using Eq, (20.18) the entropy change for the free expansion willbe the EXECUTE: We found in Example 19.1 (Section 19.2) thatthe work ‘done by moles of ideal yas in an isothermal expansion from V,t0 Vois W = mI (V./V,). Using V; = V and Vs = 2V, we have wv 5 = W= nT In = nk In ‘Thus the entropy change is 2 2 As = T= mein? ‘which is also the entropy change for the free expansion with the ‘se inital and final states. or one mole, AS = (1 mol) (8.314 J/mol-K) (In 2) = $.763K EVALUATE: The entropy change is positive, as we predicted. The factor (In 2) in our answer is a result of the volume having. ireased by a factor of 2. Can you show that if the volume had increase inthe free expansion from V0 xV, where x isan arbitrary ‘number, the entropy change would have been AS = nf In? AS for each step, then add the entropy changes to get the total AS for the cycle as a whole EXECUTE: There is no entropy’ change during the adiabatic expan- sion or adiabatic compression. During the isothermal expansion at Tj, = 500K the engine takes in 2000 J of heat, and its entropy change, from Eg. 2018).Ty During the isothermal compression at Te = 38 ‘off 1400 Jot heat, a its entropy change is Qe _ = 14005 B8e= T= Saye 7 TOK ‘The total entropy change in the engine during one eycle is, 404K + ( MS was = AS + AS. 401K) = 0. EVALUATE: The result AS, = cells us that when the Carnot engine completesacyele,ithasthe sameentropyasitdid athe beg ring of theeycle. We'llexplorethiresltinthe following subsection, ‘What isthe total entropy change of the engine's environment during this eycle? The hot (500 K) reservoir gives off 2008 of heat Entropy in Cyclic Processes 7 20.7 | Enteopy “ducing the reversible isothermal expansion, so its entropy change is (2000 5)/(500 K) = =4.0 J/K: the cold 350 K) reservoir albsorhs 1400 J of heat during the reversible isothermal compres sion, so its entropy change is (+1400 3)/(380 K) = +40 9K. ‘Ths each individual reservoir has an entropy hiange: however, the sum of these changes—th fatal entropy change ofthe s9s- These results apply tothe special ease of the Carnot eye, for which ail ofthe processes are reversible. In this ease we find that the toral entropy change of the system and the environment together is zero, We will se that i the eycle includes reversible processes (a isthe ease forthe Otto eycle or Diesel eyele of Sec- tion 20.3), the total entropy’ change of the system and the ment cannot be zero, but rather must be positive Example 20.9 showed that the total entropy change for a cycle of a particular Carnot engine, which uses an ideal gas as its working substance, is zero. This result follows directly from Eq. (20.13), which we can rewrite as Oy, Qe 0 Ty” Te (20.20) The quotient Qy/Ty equals AS, the entropy change of the engine that occurs at T= Ty. Likewise, Oc/T, equals ASe, the (negative) entropy chs T= Te. Hence Eq, (20.20) says that AS + AS is zero net entropy change in one eycle What about Carnot engines that use a different working sub to the second law, any Camot engine operating between giver and Te has the same eff expression fore with Eq. (20:4), € Ey, (20.20) is valid for any Carnot working betwer whether its working substance is an ideal gas or not. We cor ‘entropy change in one cycle of any Carnot engine is cero, “This result can be generalized to show thatthe total entropy’ reversible cyclic process is zero, A reversible cyclic process apy ‘gram as a closed path (Fig. 20.16a). We can approximate such as we Tike hy a sequence of isothermal and adiabatic processes that occurs a P—Reversible cyte process hy orem Kt || 1\ by aseris of 1 FefTy (Eg, 20.14). Combining this + Qci just reproduces Eg, (20.20) 80 these temperatures, Approximating the path yan he cyl process Camor cycles ange of the engine = 0: that is, there stance? According n temperatures Ty 20.16 (a) A reversible eyelic process for tn ideal gas is shown as a red closed path toma pV-diggram. Several ideal- isotherms ate shown in blue, (b) We can approximate the pat in (a) by a series of Tome. thin Carmot cycles: one cyele is high lighted in gold, The total entropy change is zero for eact Carnot eycle and forthe elie process. (¢) The entropy hetween points « and bis indepen: dent ofthe path ude that the toral change during any spears on a pV-dia- ‘a path as closely forming parts of “wo paths (1 and 2) frm point to point enteopy ange s same fo either ath778 Harter 20 1 The Second Law of Thermodynamics ‘many long, thin Carnot cycles (Fig. 20,16), The tot cyele isthe sum ofthe entropy changes for is ero. So the total entropy change during any reversible cyele is zero: centropy change for the full (reversible eyelie provess) 021) [It follows that when a system undergoes a reversible process leading from any sate « to any other state b, she entropy change of the system is independent of the path (Fig, 20.16e). I the entropy change for path 1 were different from the change for path 2, the system could be taken along path 1 and then backward 2 to the starting point, with a nonzero net change in entropy. This would violate the conclusion that the total entropy change in such a cyelie process must be zero. Because the entropy change in such processes is independent of path, we con: clude that in any given state, the system has a definite value of entropy that depends only on the state, not on the processes that led to that state aici path Entropy in Irreversible Processes In an idealized, reversible process involving only equilibrium states, the total entropy change of the system and its surroundings is zero. But all ireversible processes involve an increase in entropy. Unlike energy. entropy is nota con served quantity: The entropy ofan isolated system cart change. but as we shall see it can never decrease. The free expansion of a gas, ‘an irreversible process in an isolated system in wh Suppore LOD kg of water at 100°C is placed in thermal contat with 1.00 ky of water at °C. What is the total change in entropy”? Assume thatthe specific heat of water is constant at 4190 Vkg=K IDENTIFY and SET UP: This process involves ireversible heat flow because ofthe temperature differences. We nonetheless calculate the entropy changes forthe (initially) hot water andl the (intially) cold water inthe same way asin Example 20.6 by assuring thatthe process occurs reversibly, We calewlate AS foreach substance using Fg (20.19) because the temperatures eange inthe process EXECUTE: The fin change ofthe hot water is perature is SOPC = 323 K, The entropy ar Rar Mx f oF = (1.002) (4190s of. £ ; ma) (219011) m 28) — as. The entropy chang ofthe cok water is 23K 88. © (21904) 557K) = $7051 described in Example 20.8, i$ Hp there isan entropy inerease Te 1o1al entropy change ofthe syst BS nat = AS + AS) = 603 YK) + 708 HK = +102 1K EVALUATE: An irreversible heat flow in an isolated system is secompanied by an inerease in entropy. We could have reached the same end state by simply mixing the two quantities of water. This, too, isan reversible process; because the entropy depends only on the state ofthe system, the total entropy change would be the same 102K. 1s worth noting that the entropy’of the system) increases con rinuonsly a the 40 quantities of water come to equilibrium. For ‘example, the first 4190 J of heat transfered cools the hot water to ‘99°C and warms the cold water to 1°C. The net change in entropy for this step is approximately 41903 | 41903 as = Se Oe aK ‘Can you show ina similar way thatthe net enteopy ch live for any one-degree temperature change leading tothe equilib rium conition?2008 | Microscopie Interpretation of Entropy Entropy and the Second Law The results of Example 20.10 about the flow of heat from a higher toa lower tem- perature, or the mixing of substances at different temperatures, are characteristic ‘of all natural (that is, irreversible) processes. When we include the entropy changes of all the systems taking part in the process, the inereases in entropy are always greater than the decreases. In the special case of a reversible process, the increases and decreases are equal. Hence we can state the general principle: When all systems taking part in a process are included, the entropy either remains constant or increases. In other words, no process is possible in whieh the total entropy decreases, when all systems taking part in the process are ineluded. This is un alternative statement of the second aw of thermodynamics in ‘entropy. Thus it is equivalent to the “engine” and “refrigerator” ‘cussed earlier. Figure 20.17 shows a specific example of this zen ‘The increase of entropy in every natural, irreversible process measures the increase of divorder or randommess in the universe associated with that process. Consider again the example of mixing hot and cold water (Example 20.10). We ‘might have used the hot and cold water as the high- and low-temperature reser voirs of a heat engine. While removing heat from the hot water and giving heat to the cold water, we could have obtained some mechanical work. But once the hot and cold water have been mixed and have come to a uniform temperature, this ‘opportunity to convert heat (9 mechanical work is lost irretrievably. The luke- warm water will never unis itself and separate into hotter and colder portions No deci been lost is not ener; but opportunity, the opportunity to convert part ofthe heat from the hot water into mechanical work, Hence when entropy increases, ener) becomes less available, and the universe becomes more andor of Suppose 2.00 kg of water at 50°C spontaneously changes temperature, so that half of the water cools to 0°C while the other half spontaneously warms «© 100°C, (All of the water remains liquid, so it doesn’t freeze of boil.) What would be the entropy change of the water? Is this process possible? Why or why not? ave in energy occurs when the hot and cold water are mixed, What has “20.8 | Microscopic Interpretation of Entropy We described in Section 19.4 how the internal energy ofa system could be ealeu- lated, at least in principle, by adding up all the kinetic energies of its constituent particles and all the po scopic calculation of the entropy S of a system, Unlike energy, however, entropy js not something that belongs to each individual particle or pair of particles in the system, Rather, entropy is a measure of the disorder of the system a a whole. To e how to calculate entropy microscopically, we first have to introduce the idea of macroscopic and microscopic states, ‘Suppose you toss WV identical coins on the floor, and half of them show heads and half show tails. This isa description of the large-seale or macroscopic state of the system of NV’ coins. A description of the mieroseopie state of the system includes information about each individual coin: Coin 1 was heads, coin 2 was tails, coin 3 was tails, and so on, There can be many microscopic states that cor- 779 20.17 The mixing of colored ink snd ater starts from a sate of relative omer low entropy) in which each Mud is sepa rate and stint from the ether. The final Sate after mixing is more disordered (has sreater entropy). Spontaneous wnmixing of the ink and water, a process in which there would be a net decrease in entropy, is never observed780 20.18 All possible microscopic states of four coins. There can be se ciate 20 1 The Second Law of Thermodynamics Macroscopic state CCorespnding microscopic states voir OOO @ OOo “Tse heal one tile cag eee coneeel eee Gece ooee One the tal = eee o¢e + 9999 9909 wr GOOQ respond to the same maeroscopic description. For instance, with N = 4 coins there are six possible states in whieh half are heads and half ae tails (Fig. 20.18) “The number of mieroscopie states grows rapidly with increas there are 2" = 1.27 X 10" microscopic states, of which 1.01 x 10” are half Fneads and hal tails “The least probable outcomes of the coin toss are the states that are either all heads or all tails. It is certainly possible that you could throw 1¢K) heals in a row, but don’t bet on it: the probability of doing this isonly Vin 1.27 % 10%. The most probable outcome of tossing N’ coins is that half are heads and half are tails. The reason is that this macroscopic state has the greatest number of corresponding _microscopic states, as shown in Fig, 20.18 ‘To make the connection to the concept of entropy. note that N coins that are all heads constitutes a completely ordered macroscopic state; the description “all heads” completely specifies the state of each one of the NV coins, The same is tue it the coins ae al tals. But the macroseopie description “half heads, half tails” by itself tells you very litte about the state (heads or tails) of each individual coin, We say thatthe system is disordered because we know so Title about its micro- scopic state. Compared to the state “all heads” or “all tails.” the state “half head, half tails” has a much greater number of possible microscopie states, much greater disorder, and hence much greater entropy (which isa quantitative measure of dis order) Now instead of N coins, considera mole of an ideal gas containing Avogadro's number of molecules. The macroscopic state of this gas is given by its pressure p, ‘Volume V, an! temperature 7; description of the microscopic state involves stat-208 | Microscopic nt ion of Entropy 781 ing the position and velocity for each molecule in the gas. At a given pressure, volume, and temperature, the gas may be in any one of an astronomically large rhumber of microscopic states, depending on the positions and velocities of its 6.02 x 10* molecules. Ifthe gas undergoes a free expansion into a greater vol ‘ume, the range of possible positions inereases, as does the number of possible The system becomes more disordered, and the entropy increases as calculated in Example 20.8 (Section 20.7). ‘We can draw the following general conclusion: For any system, the most prob table macroscopic state is the one with the greatest number of corresponding ‘microscopie states, which is also the macroscopic stare with the greatest disorder «and the greatest entropy Let w represent the number of possible microscopic states for a given macro- scopic state. (For the four coins shown in Fig. 20.18 the state of four heads has w = 1, the state of three heads and one tails has w = 4, and so on.) Then the entropy S of a macroscopic state can be shown to be given by microscopic state: S=kInw (microscopic expression for entropy) (20.22) where & = RIN, isthe Boltzmann constant (gas constant per molecule) into duced in Section 18.3. As Eg. 20.22) shows, inereasing the number of possible microscopic states increases the entopy S ‘What matters in athermodynamie progess is not the absolute enteopy S but the difference in entropy between the intial and final states. Henee an equally valid and useful definition would be § = &In w + C, where Cis a constant, since C cancels in any calculation of an entropy difference between two states, But it's convenient to set this constant equal 0 zero and use Eq. (20.22). With this choice, since the smallest possible value of w is unity the smallest possible value ofS for any system is & In 1 = 0. Entropy can never be negative. In practice. calculating w isa difficult task, so Eq, (20.22) s typically used only to calculate the absolute entropy Sof certain special systems, But we ean use this relation ocaleulate differences in entropy between one state and another. Consider system that undergoes a thermodynamic process that takes it fom macroscopic state I. for which there are w, possible microscopic states, to macroscopic state 2, with ws associated microscopic states. The change in entropy in this process is Kinws =n, =n“? 033) ‘The difference in entropy between the two macroscopic states depends on the ratio of the numbers of possible microscopie states. AS the following example shows, using Eq. (20,23) to calculate a change in entropy from one macroscopic state to another gives the same results as consider: ing a reversible process connecting those (v0 states and using Eq. (20.19). Use Eq. (20.28) 10 calculate the entopy change in the fre expan sion of n moles of gas at temperature F describe in Example 20.8 Nata a ole lle IDENTIFY and SET UP: The situation i shown in Figure 20.19. ‘When the pastition is broken, the velocities of the molecules are782 ‘CHAPTER 20 | The Second Law of Thermodynamics unaffected, sine no work is done, But each molecule now has tssige as much volume in which it ean move and hence has twice the number of possible positions. This is all we need to calculate the entropy change using Bq, (20.23). EXECUTE: Let w; he the number of microscopie sates of the sys temasa whole when the gas occupies volume V; (Fig. 20.193)-The umber of molecules is N = ng, and each molecule has twive as ‘many possible states after the partition is broken, Hence the nu 20.19 In. ber i of microscopic states when the volume occu (Fig. 20.196) is greater by a factor of 2": that i, ws = 2. The change in entropy in this process is as Since N= nNy and k = RUN this becomes: (a) Gas oeupies soume ¥: namber 6) Gas occupies AS = (ng) (RIN) In2 = nin ‘oa 22 mabe ot mrttes EVALUATE: We have found the same result a in Example 20.8, but ™ without any reference tothe thermodynamic path taken snsion of W molecules in which the volume ‘doubles the numberof possible microscopic sates inereases by “The relationship between entropy and the number of microscopic states gives us new insight into the entropy statement of the second law of thermodyn that the entropy of a closed system ean never decrease. From Eq, (20.22) this ‘means that a closed system can never spontaneously undergo a process that decreases the number of possible microscopic states ‘An example of such a forbidden process would be ill of the air in your room spontaneously moved to one half ofthe room. k tum in the other halt Such a “free compression” would be the reverse of the free expansion of Examples 20.8 and 20,11. This would decrease the numberof possible microscopic states by «factor of 2, Strictly speaking, this process is not impossible! The probability of finding a given molecule in one half of the room is 1/2. so the probability of ing all ofthe molecules in one half of the room at once is (1/2). (This is exactly the same as the probability of having a tossed coin come up hy row.) This probability isnot zero, But les you worry about suddenly finding your self gasping for breath inthe evacuated half of your room, consider that a typical room might hold 1000 moles of air, and soN = 1000N, = 6.02 x 10° mole cules. The probability of all the molecules being in the same half of the room is, therefore (1/2). Expressed as a decimal, this number has more than 10° zeros tothe right of the decimal point! Because the probability of such a “free compression” taking place is so van ishingly small, it has almost certainly never occurred anywhere in the universe since the beginning of time. We conclude that forall practical purposes the second Jaw of thermodynamics is never violated, ‘A quantity of N molecules of an ideal gas initially occupies volume V. By what factor does the number of microscopic states ofthe gas inerease ifthe gas expands 10 volume 2V in (a) a reversible isothermal process or (b) a reversible adiabatic process? Are the answers the same or different? Why? ving a va ads IV times in aSummary 783 SE A reversible process is one whose direction can be reversed by an infinitesimal change in the conslitons ofthe process, and in which the system is always in oF very close to thermal ‘equilibrium, Al other thermodynamic processes are ineversile takes heat Qy from a source, con- w o work W, and discards the “a remainder |Qc| ata lower temperature. The ther a mal efficiency eof heat engine measures how uch ofthe absorbed heat is converted 10 work (See Example 20.1) 4% oy A gasoline engine operating on the Oto eyele has theoretical maximum thermal efficiency e that ‘depends on the compression ratio rand the ratio ‘of heat capacities ofthe working substance. (206) A refrigerator tkes heat Qc from a colder place, IQe| 2 tas a work input WV) and discards eat |Qy| a ceil al ‘warmer place. The effectiveness of the refrigera- ' Gl tors given by its coeticient of performance K 209) “The second! lw of thermodynamics describes the directionality of natural thermodynamic processes, can be stated in several equivalent forms. The engine statement is that no yee process can con: ver heat completly into work. The refrigerator staement is that no eyelie process ean transfer eat from a colder place to a hotter place with no input of mechanical work, “The Carnot cycle operates bewween wo heat Te_Tu- reservoirs a temperatures Ty and Te and uses Seam = 15 ‘only reversible processes. Is thermal efficiency depends only on Ty and T,. An additional equiva lent statement of the second! law is that no engine ‘operating between the same two temperatures can ‘he more efficient than a Carnot engine. (See Examples 20.2 and 20.3) 20.14) 2784 HAPTER 20 I The Second Law of Thermodynamics A Carnot engine run backward is a Carnot relrigeratr. Ils coefficient of performance depends ee only on Ty and Te. Another Form of the second law states that no refrigerator operating between Ken i the same two temperatures can have a larger coefficient of performance than a Carnot refigers- (20.15) tor. (See Example 20.4) ‘The Kelvin temperature scale is based on the efficiency ofthe Carmot cycle and is independent of the properties of any specific material. The evo point on the Kelvin scale is called absolute zero, change in any reversible process depends on the amount of heat flow and the 20.19) absolute temperature 7; Entropy depends only onthe state ofthe system, and the change in entropy between given initial an inal states isthe same forall processes leading from one sate to the other. This fat can be used to find the entropy change in an ieversible process, (See Examples 20.5 through 20.10) Enropy i quantitative measure ofthe disorder of a ystem. The entropy a0 as- [2 aeversible process) {An important statement ofthe second law of thermodynamics is thatthe entropy ofan isolated \ system may increase hut can never decrease. When a system interacts with its surroundings, the ‘otal entropy change of system and surroundings ean never deerease. When the ryan nzelaey Pi Imre roe en ary bona nS > involves only reversible processes, the toll entropy is constant and AS ‘When a system isin a particular macroscopic S=kinw (20.22) 7 state, the particles that make up the system may Eeesht ‘he in any of w possible microscopic states, The eater the number whe greater the entropy: (See Example 20.11) Key Terms irreversible process, 755, ‘compression ratio, 759 second law of thermodynamics, 764 reversible process, 755 ‘Otto eyele, 759 Carnot eyele, 766 equilibrium process, 755 Diesel eycle, 761 Kelvin temperature scale, 772 heat engine, 756 refrigerator, 761 absolute zero, 772 working substance, 75 coefficient af performance, 762 ‘entropy, 773 eyelic process, 756 ‘energy efficiency rating, 763 macroscopic state, 779 thermal efficiency, 757 heat pump, 763, Your Notes