Surface Tension & Capillarity

A steel needle is eight times denser than water. Archimedes says it must sink — and dropped point-first, it does. But lay it gently, flat on the surface, and it rests there, cradled in a visible dimple, as if the water had a skin. Pond skaters run on that skin all day on legs that never get wet. Yet the moment you add a drop of dishwashing liquid, the needle plunges and the insect is in mortal danger. Whatever this "skin" is, it is real, it is measurable — and soap switches it off. The skin is surface tension: the surface of a liquid behaves like a stretched elastic membrane, because surface molecules are energetically short-changed. A molecule in the bulk is pulled by neighbours on all sides; a molecule at the surface has lost its upstairs neighbours and clings harder to the rest. Surfaces therefore cost energy, and liquids do everything they can to shrink them — which is why free drops are spheres, why bubbles pop, and why water climbs up narrow tubes against gravity, a trick called capillarity that waters every plant you have ever seen. This is a CBSE/ISC derivation-and-definition favourite and a JEE/NEET numerical factory: capillary rise, work to stretch films, drops splitting and merging. (It is also, pleasingly, a chapter founded partly at a kitchen sink — meet Agnes Pockels below.) The excess-pressure-in-bubbles machinery gets its own archetype next; here we build the concepts it will spend.

Key takeaways
  • Surface molecules have fewer attracting neighbours than bulk molecules, so every square metre of liquid surface costs energy — liquids minimize their surface area.
  • Surface tension TT is that cost expressed two equivalent ways: force per unit length (T=F/LT = F/L, N/m) along any line in the surface, or energy per unit area (T=W/ΔAT = W/\Delta A, J/m²).
  • A soap film has two surfaces: stretching work W=T (2 ΔAfilm)W = T\,(2\,\Delta A_{\text{film}}), and film pulls on a wire of length LL with 2TL2TL.
  • Contact angle θ\theta decides wetting: θ<90°\theta < 90° (water on glass) — liquid climbs; θ>90°\theta > 90° (mercury on glass) — liquid retreats.
  • Capillary rise (Jurin's law): h=2Tcos⁡θρgrh = \dfrac{2T\cos\theta}{\rho g r} — thinner tube, higher climb; mercury descends.
  • Heating weakens cohesion: TT decreases with temperature; surfactants (soap) slash it.
  • Key idea: liquid surfaces cost energy, so they act like stretched skins — pulling flat, pulling spherical, and hauling liquid up capillaries.
The core insight

Because surface molecules lack neighbours above, a liquid surface carries an energy cost per unit area — equivalently a force per unit length TT — making it behave as a stretched membrane that minimizes area, sets contact angles, and drives capillary rise h=2Tcos⁡θ/ρgrh = 2T\cos\theta/\rho g r.

Where this lives today

The highest-stakes application of this article is in newborn lungs. Alveoli are micrometre-scale wet bubbles; by the curvature–pressure logic previewed above, small wet bubbles need large pressures to stay open. Healthy lungs coat themselves with a pulmonary surfactant that slashes surface tension exactly where alveoli are smallest; premature babies born before the coating develops suffer respiratory distress syndrome, and the therapy — surfactant replacement, delivered since the 1980s — is this chapter administered by nebulizer, credited with cutting RDS mortality several-fold. Elsewhere: inkjet printers fire picolitre drops whose formation and satellite-droplet behaviour are pure surface-tension engineering; lab-on-a-chip diagnostics move blood through capillary channels with no pump at all — Jurin's law as plumbing; water-repellent "lotus-effect" fabrics and self-cleaning glass tune contact angles past 150°; and espresso machines, fountain pens, candle wicks, and paper towels all run on capillarity you can now price with equation (3).

The physical picture

Follow one water molecule. Deep in the bulk, it is bonded to neighbours in every direction — pulled equally all around, net force zero, energetically content. Lift it to the surface and half its social circle vanishes: no neighbours above, only air's occasional molecule. It keeps all its downward and sideways attractions but loses the upward ones. Two consequences: it is pulled inward (which is why surfaces flatten and drops round themselves), and it sits at higher energy than a bulk molecule — creating new surface means promoting molecules into this expensive position. So a liquid surface is a workforce of underpaid molecules, and the liquid economizes: it exposes the fewest molecules it can. A free drop takes the shape with the least surface for its volume — a sphere. A film stretched on a wire frame snaps flat. Two touching droplets merge (less total surface than two). And along any line you draw in the surface, the molecules on one side pull the molecules on the other side with a tangential force proportional to the line's length: the "skin tension" the needle and the pond skater ride on.

Figure
The molecular origin — surface molecules are missing their upstairs neighbours
surface bulk: pulled all ways — net zero surface: no neighbours above — net inward pull, higher energy strider's foot dimpled skin pulls up along the contact line — T per metre surface acts as a stretched membrane: tension T along every line in the surface Surfaces cost energy — liquids expose as few molecules as they can

One more character enters when a liquid meets a solid wall: the contact angle θ\theta, the angle (measured inside the liquid) between the solid surface and the liquid surface at the line where they meet. It is set by a tug-of-war between cohesion (liquid–liquid attraction) and adhesion (liquid–solid attraction). Water on clean glass: adhesion wins, the water spreads and climbs the wall, θ≈0°\theta \approx 0°–8°8° — a wetting liquid with a concave meniscus. Mercury on glass: cohesion wins overwhelmingly, the mercury balls up and pulls away, θ≈140°\theta \approx 140° — non-wetting, convex meniscus. Water on a waxed jacket: θ>90°\theta > 90° by design, so rain beads and rolls. The contact angle is the boundary condition that decides which way capillarity goes.

The math: one constant, two costumes

Surface tension TT (some books write SS or γ\gamma) is defined along a line: if you draw any line of length LL in a liquid surface, the surface on one side pulls the surface on the other side with force

F=T L(1)F = T\,L \qquad (1)

where: TT — surface tension in N/m, tangent to the surface, perpendicular to the line. Water's T≈0.072T \approx 0.072 N/m at room temperature; soap solution ~0.030; mercury a muscular 0.465. These are small forces — 7 grams-weight per metre of line for water — which is why surface tension rules the millimetre world (insects, drops, capillaries) and is irrelevant to ships. The same constant wears an energy costume. Creating new surface area ΔA\Delta A promotes molecules to the expensive surface layer, costing work

W=T ΔA(2)W = T\,\Delta A \qquad (2)

where: TT now reads as surface energy per unit area, J/m². The units agree — N/m=J/m2\mathrm{N/m} = \mathrm{J/m^2} — and so does the physics: a membrane that costs energy per area is a membrane under tension. Exam questions switch costumes freely: "force on a wire" wants (1), "work to split a drop" wants (2).

Derivation AThe film factor of two — stretching a soap film on a sliding wire

We derive the force–energy equivalence and the most-tested subtlety of the chapter: a film is a liquid sandwich with two faces. Setup: a U-shaped wire frame with a frictionless slider of length LL carries a soap film. The film pulls the slider inward; we compute the force to hold it, then the work to pull it out a distance xx, and check the two costumes agree. Watch every appearance of the 2.

Step 1. The film has a front surface and a back surface. Each pulls the slider with T LT\,L (equation 1), so holding the slider still takes

F=2 T LF = 2\,T\,L

Step 2. Pull the slider out by xx at constant speed. Work done by you: W=Fx=2TLxW = Fx = 2TLx.

Step 3. New area created: the film's rectangle grows by LxLx — but both faces grow, so the new surface area is ΔA=2Lx\Delta A = 2Lx. Energy cost at TT per unit area: W=T(2Lx)W = T(2Lx) ✓ — identical. The two costumes are one constant. Why the 2 is physical, not bookkeeping: a bubble and a film really do have two liquid–air interfaces separated by a thin water layer; a drop or a tank surface has one. Every classic slip in this chapter — film pull, bubble pressure (4T/r vs 2T/r), needle lift-off — is a miscounted 2. *Verified: units N = (N/m)(m) ✓; the measured pull on a 5 cm slider in soap solution: 2(0.030)(0.05)=3×10−32(0.030)(0.05) = 3\times10^{-3} N ≈ 0.3 gram-weight — matches the classroom demo's feel. Standard: NCERT Ch. 9 §9.6.*

Drops, bubbles, and pressure — the preview. A curved surface under tension squeezes what it wraps: the inside of a drop sits at higher pressure than the outside, by ΔP=2T/r\Delta P = 2T/r for a liquid drop and 4T/r4T/r for a soap bubble (two surfaces again). These results — and the lovely traps hiding in them (smaller bubble, bigger pressure!) — are derived and drilled in their own archetype at the end of this unit; here, carry only the qualitative point: tension + curvature = pressure difference. It is the engine behind capillarity, coming right now.

Capillarity: the climb up the tube

Dip a clean glass capillary (bore radius rr) into water. Adhesion curves the water's edge up the glass (θ≈0\theta \approx 0), giving the surface a concave meniscus — approximately a hemisphere of radius ≈r\approx r. The pressure just below that curved surface is lower than atmospheric (the curved skin holds the difference), so the atmosphere pushes water up the tube until the weight of the raised column repays the pressure deficit. Equilibrium height is Jurin's law:

h=2Tcos⁡θρ g r(3)h = \frac{2T\cos\theta}{\rho\, g\, r} \qquad (3)

where: θ\theta — the contact angle (cos⁡θ≈1\cos\theta \approx 1 for water on clean glass); rr — the tube's inner radius (not diameter, not the meniscus of some other geometry); ρ\rho — the liquid's density. Read the scalings: thinner is higher (h∝1/rh \propto 1/r — halve the bore, double the climb), and the sign of cos⁡θ\cos\theta steers everything: mercury's θ≈140°\theta \approx 140° makes cos⁡θ<0\cos\theta < 0, so h<0h < 0 — mercury is depressed below the outside level, the meniscus convex. That sign flip is a fixture of assertion–reason questions.

Figure
Capillarity both ways — water climbs the glass, mercury refuses
h = 2T cosθ / ρgr concave meniscus, θ ≈ 0 water in glass: RISES depressed: θ ≈ 140° cos θ < 0 ⇒ h < 0 mercury in glass: DESCENDS, convex dome
Go deeper · Derivation B · 1 minJurin's law by weighing the lifted column
Derivation BJurin's law by weighing the lifted column

We derive equation (3) by the force route — balancing the surface-tension pull against the lifted water's weight — the derivation CBSE marking schemes expect. Setup: tube radius rr, contact angle θ\theta, liquid density ρ\rho, rise height hh (to the bottom of the meniscus). Strategy: the surface tension acts along the contact line where water meets glass; take its vertical component and equate to the column's weight. Watch which length the "per unit length" applies to: the contact line is the tube's inner circumference.

Step 1. The liquid surface meets the glass along a circle of length 2πr2\pi r. The surface pulls along its own plane — at angle θ\theta to the wall — so the vertical component of the pull on the liquid is

Fup=T (2πr)cos⁡θF_{\text{up}} = T\,(2\pi r)\cos\theta

Step 2. Weight of the lifted column of height hh:

W=ρ g (πr2h)W = \rho\,g\,(\pi r^2 h)

Step 3. Equilibrium — set them equal and solve:

T(2πr)cos⁡θ=ρgπr2h⇒h=2Tcos⁡θρgrT(2\pi r)\cos\theta = \rho g \pi r^2 h \quad\Rightarrow\quad h = \frac{2T\cos\theta}{\rho g r}

Why 1/r1/r: the driving force lives on the circumference (∝r\propto r) while the bill is the column's volume (∝r2\propto r^2) — narrow tubes have more perimeter per unit of payload. The same perimeter-versus-volume logic decided terminal velocity's r2r^2 last article; small worlds are ruled by their surfaces. *Why mercury descends: for θ>90°\theta > 90° the vertical component points down — the surface pushes the column below the outside level until the "negative column" balances. Verified: units — (N/m)/(kg/m3⋅m/s2⋅m)=m(\mathrm{N/m})/(\mathrm{kg/m^3 \cdot m/s^2 \cdot m}) = \mathrm{m} ✓. Anchor — water, r=0.20r = 0.20 mm: h=2(0.072)/(1000)(9.8)(2.0×10−4)=7.3h = 2(0.072)/(1000)(9.8)(2.0\times10^{-4}) = 7.3 cm, recomputed in worked example 1 ✓. Standard: NCERT Ch. 9; the pressure-route derivation (via 2T/r2T/r across the meniscus) gives the identical result and lives in the bubbles archetype.*

Limiting cases

caseresultmeaning
r→r/2r \to r/2hh doublescapillarity is a thin-tube phenomenon; in a bucket it's invisible
θ<90°\theta < 90° (wetting)h>0h > 0, concave meniscuswater in glass: climbs
θ>90°\theta > 90° (non-wetting)h<0h < 0, convex meniscusmercury in glass: descends
θ=90°\theta = 90° exactlyh=0h = 0, flat meniscussilver–water case; no climb either way
soap addedTT drops ~2×; hh dropssurfactants sabotage the skin — the sinking needle
temperature upTT down (cohesion loosens)hot water wets and spreads better; T→0T \to 0 at boiling-critical point
tube shorter than hhliquid rises to the top, meniscus flattens, does NOT overflowthe curvature adjusts to the available height — a JEE favourite

Worked examples

1 · Board level — capillary rise (CBSE/ICSE pattern, 3 marks)

A clean glass capillary of inner radius 0.20 mm0.20\ \mathrm{mm} is dipped vertically into water (T=0.072 N/mT = 0.072\ \mathrm{N/m}, ρ=1000 kg/m3\rho = 1000\ \mathrm{kg/m^3}, contact angle ≈0°\approx 0°, g=9.8 m/s2g = 9.8\ \mathrm{m/s^2}). Find the height of capillary rise.

Step 1. Convert the radius: r=0.20 mm=2.0×10−4 mr = 0.20\ \mathrm{mm} = 2.0\times10^{-4}\ \mathrm{m}. (Check the question's word: radius given here; had it said diameter, halve first.)

Step 2. Jurin's law with cos⁡0°=1\cos 0° = 1: h=2Tρgr=2×0.0721000×9.8×2.0×10−4h = \dfrac{2T}{\rho g r} = \dfrac{2\times0.072}{1000\times9.8\times2.0\times10^{-4}}.

Step 3. h=0.1441.96=7.3×10−2 m≈7.3 cmh = \dfrac{0.144}{1.96} = \boxed{7.3\times10^{-2}\ \mathrm{m}} \approx 7.3\ \mathrm{cm}.

Step 4. Sanity: a fifth-of-a-millimetre bore lifts water 7 cm; xylem vessels in plants are finer still — capillarity is one of the mechanisms sap exploits. In a 2 mm bore the rise would be a barely visible 7 mm: the 1/r1/r law in action.

2 · Competitive level — work to atomize a drop (JEE-Main/NEET pattern, ~2.5 minutes)

A spherical water drop of radius R=1.0 cmR = 1.0\ \mathrm{cm} is sprayed into n=1000n = 1000 identical droplets (T=0.072 N/mT = 0.072\ \mathrm{N/m}). Find (a) the radius of each droplet and (b) the work required. (c) Does the water warm or cool slightly as this happens?

Step 1. (a) Volume conservation: n⋅43πr3=43πR3⇒r=R n−1/3=1.0 cm/10=1.0 mmn\cdot\tfrac43\pi r^3 = \tfrac43\pi R^3 \Rightarrow r = R\,n^{-1/3} = 1.0\ \mathrm{cm}/10 = \boxed{1.0\ \mathrm{mm}}.

Step 2. (b) Surface area jumps from 4πR24\pi R^2 to n⋅4πr2=4πR2 n1/3n\cdot4\pi r^2 = 4\pi R^2\, n^{1/3}: a factor of n1/3=10n^{1/3} = 10.

Step 3. Work = T ΔA=T 4πR2 (n1/3−1)=0.072×4π×(10−2)2×9=8.1×10−4 JT\,\Delta A = T\,4\pi R^2\,(n^{1/3} - 1) = 0.072\times4\pi\times(10^{-2})^2\times9 = \boxed{8.1\times10^{-4}\ \mathrm{J}}.

Step 4. (c) You supplied energy that now lives in the enlarged surface; the process absorbs energy, and if done adiabatically the liquid cools slightly. (Merging droplets — the reverse — releases surface energy and warms: that direction powers a whole archetype later in this unit.) Timing note: memorize the skeleton W=4πR2T(n1/3−1)W = 4\pi R^2 T(n^{1/3}-1) — deriving it via volume conservation eats a minute you can spend elsewhere. The paired formula for merging is the same with the sign flipped. Both are JEE-Main regulars.

3 · Competitive level — lifting a ring off water (JEE-Main pattern, ~2 minutes)

A thin wire ring of radius 5.0 cm5.0\ \mathrm{cm} and weight 0.030 N0.030\ \mathrm{N} rests flat on a water surface. Find the minimum vertical force needed to pull it free (T=0.072 N/mT = 0.072\ \mathrm{N/m}; the water wets the wire).

Step 1. As the ring lifts, it drags up a collar of water attached along both the inner and outer edges of the wire — two contact circles, each of circumference 2πr2\pi r (thin wire: both ≈ same rr).

Step 2. Surface-tension pull-down at breakaway: FT=T×2(2πr)=0.072×2×2π×0.05=4.5×10−2 NF_T = T \times 2(2\pi r) = 0.072\times2\times2\pi\times0.05 = 4.5\times10^{-2}\ \mathrm{N}.

Step 3. Total minimum pull: F=W+FT=0.030+0.045=7.5×10−2 NF = W + F_T = 0.030 + 0.045 = \boxed{7.5\times10^{-2}\ \mathrm{N}}.

Step 4. Note the structure: the tension term (0.0450.045 N) exceeds the ring's weight — at centimetre scales the skin outmuscles gravity, which is the entire theme of this chapter. Timing note: the examiner's coin-flip is the contact-line count — a ring or a needle pulls TWO lines of water (factor 2); a vertical plate's perimeter counts once around. Draw one cross-section, count the liquid junctions, then compute. Ninety seconds with the picture; a guaranteed error without it.

Exam traps

The factor of 2 census. Film on a slider: 2 surfaces → 2TL2TL. Soap bubble: 2 surfaces → 4T/r4T/r. Water drop or air bubble inside water: 1 surface → 2T/r2T/r. Ring/needle pulled off water: 2 contact lines. Miscounting the 2 is this chapter's signature error — take the census before substituting.

Radius vs diameter in Jurin's law. h=2Tcos⁡θ/ρgrh = 2T\cos\theta/\rho g r uses the tube's inner radius. Questions love giving the diameter.

Units of TT. N/m, equivalently J/m² — not N/m² (that's pressure/stress). An option quoting surface tension in pascals is an auto-eliminate.

Contact angle direction. Measured inside the liquid, from the solid wall to the surface tangent. Wetting: θ<90°\theta < 90°, concave meniscus, rise. Non-wetting: θ>90°\theta > 90°, convex, depression. Mercury numericals arrive with θ≈140°\theta \approx 140° and cos⁡θ≈−0.77\cos\theta \approx -0.77: keep the minus sign and report a depression.

Short-tube "overflow". If the tube's length above the liquid is less than the predicted hh, water rises to the top and the meniscus reduces its curvature until the balance closes — it never fountains out. Energy conservation is not negotiable; this is a stock assertion–reason item.

Temperature and impurities. Heating lowers TT (near-linear over exam ranges); soluble surfactants lower it dramatically; highly soluble salts raise it slightly. "Hot soapy water cleans best" is this trap's real-world mnemonic.

Small means surface-ruled. Surface forces scale with length; weights scale with volume. Below a few millimetres, tension wins (striders, drops, capillaries); above, gravity wins (lakes are flat). If your capillary-rise answer is in metres for a millimetre bore, re-check powers of ten.

Which gg. CBSE 9.8; JEE/NEET usually 10. Jurin's hh scales as 1/g1/g, so the choice shifts answers by 2% — match the paper's declaration.

Origins · 1 minAgnes Pockels and how this was discovered
AP
Agnes Pockels
1862–1935 · Braunschweig, Germany

Agnes Pockels was barred from university — her family judged higher education improper for a woman — so she did physics in the kitchen. Washing dishes, she noticed how films and greases changed water's surface, and built (in 1882, aged twenty) the "Pockels trough": a shallow tray with movable barriers to compress a surface film and a button-on-a-balance to measure the surface's pull. With it she quantified how oils and soaps alter surface tension, monolayer by monolayer. In 1891 she wrote — in German, apologetically — to Lord Rayleigh, who was so impressed he had the letter translated and published in Nature. Her trough, refined by Irving Langmuir (Nobel 1932), remains standard lab equipment as the Langmuir–Pockels trough: kitchen-sink science that founded surface chemistry.

History
1718
— James Jurin establishes that capillary rise varies inversely with tube radius — Jurin's law.
1805
— Thomas Young and Pierre-Simon Laplace connect surface curvature to pressure difference: the Young–Laplace law behind menisci and bubbles.
1871
— Joseph Plateau's soap-film experiments codify the geometry of minimal surfaces (Plateau's laws of film junctions).
1891
— Agnes Pockels' kitchen-trough measurements of surface films appear in Nature at Rayleigh's urging.
1917
— Irving Langmuir turns monolayer studies into quantitative surface chemistry (Nobel Prize, 1932).

Going further

Texts. NCERT Physics Class 11, Ch. 9 §9.6 · HC Verma Vol. 1, Ch. 14 · Resnick–Halliday–Walker Ch. 14. The excess-pressure machinery (2T/r2T/r, 4T/r4T/r, joined bubbles) is derived and drilled in the Bubbles, Drops & Excess Pressure archetype.

Provenance. All results standard. A verifier should recompute: the slider's 2TLx=T(2Lx)2TLx = T(2Lx) agreement; example 1's h=0.144/1.96=7.3h = 0.144/1.96 = 7.3 cm; example 2's r=1r = 1 mm, ΔA\Delta A factor 10, W=8.1×10−4W = 8.1\times10^{-4} J; example 3's FT=0.045F_T = 0.045 N and total 0.075 N.

Prerequisites. Fluids at Rest: Pressure & Pascal's Law. Leads to. Bubbles, Drops & Excess Pressure (archetype) · Drops Merging & Terminal Velocity (archetype).

If the surface is like a stretched membrane, why doesn't it snap back like rubber when I poke it?

It does — watch the dimple under a strider's foot vanish when the foot lifts. The difference from rubber: a rubber sheet's tension grows as you stretch it (more strain, more stress), while a liquid surface's tension is constant — stretching a surface doesn't strain existing bonds, it recruits new molecules up from the bulk at a fixed energy price per unit area. That's why TT is a material constant, not a Hooke's-law spring "constant" that depends on deformation.

How does the needle actually float? Is it buoyancy?

Mostly not. The gently laid needle never breaks the surface: the water's skin deforms into a trough under it, and the surface tension along the two lines of contact pulls upward (plus a little buoyancy from the displaced trough volume). Estimate: a 5 cm needle enjoys up to 2TL≈2(0.072)(0.05)=7.2×10−32TL \approx 2(0.072)(0.05) = 7.2\times10^{-3} N of skin support — enough for a light needle's ~5×10−35\times10^{-3} N weight, with the dimple's geometry making up angles. Add detergent, halve TT, and the ledger fails: down it goes. Dropped point-first, the skin is pierced instantly and plain Archimedes takes over.

Why are raindrops spheres — and are they really teardrop-shaped?

A free liquid mass minimizes surface energy, and the sphere holds a given volume with the least possible surface — so small drops (where tension dominates gravity and drag) are near-perfect spheres. The "teardrop" icon is fiction: falling raindrops are spheres when small, then flatten into hamburger-bun shapes as aerodynamic pressure competes (millimetre-plus sizes), and break apart beyond ~4 mm. Astronauts' water blobs aboard the ISS are the clean experiment: no gravity, no drag — flawless spheres.

How high can capillarity really lift water — do trees use it to reach 100 m?

Jurin's law with the finest xylem conduits (~10 μm radius) gives h=2(0.072)/(1000)(9.8)(10−5)≈1.5h = 2(0.072)/(1000)(9.8)(10^{-5}) \approx 1.5 m. Helpful, nowhere near 100 m. Tall trees mainly pull water from the top: evaporation in leaf pores creates enormous negative pressures (tension in the water column — cohesion keeps the column intact), a mechanism called the cohesion–tension theory. Capillarity's role is real but supporting: it wets the fine pathways and holds water in the smallest pores. So "trees = capillarity" oversimplifies; "surface forces, including capillarity and evaporation-driven tension" is the honest sentence.

Why does hot, soapy water clean so much better than cold plain water?

Both knobs lower surface tension. Cold plain water at T=0.072T = 0.072 N/m beads up on greasy fabric — high-tension surfaces refuse to spread into narrow pores (capillary entry needs wetting). Heat drops TT by several percent and, more importantly, soap drops it by half while its molecules coat grease particles (hydrophobic tails inward) so they detach and stay suspended. Result: water that spreads, penetrates pores, and carries oil away. The physics half of detergency is literally this article; the chemistry half is the surfactant molecule's two-faced structure.

Is the meniscus correction why burettes are read at the bottom of the curve?

Yes. In glass, water-based solutions wet the wall and curve up at the edges (concave meniscus); the physically meaningful level — where the bulk liquid's surface sits — is the meniscus's bottom, so titration protocol says read there, at eye level. Mercury (in old thermometers/barometers) curves the other way, and you read the top of its convex dome. The reading convention is applied contact-angle physics, and "why do we read the lower meniscus?" is a genuine lab-viva favourite.

Does surface tension exist between two liquids, or only liquid–air?

Every interface between phases carries an interfacial tension: oil–water ~0.05 N/m (why vinaigrette separates and shaken droplets re-merge), liquid–solid interfaces too (they set the contact angle via the balance of tensions along the contact line — Young's relation, stated qualitatively at HS). "Surface tension" is the special name for the liquid–vapour case. Emulsifiers like egg yolk in mayonnaise are surfactants for the oil–water interface — same physics as soap, aimed at a different boundary.

Before this

Where this leads