How Molecules Cross Cell Membranes

Every cell is surrounded by a membrane only about 5 nm thick — thousands of times thinner than a sheet of paper. Yet this film controls, molecule by molecule, what gets in and out. Oxygen slips straight through; glucose needs a door; sodium ions need a channel or a pump that burns ATP. Understanding how molecules cross membranes brings together polarity, intermolecular forces, concentration gradients, electrochemistry and thermodynamics. This article works through the main routes, from simplest to most elaborate.

The barrier: a lipid bilayer

The membrane is built from phospholipids (see phospholipids and membranes). Each has a polar, charged head (phosphate plus a group such as choline) and two long, non-polar fatty acid tails. In water, they self-assemble into a bilayer: heads facing the water on both sides, tails tucked together in the middle.

That middle is a slab roughly 3 nm thick of hydrocarbon — chemically similar to a thin layer of oil. Cholesterol molecules sit among the tails, adjusting fluidity. Proteins are embedded throughout, making up about half of the membrane’s mass in many cells.

To cross the bilayer unaided, a molecule must leave the water, dissolve in the oily core, and re-enter water on the other side. How easily it does this depends on its size and above all its polarity.

Route 1: simple diffusion through the lipid

Small, non-polar molecules cross easily: O₂, CO₂, N₂, and lipid-soluble substances such as steroid hormones (see cortisol) and many drugs. Small uncharged polar molecules such as water, ethanol and urea cross more slowly. Ions — Na⁺, K⁺, Cl⁻, Ca²⁺ — essentially can’t cross at all. The reason is energetic: an ion in water is surrounded by a shell of water molecules attracted to its charge. Stripping that shell away and moving the ion into a low-polarity hydrocarbon costs a very large amount of energy. Large polar molecules such as glucose are also effectively blocked.

A useful rule of thumb: permeability correlates with how well a substance dissolves in oil compared with water (its partition coefficient). This is why drug designers pay close attention to log P (see how drugs are designed).

Fick’s law

The rate of diffusion across a membrane is described by Fick’s first law:

rate of diffusion ∝ (surface area × concentration difference) ÷ thickness

More precisely, flux J = P × (C₁ − C₂), where P is the permeability coefficient (which includes the partition coefficient and the diffusion coefficient in the membrane, divided by membrane thickness). This explains why organs specialised for exchange — alveoli in lungs, villi in the gut, capillaries — have huge surface areas, very thin barriers and steep concentration gradients maintained by blood flow.

Diffusion is passive: molecules move down their concentration gradient, and no ATP is used. The driving force is the increase in entropy as concentrations even out.

Route 2: facilitated diffusion

Polar molecules and ions cross through membrane proteins, still moving down their gradient and still without ATP. There are two kinds of protein.

Channels

Channel proteins form water-filled pores. They’re usually highly selective. The potassium channel is a famous example: its narrow selectivity filter is lined with carbonyl oxygen atoms positioned to replace exactly the water molecules around a K⁺ ion. K⁺ passes through with almost no energy cost. Na⁺ is smaller, so it can’t touch all the carbonyl oxygens at once, and it stays with its water shell, which is too big to fit. The result is selectivity of about 1,000 to 1 for K⁺ over Na⁺, at rates of up to around 10⁸ ions per second per channel. Roderick MacKinnon shared the 2003 Nobel Prize in Chemistry for revealing this structure.

Aquaporins are channels for water, letting it cross far faster than through the bilayer alone (Peter Agre shared the same 2003 prize for discovering them). Their narrow pores and a positively charged region block protons, so water can pass but H⁺ can’t hop through. Many channels are gated — opened by voltage, by binding a signal molecule, or by stretching — which is the basis of nerve impulses (see nerve impulses).

Carriers

Carrier proteins bind their passenger on one side, change shape, and release it on the other. The GLUT family of glucose transporters works this way. Carriers are slower than channels (hundreds to thousands of molecules per second) and they saturate: at high concentrations every carrier is busy, so the rate levels off at a maximum — kinetics that look just like an enzyme’s Michaelis–Menten curve (see enzyme kinetics). Simple diffusion, by contrast, keeps increasing linearly with concentration.

Route 3: active transport

To move something against its gradient, a cell must supply energy.

Primary active transport: pumps

Pumps use ATP directly. The sodium–potassium pump (Na⁺/K⁺-ATPase) is the best-known:

1. Three Na⁺ bind on the inside.
2. ATP transfers a phosphate to the pump, changing its shape; the Na⁺ sites open to the outside and release the ions.
3. Two K⁺ bind from the outside.
4. The phosphate is released, the pump returns to its original shape and releases the K⁺ inside.

Each cycle moves 3 Na⁺ out and 2 K⁺ in per ATP hydrolysed (see ATP vs ADP). Because it moves more positive charge out than in, it also contributes directly to the negative voltage inside the cell. Jens Skou discovered this pump and shared the 1997 Nobel Prize in Chemistry.

Other pumps include the calcium pump that clears Ca²⁺ after muscle contraction, and the proton pump in the stomach lining that makes acid, concentrating H⁺ about a million-fold relative to blood.

The energy of a gradient

How much energy does it cost to move a substance against a gradient? For an uncharged molecule:

ΔG = RT ln(C_in / C_out)

For an ion, an electrical term is added: ΔG = RT ln(C_in / C_out) + zFΔψ, where z is the ion’s charge, F is the Faraday constant and Δψ the membrane potential.

Worked example. At 310 K, moving glucose from 5 mmol dm⁻³ outside to 50 mmol dm⁻³ inside:

ΔG = 8.314 × 310 × ln(50/5) = 8.314 × 310 × 2.303 ≈ +5.9 kJ mol⁻¹

A tenfold gradient costs about 5.9 kJ mol⁻¹ — well within the energy available from ATP hydrolysis (around −50 kJ mol⁻¹ under cell conditions).

Secondary active transport: co-transport

Cells often use the sodium gradient built by the Na⁺/K⁺ pump to drive other transport. Co-transporters couple the downhill flow of Na⁺ into the cell with the uphill movement of another molecule:

• Symport: both move in the same direction. The SGLT1 transporter in the gut brings in 2 Na⁺ with each glucose, letting cells absorb glucose even when it’s more concentrated inside. This is the chemistry behind oral rehydration salts (see electrolytes in the body).
• Antiport: they move in opposite directions. The sodium–calcium exchanger in heart cells uses Na⁺ entry to push Ca²⁺ out.

No ATP is used by the co-transporter itself, but the gradient it depends on was paid for with ATP by the pump — hence “secondary”.

Osmosis: water on the move

Osmosis is the net movement of water across a partially permeable membrane from a region of lower solute concentration to a region of higher solute concentration (higher water potential to lower). It happens through the bilayer and, much faster, through aquaporins.

What matters is the number of dissolved particles, not their identity. The osmotic pressure of a dilute solution is approximately:

Π = iMRT

where i is the number of particles per formula unit (2 for NaCl), M the molarity and T the temperature. Blood plasma is about 290 mOsm per kg of water; a “normal saline” drip (0.9 % NaCl, about 0.154 mol dm⁻³, giving about 308 mOsm dm⁻³ if fully dissociated) is close to this, so red blood cells neither swell nor shrink in it. In pure water, red cells swell and burst; in concentrated salt solution, they shrink. Plant cells don’t burst, because their cellulose walls resist expansion — creating the turgor pressure that keeps plants upright (see water in the human body).

Route 4: bulk transport in vesicles

Very large molecules and particles cross in membrane bubbles:

• Endocytosis: the membrane folds inward and pinches off, bringing material in. Phagocytosis (“cell eating”) by white blood cells engulfs bacteria (see the immune system as chemistry). Receptor-mediated endocytosis takes up specific molecules — for example, LDL particles carrying cholesterol.
• Exocytosis: vesicles fuse with the membrane and release their contents outside — how nerve cells release neurotransmitters and how cells secrete hormones such as insulin.

These processes need energy, both to reshape the membrane and to move vesicles along the cell’s internal skeleton.

Summary table

Route	Energy	Direction	Protein?	Examples
Simple diffusion	None	Down gradient	No	O₂, CO₂, steroids
Facilitated diffusion (channel)	None	Down gradient	Channel	K⁺, water (aquaporins)
Facilitated diffusion (carrier)	None	Down gradient	Carrier; saturates	Glucose via GLUT
Primary active transport	ATP	Against gradient	Pump	Na⁺/K⁺-ATPase, H⁺ pump
Secondary active transport	Ion gradient	Against gradient (for the passenger)	Co-transporter	SGLT1 (Na⁺ + glucose)
Osmosis	None	Water to higher solute concentration	Bilayer or aquaporins	Water
Endocytosis / exocytosis	ATP	In / out	Many	Bacteria, hormones, neurotransmitters

Key takeaways

• The bilayer’s hydrocarbon core lets small non-polar molecules through but blocks ions and large polar molecules.
• Fick’s law: diffusion rate depends on area, gradient and thickness.
• Channels and carriers allow facilitated diffusion; carriers saturate.
• Pumps use ATP to move substances against gradients; co-transporters use the Na⁺ gradient.
• The cost of moving a molecule against a gradient is ΔG = RT ln(C_in/C_out) (plus an electrical term for ions).
• Osmosis depends on the number of dissolved particles; vesicles move large cargo.