In short

Heat is energy transferred from one body to another because they are at different temperatures. Its SI unit is the joule. The heat that must flow into a body of mass m to raise its temperature by \Delta T is

\boxed{\; Q = m c \Delta T \;}

where c is the specific heat capacity of the material (J/kg·K), a property of the substance that measures how much energy one kilogram of it stores per kelvin of temperature rise. Water has the highest specific heat of any common substance: c_{\text{water}} = 4186\,\text{J/kg·K}, about five times that of steel and ten times that of lead.

Put two bodies at different temperatures into thermal contact and (if nothing else can absorb or release heat) they exchange energy until they reach a common equilibrium temperature. Energy conservation then says

\boxed{\; \text{heat lost by hot body} = \text{heat gained by cold body}\;}

— the principle of calorimetry. Solving this single equation gives the final temperature of any mixture, the specific heat of an unknown metal, or the water equivalent of a calorimeter vessel.

Water's large c is the thermostat of the Indian coast: it is why Mumbai barely cools at night while Delhi drops ten degrees, why a single monsoon cloud carries enough latent and sensible heat to change the temperature of a small town, and why every power plant in India — thermal, nuclear, solar — moves its heat out through water and not through any other fluid.

Turn on the stove in an Indian kitchen and put a large tadka pan of mustard oil on one burner and a small steel katori of water on another. Feed both the same gas flame for two minutes. Drop a match-head of chilli into each. The chilli in the oil crackles and blackens almost instantly; the chilli in the water just floats there, barely warming. Both pots received the same amount of heat from the same flame for the same time — so why did one reach 180°C in two minutes and the other stay below 40°C?

The answer is that water and oil are very different things, thermally. Water refuses to get hot. You can pour joule after joule of energy into a bucket of water and see the temperature rise only slowly. Mustard oil, by contrast, takes what heat you give it and leaps upward in temperature. The same is true of every pair of materials you compare: hand a kilogram of aluminium the same energy you hand a kilogram of lead, and the aluminium goes up a few degrees where the lead goes up by twenty. The quantity that captures this difference — one number per material — is specific heat capacity, and the equations built on it describe every problem in which "a hot thing touches a cold thing and they settle at some middling temperature."

This article takes that story from the start. You should have read temperature and thermometers — the \Delta T in all the formulas below only makes sense once you have a scale to measure it on — and it helps to have met conservation of energy, because the central equation of calorimetry is energy conservation applied to a specific situation. We derive Q = mc\Delta T from the definition of heat, prove the mixing equation, work through three examples (bath water, a copper immersion heater, and a dropped iron shot — the textbook way of measuring the specific heat of an unknown metal), and close with why water's staggering c shapes climate, cooking, and the entire thermal infrastructure of modern India.

What heat is — and what it is not

Before any formula, pin down the word. In everyday language "heat" does double duty: a hot stove is "heat", a warm day is "heat", the energy that escapes from a cup of tea is also "heat". Physics uses it for exactly one thing: the energy transferred between two bodies because they are at different temperatures. Nothing else.

A body does not "contain heat." It contains internal energy — the total kinetic and potential energy of its molecules. When it is hotter than its surroundings, some of that internal energy flows out as heat; when it is colder, heat flows in. The same hot stove that "has a lot of heat" in colloquial usage is really a stove with a lot of internal energy and a temperature much higher than the kitchen's — so heat flows out of it, into the pan, into the dal, and eventually into the air of the room.

This distinction matters because the equations below track energy flowing across a boundary, not "how much heat a body has." A body at rest with no heat flowing in or out has zero heat transfer even though its internal energy is huge.

The SI unit of heat is therefore the same as any other energy — the joule. The older unit, the calorie, is defined to be the heat needed to warm 1 gram of water by 1 K; it equals 1\,\text{cal} = 4.186\,\text{J}. The food calorie (Cal, with a capital C) quoted on packets of biscuits and instant noodles is actually a kilocalorie: 1\,\text{Cal} = 1000\,\text{cal} = 4186\,\text{J}. Every physics equation in this article uses joules.

Specific heat — the resistance to heating

Hand a body some heat Q. Its temperature rises by some \Delta T. Experiment shows that, for reasonable temperature changes (not near a phase transition),

Write this as

\boxed{\; Q = m c \Delta T \;} \tag{1}

where c is the specific heat capacity of the material — the heat required to raise the temperature of 1 kg of the substance by 1 K. Its SI unit is \text{J}/(\text{kg·K}).

Why: divide through by m\Delta T and you recover the definition of c — heat per unit mass per unit temperature rise. It is an intrinsic material property, the same for any amount of the substance, like density. A kilogram of water has the same c as a milligram of water; the constant characterises the material, not the sample.

Some values to have in the back of your mind:

Substance c (J/kg·K)
Water (liquid) 4186
Ice (at 0°C) 2100
Steam (at 100°C) 2010
Mustard oil ≈ 2000
Milk ≈ 3900
Ghee ≈ 2200
Ethanol 2440
Human body (avg) 3500
Wood 1700
Aluminium 897
Glass 840
Iron / mild steel 450
Copper 385
Brass 377
Silver 235
Lead 129

The pattern is stark. Water's specific heat is enormous — larger than almost anything in daily use. Lead needs one joule to raise a gram's worth by 1 K; water needs over thirty. That single row of the table is why bucket-bath water stays warm all morning (you would have lost heat far faster if the bucket were filled with mustard oil), why a clay pot of thickened milk on the stovetop takes twenty minutes to come to a boil even though ghee in a pan reaches frying temperature in two, and why every thermal power plant in India cools its steam condensers with water rather than any other fluid cheap enough to flow in bulk.

Why water's specific heat is so large

This deserves a paragraph. Water molecules are held together by hydrogen bonds — a weak attraction between the partially positive hydrogen on one molecule and the partially negative oxygen on another. Pouring energy into liquid water goes partly into molecular translation (raising the temperature) but also partly into breaking and reforming these hydrogen bonds (at a given temperature the network is constantly rearranging). The bond-rearrangement sink absorbs a lot of heat per degree, so each kelvin of temperature rise costs more energy than it would for a liquid with no hydrogen bonding. Every comparable liquid — mustard oil, ghee, ethanol — lacks this network and has a correspondingly smaller c.

Bodies of water on the Earth's surface, stitched together by this hydrogen-bonded network, behave as a giant thermal flywheel. They absorb heat during the day without warming very much, release it during the night without cooling very much, and drag nearby land temperatures toward an average. This is why a summer afternoon in Mumbai (on the Arabian Sea) rarely crosses 34°C while Delhi (inland) routinely touches 45°C.

Heat capacity — the whole-body version

For a single object of mass m made of a specific material, it is sometimes handy to define the heat capacity C of that object:

C = mc, \qquad Q = C\Delta T.

Heat capacity is extensive — proportional to the amount of substance. Two kilograms of water has twice the heat capacity of one kilogram. Specific heat is intensive — independent of the amount. This terminology lets you say things like "the heat capacity of this room's air is about 5 kJ/K" without specifying a unit mass.

A special case that appears over and over: the water equivalent of a calorimeter. A calorimeter vessel (a glass or metal container) has its own heat capacity. You can always write C_{\text{calorimeter}} = m_e c_{\text{water}} where m_e is the hypothetical mass of water that would have the same heat capacity. m_e is called the water equivalent of the calorimeter. It saves arithmetic in mixing problems, because you can simply add m_e to the actual water mass and treat the whole thing as water.

The principle of calorimetry — from energy conservation

Put two bodies in thermal contact, insulate them from the rest of the world, and wait. What determines the final temperature?

Take body 1 at temperature T_1, mass m_1, specific heat c_1; body 2 at T_2, m_2, c_2. Assume T_1 > T_2. Heat flows from body 1 to body 2 until they reach a common equilibrium temperature T_f with T_2 < T_f < T_1.

Body 1 loses heat:

Q_{\text{lost}} = m_1 c_1 (T_1 - T_f).

Body 2 gains heat:

Q_{\text{gained}} = m_2 c_2 (T_f - T_2).

If the pair is thermally insulated from everything else, energy conservation says the heat that left body 1 must equal the heat that arrived at body 2 — no other pathway exists:

\boxed{\; m_1 c_1 (T_1 - T_f) = m_2 c_2 (T_f - T_2) \;} \tag{2}

Why: the only thing that can happen to the heat leaving body 1 is that it enters body 2. There is no third party to receive it, no chemical reaction converting it to other forms, no work being done against a piston. So "energy out of 1" equals "energy into 2", which is equation (2). Rearranged: heat lost = heat gained.

Solve for T_f:

T_f = \frac{m_1 c_1 T_1 + m_2 c_2 T_2}{m_1 c_1 + m_2 c_2}. \tag{3}

Why: expand equation (2), m_1c_1 T_1 - m_1 c_1 T_f = m_2 c_2 T_f - m_2 c_2 T_2, collect T_f terms on one side, factor. The final temperature is a weighted average of the starting temperatures, weighted by each body's heat capacity m_i c_i. Big heat capacities pull T_f toward them; small heat capacities give way.

This single equation is the whole of basic calorimetry. With multiple bodies the generalisation is immediate — heat lost by the hot ones equals heat gained by the cold ones, and T_f is the weighted average of all the starting temperatures:

T_f = \frac{\sum_i m_i c_i T_i}{\sum_i m_i c_i}.

Sign convention and the "always positive" trick

Students routinely trip over signs in calorimetry. The cleanest convention:

If you prefer, you can write every transfer as Q_i = m_i c_i (T_f - T_i) and require \sum_i Q_i = 0 — the "zero-sum" formulation. Hot bodies give negative Q, cold bodies positive, and they cancel. Both formulations are algebraically identical; pick whichever makes fewer arithmetic errors in your work.

Measuring an unknown specific heat — the method of mixtures

Suppose you have a metal of unknown composition — a brass nut, an iron washer, a copper coin — and you want its specific heat. The classical method, still used in every good undergraduate physics lab, is:

  1. Heat the metal to a known high temperature T_m by boiling it in a water bath (so T_m = 100°\text{C} in Delhi, about 93°C in Shimla where water boils lower).
  2. Have a calorimeter — a light metal cup, usually copper, insulated in a wooden box — holding a known mass m_w of water at a known temperature T_w.
  3. Drop the hot metal into the calorimeter. Stir. Watch the thermometer climb.
  4. Record the final equilibrium temperature T_f.

Now write heat lost = heat gained. Heat lost by the metal is m_m c_m (T_m - T_f). Heat gained by water plus calorimeter is (m_w c_w + C_{\text{cal}})(T_f - T_w), where C_{\text{cal}} is the calorimeter's heat capacity.

m_m c_m (T_m - T_f) = (m_w c_w + C_{\text{cal}})(T_f - T_w).

Solve for c_m:

\boxed{\; c_m = \frac{(m_w c_w + C_{\text{cal}})(T_f - T_w)}{m_m (T_m - T_f)} \;} \tag{4}

Why: every quantity on the right is measurable with a balance, a thermometer, and the known c_w = 4186\,\text{J/(kg·K)}. The only unknown is c_m, which equation (4) hands you. This is how the specific heats of every metal in the table above were first measured, and the experiment is still in every JEE-level practical syllabus.

Common sources of error, and how labs control them:

Worked examples

Example 1: How long to bring a bucket of bath water from 20°C to 40°C?

A steel bucket contains 15 litres of water at 20°C. You drop an immersion heater rated 1500 W into the bucket. Ignoring heat loss to the air and heating of the steel bucket itself (we will fix that assumption afterwards), how long does it take to warm the water to 40°C? Water has \rho = 1000\,\text{kg/m}^3 and c_w = 4186\,\text{J/(kg·K)}.

A steel bucket of water with an immersion heaterA cylindrical bucket containing water, with a rod-shaped immersion heater plugged into a wall socket via a cable. The heater dissipates electrical power P into the water.230 V1500 W15 L water, initially 20°C
Electrical energy is dissipated by the heating element into the water. Power times time equals the heat delivered; set that equal to $mc\Delta T$ to find the time.

Step 1. Compute the mass of water.

m = \rho V = 1000\,\text{kg/m}^3 \times 15 \times 10^{-3}\,\text{m}^3 = 15\,\text{kg}.

Why: 1 litre of water weighs 1 kg by definition at 4°C, near enough for this problem; 15 litres is 15 kg.

Step 2. Compute the heat required.

Q = mc\Delta T = 15 \times 4186 \times 20 = 1.26 \times 10^{6}\,\text{J} = 1.26\,\text{MJ}.

Why: \Delta T = 40 - 20 = 20 K. Plug in. Over a megajoule to heat a bath bucket — which sounds enormous but is exactly the energy in roughly 30 g of petrol.

Step 3. Set heat equal to power times time.

Q = Pt \implies t = \frac{Q}{P} = \frac{1.26\times 10^6}{1500} = 837\,\text{s} \approx 14\,\text{minutes}.

Why: the heater dissipates 1500 J every second; we need 1.26 million joules; divide. The time scales inversely with power, so a 2500 W heater would do the job in 8-9 minutes.

Step 4. What if we include the bucket?

A steel bucket of 1 kg has heat capacity 1 \times 450 = 450 J/K; to warm it through 20 K takes 9000 J. This is less than 1% of the 1.26 MJ, so the answer barely shifts — the steel's small c and modest mass makes it nearly invisible next to the water. For the same calculation with an aluminium karahi, C = 0.9\,\text{kg}\times 897 = 800\,\text{J/K}, still under 2%.

Result: about 14 minutes for the water to warm from 20 to 40°C. Including the bucket adds under a minute.

What this shows: the enormous c of water is what dominates almost every kitchen and bathroom heating problem. Whatever metal vessel you use is thermally "invisible" next to the water inside it — a useful rule of thumb. This is also why Indian household geysers are rated at 2500 W or more: with less, a morning bath takes too long.

Example 2: Mixing hot tea with cold milk

A chai wallah pours 200 mL of milk (specific heat 3900 J/kg·K, density 1030 kg/m³) at 10°C into a steel glass holding 250 mL of tea (essentially water, c = 4186\,\text{J/(kg·K)}) at 85°C. The steel glass has mass 120 g and c_{\text{steel}} = 450\,\text{J/(kg·K)} — and was already at 85°C from a previous rinse. What temperature does the final mixture settle at? Assume no heat loss to the air.

Pouring cold milk into hot tea to reach a uniform warm temperatureA steel tumbler originally containing hot tea at 85°C. Cold milk at 10°C is poured in. The mixture settles at a single equilibrium temperature between these values.tea (250 mL, 85°C)+ steel 120 gmilk 200 mL 10°C
Three masses come into thermal equilibrium: hot tea, cold milk, hot steel tumbler. Heat flows from tea and steel into the milk until a common temperature is reached.

Step 1. Convert volumes to masses.

  • Tea (water): m_t = 1000 \times 0.250 \times 10^{-3} = 0.250 kg.
  • Milk: m_m = 1030 \times 0.200\times 10^{-3} = 0.206 kg.
  • Steel glass: m_s = 0.120 kg.

Why: always convert to kg before plugging into the formula — mixing grams and kilograms is the single commonest arithmetic error in calorimetry problems.

Step 2. Identify who is hot and who is cold, and write heat lost = heat gained.

  • Hot (starting at 85°C): tea and steel glass. Heat lost = (m_t c_t + m_s c_s)(85 - T_f).
  • Cold (starting at 10°C): milk. Heat gained = m_m c_m (T_f - 10).

Set them equal:

(m_t c_t + m_s c_s)(85 - T_f) = m_m c_m (T_f - 10).

Why: the tea and the steel tumbler are initially at the same temperature, so they lose heat together as if they were one combined body of effective heat capacity m_t c_t + m_s c_s. This combining-of-heat-capacities trick is used constantly in calorimetry.

Step 3. Plug in numbers.

  • m_t c_t = 0.250 \times 4186 = 1046.5 J/K.
  • m_s c_s = 0.120 \times 450 = 54 J/K.
  • m_m c_m = 0.206 \times 3900 = 803.4 J/K.

Combined hot side: 1046.5 + 54 = 1100.5 J/K.

1100.5 \times (85 - T_f) = 803.4 \times (T_f - 10).

Step 4. Solve for T_f.

Expand: 93542.5 - 1100.5\, T_f = 803.4\, T_f - 8034.

93542.5 + 8034 = (803.4 + 1100.5)\, T_f
101576.5 = 1903.9\, T_f
T_f = \frac{101576.5}{1903.9} \approx 53.4{ }^\circ\text{C}.

Why: this is equation (3) with three bodies collapsed to two by combining the tea and glass. The final temperature is a weighted average: heat-capacity-weighted average of 85 and 10 is, because the hot side's capacity is larger, pulled slightly toward 85.

Result: the chai settles at about 53°C — a reasonable drinking temperature.

What this shows: a cold milk addition can cool a scalding cup of chai to drinkable temperature in seconds, and equation (3) tells you exactly how much. Chai wallahs do this intuitively. You can do it quantitatively. Notice the steel glass contributed only about 5% of the hot-side heat capacity — as predicted, the metal vessel is thermally almost invisible next to the liquid contents.

Example 3: The specific heat of an unknown metal

A student drops a 250 g piece of an unknown metal, heated to 98°C in boiling water, into a copper calorimeter of mass 150 g (water equivalent 14 g) containing 300 g of water initially at 22°C. The final equilibrium temperature is 28.5°C. Find the specific heat of the metal. c_w = 4186\,\text{J/(kg·K)}.

Method of mixtures for measuring an unknown specific heatA hot metal block at 98°C is dropped into a copper calorimeter containing water at 22°C. The system reaches 28.5°C. The specific heat of the metal is found from heat lost equals heat gained.250 g, 98°Ccopper calorimeter300 g water, 22°C→ final 28.5°C
Heat flows from the hot metal block into the water and the copper calorimeter, all reaching a common final temperature. The equation heat lost = heat gained yields the unknown specific heat.

Step 1. Convert masses and identify heat capacities.

  • m_m = 0.250 kg (unknown metal), initially at T_m = 98°\text{C}.
  • m_w = 0.300 kg of water, initial T_w = 22°\text{C}.
  • Calorimeter water equivalent: m_e = 0.014 kg — so the calorimeter's heat capacity is m_e c_w = 0.014 \times 4186 = 58.6 J/K.
  • Final temperature T_f = 28.5°\text{C}.

Step 2. Apply equation (4).

Heat gained by water plus calorimeter:

Q_{\text{gained}} = (m_w + m_e) c_w (T_f - T_w) = (0.300 + 0.014) \times 4186 \times (28.5 - 22).
= 0.314 \times 4186 \times 6.5 = 8544\,\text{J}.

Why: using the water equivalent m_e lets us bundle the calorimeter's heat capacity with the water's. Both gain heat, both reach T_f from T_w, so we can add them and treat as one 0.314 kg of effective water.

Step 3. Heat lost by the metal.

Q_{\text{lost}} = m_m c_m (T_m - T_f) = 0.250 \times c_m \times (98 - 28.5) = 0.250 \times c_m \times 69.5 = 17.375\,c_m.

Why: the metal cools by 98 - 28.5 = 69.5 K. The product m_m \Delta T = 17.375 kg·K is a known number; the only unknown in Q_{\text{lost}} is c_m.

Step 4. Set Q_{\text{lost}} = Q_{\text{gained}} and solve.

17.375\, c_m = 8544 \implies c_m = \frac{8544}{17.375} \approx 492\,\text{J/(kg·K)}.

Why: arithmetic. The answer lies between copper (385) and iron (450) — it is suspiciously close to mild steel / iron, which has a tabulated value of about 450-490 J/(kg·K). Our student has probably been given a chunk of steel.

Step 5. Estimate uncertainty.

Suppose each temperature reading is good to \pm 0.5 K. Then the crucial difference T_f - T_w = 6.5 \pm 0.7 K — a relative error of about 11%. Similarly, T_m - T_f = 69.5 \pm 0.7 K, only 1% error. The dominant error is in the small cold-side temperature rise. Propagating, c_m is known to roughly \pm 12\%, so c_m = 492 \pm 60\,\text{J/(kg·K)}. That error bar comfortably covers the true value for mild steel.

Result: c_m \approx 490\,\text{J/(kg·K)}, consistent with mild steel or iron.

What this shows: the method of mixtures gives the specific heat of any metal with a simple thermometer, a balance, and a calorimeter. The experiment's accuracy is usually limited by how well you measure the smallest temperature change — the cold-side rise — which is exactly why you want a small mass of water and a large mass of metal, to make that rise big enough to measure well.

Explore calorimetric mixing yourself

The interactive below lets you drag the mass ratio m_{\text{hot}}/m_{\text{cold}} (with water on both sides for simplicity — c cancels) and watch the equilibrium temperature move. Two masses of water are brought into contact at 80°C and 20°C; the final temperature depends only on the mass ratio.

Interactive: equilibrium temperature as a function of mass ratio Curve showing the equilibrium temperature Tf as a function of the mass ratio r = m_hot/m_cold. Hot water is at 80°C, cold at 20°C, both with the same specific heat. mass ratio m_hot / m_cold final temperature T_f (°C) 0 20 50 80 1 5 10 hot 80°C cold 20°C T_f (°C) drag the red point
Drag the red point to change the mass ratio. At ratio 1 (equal masses), $T_f$ is exactly halfway between 20 and 80 — namely 50°C. As the hot mass gets larger, $T_f$ approaches 80°C; as it gets smaller, $T_f$ approaches 20°C. The curve is a simple weighted average.

Common confusions

If you came here to solve textbook calorimetry and set up simple mixing problems, you already have what you need. What follows is for readers who want to see the subtleties: the difference between c_p and c_v, how Dulong–Petit connects specific heats of solid elements to atomic weight, why water's c is literally unique among common liquids, and how the method of mixtures extends to non-adiabatic calorimeters.

Two specific heats: at constant pressure and constant volume

For solids and liquids, one number c suffices for all practical purposes. For gases, it does not — the specific heat depends on whether you allow the gas to expand while you heat it.

  • c_v (specific heat at constant volume): the heat required per unit mass per kelvin while the gas's volume is held fixed. All the heat goes into internal energy; none into doing work on the surroundings.
  • c_p (specific heat at constant pressure): the heat required per unit mass per kelvin while the gas is allowed to expand so its pressure stays constant. Some of the heat goes into internal energy and some into pushing back the surroundings — so c_p > c_v.

For an ideal gas with n moles and N atoms per molecule, the molar specific heats are

C_{v,\text{molar}} = \tfrac{f}{2}R, \qquad C_p = C_v + R,

where f is the number of degrees of freedom (3 for monatomic gases like argon; 5 for diatomic gases like \text{O}_2 and \text{N}_2 at room temperature; higher for polyatomic). R is the gas constant, 8.31 J/(mol·K). The ratio \gamma = C_p/C_v is the adiabatic index, central to the physics of sound speed in air and of engine cycles.

For liquids and solids, the volume change on heating is tiny (the \Delta V \approx \gamma V_0\Delta T of the previous article, with \gamma_{\text{solid}} \sim 10^{-5}\,\text{K}^{-1}), so c_p - c_v is negligible. A single c suffices.

Dulong–Petit: the molar specific heat of solid elements

A remarkable empirical observation made by Pierre Dulong and Alexis Petit in 1819: the molar specific heat of most solid elements at room temperature is very nearly 3R \approx 25\,\text{J/(mol·K)}. Divide by atomic mass to get the per-kilogram specific heat; the lighter the atom, the larger the c.

Element Atomic mass (g/mol) c (J/kg·K) Molar c (J/mol·K)
Aluminium 27 897 24.2
Iron 56 450 25.2
Copper 63.5 385 24.5
Silver 108 235 25.4
Lead 207 129 26.7

Every molar specific heat hovers around 25 — a direct experimental confirmation of the equipartition theorem for solids (each atom has 6 quadratic terms in energy, giving \tfrac{6}{2}R = 3R per mole). Dulong–Petit fails at low temperature, where quantum effects freeze out modes and c falls toward zero as T^3 — a lovely consistency between classical statistical mechanics and the Debye theory that followed a century later.

This also means that knowing the specific heat of an unknown elemental solid to within about 10% is enough to estimate its atomic weight — the method Mendeleev used to correct the atomic weights of several elements when constructing the periodic table.

Water is special — the hydrogen bond audit

We said water has a large c because hydrogen bonds store energy. Quantitatively: the covalent O{-}H bond in a water molecule is about 460 kJ/mol strong, but each inter-molecular hydrogen bond is only about 20 kJ/mol. In liquid water at room temperature each molecule is hydrogen-bonded to roughly 3.5 neighbours (on average, continually rearranging). That gives an energy reservoir of about \tfrac{1}{2}\cdot 3.5 \cdot 20\,\text{kJ/mol} \approx 35\,\text{kJ/mol} that can be modestly rearranged without breaking bonds.

A useful exercise: compare water's c to that of liquid methane or neon at cryogenic temperatures (each a liquid made of molecules that cannot hydrogen-bond because they have no O-H or N-H bonds). Those liquids have molar c's of about 50-60 J/(mol·K), whereas water is 75\,\text{J/(mol·K)} — higher by the 15-20 J expected from the hydrogen-bond network. Hydrogen bonding is literally the source of water's excess c.

This has consequences far beyond the laboratory. Water moderates Earth's climate because each kilogram of the ocean can absorb four times the heat per kelvin of a kilogram of rock. Without hydrogen bonding — without water as we know it — the oceans would swing between boiling and freezing every season. The stability of our climate is in large part a consequence of the unique molecular physics of H_2O.

Non-adiabatic corrections — Newton's law of cooling

Real calorimeters are not perfectly insulated. They lose heat continuously to the surroundings at a rate proportional to the temperature difference — Newton's law of cooling:

\frac{dT}{dt} = -k(T - T_{\text{room}}).

In the method of mixtures, this means the thermometer's reading overshoots the moment of true equilibrium and then drifts downward as the calorimeter loses heat. A careful experimenter records the thermometer for several minutes after the initial settling, plots the cooling curve, and extrapolates backward to the instant of mixing to get the "true" equilibrium T_f. The correction is usually small (fractions of a degree for a well-insulated calorimeter) but crucial for precision work.

A more elegant approach, used in physics-olympiad and research-grade calorimeters, is isothermal calorimetry: the calorimeter is kept in a water bath at exactly the room temperature, so T - T_{\text{room}} = 0 and Newton's law gives zero loss. The system reaches equilibrium with no drift. The Bose Institute in Kolkata still runs isothermal calorimeters of a 1920s design for undergraduate practicals — the method is mature.

Heat capacity of a bulk room of air

A calculation to check intuition: the mass of air in a 3×4×3 m^3 bedroom at 25°C is about 3\times 4\times 3 \times 1.2\,\text{kg/m}^3 = 43 kg. Air's c_p \approx 1005 J/(kg·K). So the room-air heat capacity is 43 \times 1005 \approx 43\,\text{kJ/K}. To cool that room by 10 K (a typical AC target) takes 430 kJ, about the work of a 1.5 kW AC running at full cooling for 5 minutes — matching what every Indian flat-dweller has noticed: a sensible AC takes about 5 minutes to noticeably cool a room on a 40°C afternoon. The factor-of-2 discrepancy (real rooms take 15 minutes, not 5) is because the walls, furniture, and floor — all warm — add another ≈100 kJ/K of heat capacity that the AC must also overcome.

This is the whole power of equation (1): once you know mc for a system, you know how responsive it is to heat flow, and you can design appliances, estimate comfort times, and set up engineering problems with no further information.

Where this leads next