Check boundaries, units, conservation, entropy, and rates before accepting a mechanism or performance number.
By the end: combine macroscopic balances and microscopic assumptions without confusing model feasibility with measured performance.
Make the claim complete enough to test
Start with a system boundary and an interval. Is the device cyclic, still warming up, consuming a chemical fuel, discharging a battery, or losing mass? A warm store can produce work temporarily while its own state changes. That is different from an engine returning to its starting state after every cycle. Open systems carry energy with flowing matter as well as heat and work.
List each energy source and destination, using one set of units and an explicit work sign. Check whether reported energy is per cycle, per second, or over an interval. Record absolute reservoir temperatures and distinguish them from ambient readings or Celsius differences. An efficiency or COP must name both numerator and denominator. A number without this boundary information may be uninterpretable rather than immediately impossible.
Resolve an apparent energy surplus
Consider a fictional report covering a ten-second interval. It lists 1,000 J of heat entering a closed device, 700 J leaving as heat, and 500 J leaving as work. Before calling this impossible, write the signed account for the device alone. Net heat input is 300 J. Subtracting 500 J of work gives an internal-energy change of minus 200 J. Conservation permits these transfers if the device loses that much internal energy during the interval. The report has not yet established a cycle.
Now suppose the report also states that the device returns to exactly the same thermodynamic state. That additional statement requires zero internal-energy change, contradicting the calculated loss. One possible explanation is an omitted store, such as a battery included inside the device boundary. An omitted store does not rescue the complete-cycle claim for the combined device and store. Returning to the same temperature alone is insufficient, because composition or phase proportions might have changed while temperature stayed fixed.
One transfer, two boundary choices
Schematic arrows show energy directions, not scaled widths. The dashed boundary includes a store losing 200 J. For the device alone, the store supplies 200 J of work input, so net work output is 300 J.
Draw a larger boundary around both the device and a store that supplies 200 J. For the device alone, that supply must appear as an incoming transfer. For the combined system, it appears as decreasing stored energy rather than energy arriving from outside. Moving a boundary changes whether a contribution is internal or external. It does not create another source. Heat exchanged between blocks likewise cancels for their insulated pair.
Units can reveal a different apparent surplus. In the report, 500 J over ten seconds means an average work-output rate of 50 W. A separate statement of 500 W would mean 5,000 J over that interval if the rate remained constant. Compare energies over the same interval before balancing them. A peak rate cannot replace an integrated energy. Likewise, a per-cycle heat input cannot be compared directly with a per-second work output unless the number of cycles per second is also specified.
Temperature is another incomplete inventory. A hypothetical body with whole-sample capacity 200 J/K can lose 200 J by cooling just 1 K, under a constant-capacity, negligible-work approximation. A second body with capacity 20 J/K must cool 10 K to supply the same energy. A small temperature change therefore does not prove negligible accumulation. If capacity were instead given per kilogram, multiply by the sample mass first. If given per mole, multiply by amount in moles rather than silently treating it as a whole-device value.
Equal energy loss, different temperature changes
Hypothetical constant whole-sample capacities, negligible work. Both temperature scales run from 290 to 300 K. Each connecting line represents a 200 J loss, not a cooling trajectory or elapsed time.
For a paper exploration, replace the reported work by 400 J while retaining the original heat amounts and interval. Predict the missing accumulation before calculating it. Net heat remains 300 J, so the device must lose 100 J internally, corresponding to 0.5 K for the stipulated 200 J/K body. If the report insists on cyclic operation, identify the necessary correction instead. Either work must fall to 300 J, or another net input of 100 J must enter. These are alternative accounts, not identified mechanisms.
Apply conservation and direction as separate tests
A proposed engine absorbs 1,000 J at 600 K and rejects heat at 300 K. If it claims 700 J of work with no other changes, conservation requires 300 J rejection, but the resulting entropy decrease rules out the claim. If it instead claims 300 J work, both laws permit the account. Permission does not establish that a real device achieves it or that all losses and inputs were measured.
Suppose an accompanying refrigerator claims COP nine. Between 270 K and 300 K, nine is the reversible cooling bound; as a heating COP it would describe a different numerator. At other operating temperatures the same number can have a different feasibility status. A shared percent-looking display cannot substitute for definitions. Likewise, a negative standard reaction Gibbs value cannot establish a rate or a direction for an arbitrary composition.
Use entropy to distinguish balanced accounts
An energy-balanced account still leaves the direction of a cyclic process undecided. Consider 1,200 J taken from a 500 K reservoir and 600 J rejected to a 300 K reservoir, leaving 600 J of work. The working system returns to its initial entropy. The hot reservoir loses 1,200 divided by 500, or 2.4 J/K. The cold reservoir gains 600 divided by 300, or 2 J/K. An isolated total containing both reservoirs and the cyclic device would lose 0.4 J/K, so this account fails.
To repair that entropy deficit without adding another interaction, increase rejection to the cold reservoir. It needs at least 2.4 J/K, requiring 720 J at 300 K. Conservation then leaves at most 480 J of work. This restriction follows without specifying an internal mechanism. Rejection of 800 J would leave 400 J of work and generate about 0.267 J/K. That alternative passes these balances, although its mechanism, operating rate, and actual performance remain unspecified.
The entropy rectangles below explain why equal joules do not necessarily cancel. Each area represents energy multiplied by inverse absolute temperature. Removing energy from the hot side removes a smaller entropy per joule than adding it to the cold side supplies. Only part of the input energy must reach the cold reservoir to replace the hot reservoir's entropy loss in a reversible cycle. The remainder can leave as work, which does not carry entropy across the boundary in this ideal mechanical-work description.
The cold side must replace the entropy removed
Rectangle area encodes reservoir entropy magnitude. Width is energy at 0.3 SVG units/J; height is inverse temperature at 30,000 SVG units·K. Each area unit represents 1/9,000 J/K. The two cold rectangles are alternative accounts, not simultaneous transfers.
This reasoning requires reservoirs whose temperatures stay fixed over the transfers. Replace the hot reservoir with a cooling finite body, and dividing its whole energy loss by its initial temperature becomes wrong. For a constant-capacity body, calculate the endpoint entropy change with the logarithmic temperature ratio. Alternatively, divide successive small reversible heat increments by their changing temperatures. The reference finds a state difference, without asserting that actual exchange is reversible. The body's entropy loss alone does not violate the second law.
A refrigerator requires a related test with different signs. Suppose it extracts 800 J at 250 K and receives 100 J of work. Conservation requires 900 J rejection at 300 K. Cold-side entropy decreases by 3.2 J/K, while hot-side entropy increases by only 3 J/K. Reduce extraction to 500 J with the same work, and rejection becomes 600 J. Both entropy magnitudes are then 2 J/K. Cooling COP is five, while heating COP is six, because the latter counts the rejected heat.
Work through a third proposal on paper, extracting 400 J at 250 K for the same 100 J input and rejecting it at 300 K. Rejection is 500 J, cooling COP is four, and total reservoir entropy change is about positive 0.067 J/K. Explain why this is allowed even though it has lower COP than the reversible account. Positive entropy generation permits irreversible operation. No cycle duration is specified, so COP and heat per cycle cannot establish cooling power.
Work joins the extracted heat before rejection
The reversible limiting account extracts 500 J at 250 K and rejects 600 J at 300 K. Arrows are schematic. The bottom strip uses one SVG unit per joule; its two contributions add to the rejected heat.
Work through the mechanism
Conservation over a full cycle
Qh and Qc are positive heat magnitudes in J per cycle. The working substance returns to its initial U and S.
Nonnegative entropy generation
No other reservoirs, matter flows, or energy stores are included.
Carnot efficiency bound
η is a dimensionless fraction, not a power. Equality requires reversibility. Use kelvin.
Test a proposed cyclic energy account
Efficiency 30%; reversible bound 50%. Reservoir entropy change 0.67 J/K. Permitted by these balances; not verified device performance.
All proposals take 1,000 J from 600 K and reject heat at 300 K. Arrows show directions, not scaled flow widths; the bottom strip divides the same 1,000 J into work and rejected heat by length. Controls replace the proposed work immediately. An impossible proposal is labeled, not simulated as a functioning engine.
Claim
Necessary check
What it does not prove
Balanced energy flows
First law
Allowed entropy change
Nonnegative total entropy
Second law
Real device performance
Negative reaction ΔG
Current T, p, composition
Fast reaction
Temperature plateau
Pressure and phase state
No energy input
Apply the checks to the whole stated process, not to a favorable subset of its transfers.
Test what the displayed engine account actually establishes
Use the engine account above as a controlled comparison rather than a machine demonstration. Select the named 300 J work preset and predict the heat rejection before reading the result. With 1,000 J input, rejection must be 700 J. At the fixed 600 K and 300 K temperatures, the reservoirs gain a combined two-thirds of a joule per kelvin. Now select the 500 J preset. Rejection falls to 500 J and the entropy sum reaches zero. Only the second is a reversible limit.
Select the 700 J proposal next. Explain its failure using the reservoir entropy changes, not merely its percentage. Rejection is now 300 J, giving the cold reservoir 1 J/K while the hot reservoir loses five-thirds J/K. The deficit is two-thirds J/K. The display examines an inconsistent proposal, not a functioning engine. Finally choose zero work. All 1,000 J reaches the cold side, and the positive entropy sum describes direct heat transfer as an account with no useful work.
No route of the working substance is specified by those energy splits. To see the missing information geometrically, compare two hypothetical expansions of the earlier ideal-gas sample from 10 L to 20 L at unchanged equilibrium endpoint temperature. A reversible isothermal route performs about 693 J of work and receives the same heat. Insulated free expansion performs no work and receives no heat. Both leave internal energy unchanged. Actual work requires the route or external loading.
On a pressure-volume drawing, a quasistatic reversible gas curve can calculate boundary work because opposing external pressure approaches the equilibrium gas pressure. In free expansion, opposing pressure is zero. Drawing an isotherm through the same endpoints does not make its area the actual work. Slow piston motion also fails to guarantee reversibility if friction remains. A report might substitute an equilibrium curve for actual boundary forces and wrongly present its area as established work output.
The same endpoints do not supply the same work
Left: reversible isothermal gas pressure, p = 1,000/V with p in kPa and V in L. Shading is about 693 J of work. Right: actual external pressure in free expansion is zero, so its work area is zero. Right endpoint marks show gas pressures only; no equilibrium intermediate gas curve is asserted.
Microscopic counting has a parallel limitation. Four labeled particles independently occupying equal halves have sixteen equally weighted left/right assignments. Exactly two left has six assignments and probability 6/16. Eight particles have seventy exactly balanced assignments, yet their probability is 70/256, which is smaller. Compare a near-balance band instead. Allowing left fractions between one quarter and three quarters gives 14/16 for four particles and 238/256 for eight. Here the broader event becomes more probable. The event definition and normalization matter as much as the count.
Inspect the counting diagram and separate each assignment from the category containing it. A single balanced assignment is no more likely than a single all-left assignment under the stipulated weights. Nor are these left/right strings complete molecular states. Actual gas phase space includes detailed positions and momenta, which the coarse assignment omits. Consequently, an increase in the number printed beside a category cannot by itself establish a physical entropy change, a transition rate, or a measured probability. The constraints and probability weights must support that inference first.
Exact balance and a near-balance band are different events
Counts enumerate coarse independent equal-half assignments, not full gas phase space. Cells are categories, not individual assignments. Equal cell widths do not encode probability. The bracketed bands include left fractions from 1/4 through 3/4, including their endpoints.
Ask what observation would distinguish the alternatives
For a real thermal device, measure relevant heat flows or calibrated energy balances, work or electrical inputs, operating temperatures, and accumulation over the test interval. Include uncertainty and auxiliary equipment. Check whether the measurement interval actually returns the working system to the same state. For a material model, test whether heat capacity, phase, or interaction assumptions change in the operating range.
The laws constrain outcomes; they do not identify every microscopic mechanism or time scale. Counting configurations explains typical equilibrium behavior under specified constraints, while conductivity, transition barriers, and transport models describe how equilibrium is approached. An expert explanation should make clear which claim follows from conservation, which follows from the second law, which assumes a material model, and which needs empirical evidence.
Identify the missing evidence without changing the question
A temperature plateau can conceal a large change in a sample's state. Compare two stipulated one-kilogram water samples at the melting temperature and approximately one atmosphere. One is one-quarter liquid by mass; the other is three-quarters liquid. Their temperatures agree, but an additional half kilogram has melted in the second. Using the lesson's rounded fusion value, the enthalpy difference is 167 kJ. Matching temperatures alone misses this difference. A cycle must restore phase proportions too.
The equal-width segments in the phase-fraction diagram below represent equal masses, not equal occupied volumes. Ice and liquid have different densities, so a container drawn half full of each phase would not establish equal mass fractions. The same audit habit applies to gas models. An ideal-gas pressure-volume relation does not establish monatomic heat capacity. Rotational and vibrational energy storage can change capacities, and quantum energy spacings can suppress accessible modes. A material's energy inventory therefore needs an appropriate model, not temperature and a visually plausible drawing alone.
Restore the mass fractions to restore the state
Each segment represents 0.25 kg, not a volume. Both one-kilogram samples are at 0 °C near one atmosphere. Two more segments have melted in the second sample, increasing enthalpy by 0.5 kg × 334 kJ/kg = 167 kJ.
A chemical explanation needs a different missing variable. For the hypothetical reaction at 500 K, standard reaction Gibbs energy is minus 10 kJ/mol. At quotient one, the composition correction vanishes and forward conversion lowers Gibbs energy locally. At quotient 100, the correction is about positive 19.14 kJ/mol, so reverse conversion lowers it. Different compositions reverse the local direction despite identical standard parameters. Neither value is the total energy released by converting an entire changing mixture into product.
Read the Gibbs sketch as the changing slope of total mixture Gibbs energy, not as an activation-barrier diagram. At fixed temperature and pressure, equilibrium lies at a minimum subject to the allowed composition change. On the reactant-rich side, a small forward conversion lowers Gibbs energy. On the product-rich side, a small reverse conversion lowers it. A catalyst can change how quickly either side approaches that minimum without moving the equilibrium for the same states. A negative local slope therefore supplies no numerical time estimate and no guarantee of complete conversion.
Equilibrium energy populations require equal care about categories. With three excited states in the two-level model at 300 K, the excited level contains about 52.46% of the probability, but each excited state contains only about 17.49%. The single ground state has about 47.54%. Calling the excited state more probable confuses one state with the entire level. These equilibrium weights also say nothing about how often transitions occur. A population measurement and a transition-time measurement answer different questions.
Composition changes the direction toward equilibrium
Schematic total-mixture Gibbs energy versus forward conversion at fixed 500 K and pressure. Neither axis is numerically scaled. The marked q values locate sides of the minimum only; horizontal distances are not quotient differences. This is not an activation-barrier curve. Local reaction Gibbs energies use the hypothetical lesson parameters.
Finish with a written verdict for a fictional report claiming a constant-temperature device, a large configuration count, and negative standard reaction Gibbs energy as proof of sustained work. None supplies a complete cycle account. Request the phase and composition changes to resolve accumulation, a defined state space and weights to interpret the count, and the actual quotient to establish local chemical direction. Separate the energy, entropy, and rate questions. Missing quantities do not make every incomplete claim impossible.
Exercises
Separate feasibility from validation
A cyclic device takes 1,000 J at 600 K, rejects 700 J at 300 K, and delivers 300 J of work. Does passing the first and second laws verify its advertised real-world performance?
Hint
First calculate the balances, then distinguish consistency from a measurement.
Worked answer
Question
Result
Energy conserved?
Yes: 1,000 = 700 + 300
Entropy nondecreasing?
Yes: about +0.667 J/K
Advertised operation verified?
No; measurements and boundary checks remain
Consistency tests can reject some claims without proving the remaining ones.
The energy balance closes and total reservoir entropy rises by about 0.667 J/K. The proposal is thermodynamically permitted under the assumptions, but the arithmetic does not verify the device. You still need measurements, uncertainty, auxiliary inputs, reservoir temperatures, and evidence that the device returns to its initial state. A permitted account is not a demonstrated mechanism.
Find three different category mistakes
A report says: an insulated gas cannot gain energy; melting ice cannot absorb energy while its temperature stays fixed; and a reaction with negative standard ΔG must run rapidly to completion. Correct each statement.
Hint
Check work, latent heat, composition, and activation barriers separately.
Worked answer
Mistake
Missing distinction
Insulated means no energy transfer
Heat versus work
No temperature rise means no heat input
Sensible warming versus latent heat
Negative standard ΔG guarantees rapid completion
Standard versus actual driving force; thermodynamics versus kinetics
Identify the missing physical quantity instead of replacing one slogan with another.
Insulation restricts heat, but work can still increase internal energy. At a phase transition, added heat can change phase fractions without raising temperature. A negative standard reaction ΔG concerns reference activities; actual direction depends on composition, and rate depends on kinetic barriers. Even a favorable direction generally approaches equilibrium rather than guaranteeing complete conversion.
Sources and further reading
These optional references support the stated laws and model assumptions. The numerical examples and diagrams are authored teaching models, not experimental measurements or engineering specifications. Selected relevant passages were reviewed, not entire textbooks or a comprehensive research literature.
Selected energy-flow definitions and cooling versus heating coefficients of performance. The refrigerator paragraph says work done by the working substance, but its positive denominator Qh minus Qc is work input. Our equations explicitly use positive Win.
Selected Gibbs energy, reaction quotient, equilibrium constant, and temperature/composition distinctions. Our reaction has invented parameters. Interpret work with an explicit sign convention: this course counts output as positive. Reaction driving force is local to the current composition, not a finite complete-conversion energy.