Inside the Mitochondrial Membrane: Marcus Inversion and Magnetism

Inside almost every cell you own sits a membrane folded so tightly it looks like the pleats of a paper fan. It is thinner than a soap bubble, yet it holds back a voltage that, measured across its few nanometers, rivals a lightning storm. Electrons hop across it in jumps no classical particle could make. And a growing group of physicists, chemists and clinicians say that something as faint as Earth's magnetic field can tilt the odds of those jumps.

This piece walks through that membrane in three layers: the textbook structure, the physics of electron jumps worked out by Rudolph Marcus, and the magnetic story told by spin chemists and by Dr. Jack Kruse. Each claim is labeled with the site's three tiers and attributed to the people who hold it.

Part 1: The Centralized View: a folded battery

Two membranes, one engine

A mitochondrion has two membranes. The outer one is smooth and leaky. The inner one, the inner mitochondrial membrane, is where the energy work happens. It folds inward into ridges called cristae, which multiply its surface area many times over. In 1961, Peter Mitchell proposed that the respiratory chain pumps protons across this inner membrane and that the stored gradient, not a chemical intermediate, drives the making of ATP. He called it the chemiosmotic mechanism, and it earned him the 1978 Nobel Prize in Chemistry (Mitchell 1961). Centralized View

The bucket brigade

Four large protein complexes sit in the membrane and pass electrons down an energy slope, like a bucket brigade passing water downhill:

• Complex I takes electrons from NADH and pumps protons.
• Complex II takes electrons from succinate and feeds them in without pumping.
• Complex III moves electrons from ubiquinone to cytochrome c and pumps protons.
• Complex IV, cytochrome c oxidase, hands the electrons to oxygen, which becomes water, and pumps more protons.

The pumped protons pile up between the membranes. That difference in charge and acidity is the proton motive force. Bioenergetics textbooks put the voltage part, the membrane potential, at roughly 150 to 180 millivolts, negative on the inside. The protons flow back home through ATP synthase, a rotary motor that turns their fall into ATP. Centralized View

The shape of the fold

The folds are not random. Using cryo-electron tomography, Werner Kühlbrandt's group in Frankfurt showed that ATP synthase sits in paired dimers that form long rows along the sharply curved edges of the cristae, and that these rows help bend the membrane into its shape (Strauss et al. 2008; Davies et al. 2011). Centralized View

The bends are also packed with an unusual fat. Cardiolipin is a phospholipid found almost only in the inner mitochondrial membrane. It has four fatty-acid tails instead of the usual two, which gives it a cone shape that favors curved membranes. Nikita Ikon and Michael Ryan describe cardiolipin as a key organizer of cristae, binding to the respiratory complexes and to ATP synthase and helping hold the complexes together in larger groups (Ikon & Ryan 2017). Centralized View Membranes in the brain, retina and heart are also rich in the omega-3 fat DHA, a long, highly unsaturated chain, and DHA-containing species of cardiolipin are found in those tissues. Centralized View

Part 2: The Centralized View: how an electron jumps

Tunneling through protein

Inside the complexes, electrons sit on iron-sulfur clusters, hemes, copper centers and quinones. These redox centers are often more than a nanometer apart, separated by protein. The electron does not travel the gap like a marble. It tunnels, a quantum effect in which a particle passes through a barrier it lacks the energy to climb. Harry Gray and Jay Winkler at Caltech showed that tunneling rates in proteins fall off steeply with distance (Gray & Winkler 2003). Centralized View

Christopher Moser, P. Leslie Dutton and colleagues at the University of Pennsylvania surveyed many natural electron-transfer proteins and found a simple rule. Distance is the main control on rate, and the effect of the specific protein matrix between the centers is secondary (Moser et al. 1992). In a follow-up, Christopher Page and colleagues reported that centers meant to pass electrons to each other almost always sit within about 14 Å, while centers that should not talk are kept farther apart. They describe this as a natural engineering principle: build the chain short enough to be fast and robust, and let the chemistry vary (Page et al. 1999). Centralized View

Marcus theory in one picture

Distance is half the story. The other half is energy, and that belongs to Rudolph Marcus, who won the 1992 Nobel Prize in Chemistry for it. His theory, first published in 1956, explains why some electron jumps are fast and others slow (Marcus 1956; Marcus & Sutin 1985). Centralized View

Here is a fireside version. Picture two bowls side by side. The first bowl is the molecule and the water around it before the electron jumps. The second is the same scene after the jump. The electron can only jump at the point where the two bowls cross, because only there do the before and after scenes have the same shape and energy. To get to that crossing, the surroundings have to shake, stretch and reorient into a strained position. The energy needed to bend everything from the before shape into the after shape is the reorganization energy, written λ (lambda). The height of the step down from the first bowl to the second is the driving force, written −ΔG.

Marcus showed that the height of the hill to the crossing point equals (λ + ΔG)² / 4λ. Three cases follow:

• Normal region. When the drop is small, making it bigger lowers the hill, and the jump gets faster. This is what chemists expected.
• The sweet spot. When the driving force equals the reorganization energy, the hill vanishes. The bowls cross right at the bottom of the first one. The jump is activationless and as fast as distance allows.
• Inverted region. Push the drop further and the second bowl slides so low that the crossing climbs back up the far wall of the first. The hill reappears, and the jump slows down even though it releases more energy.

The inverted region sounded so strange that chemists doubted it for decades. In 1984, John Miller, Lidia Calcaterra and Gerhard Closs measured electron jumps across rigid molecules over a range of driving forces and saw the rate rise, peak and then fall, as Marcus had predicted (Miller et al. 1984). Centralized View

Why the chain stays near the peak

In Moser and Dutton's analysis, most biological electron transfers sit near the top of the Marcus curve, where the driving force for each hop is comparable to its reorganization energy, so small changes in energy barely move the rate. In their account, nature tunes the chain mainly by spacing, and the steps that must not happen, such as electrons running backward, are slowed by distance or by placing them far out on the curve (Moser et al. 1992; Page et al. 1999). Centralized View

Part 3: Where the inverted region meets magnetism

Spin and the radical pair

Electrons carry spin, a built-in magnetism like a tiny bar magnet. When one electron jumps from one molecule to another, it can leave behind two molecules that each hold one unpaired electron: a radical pair. The two spins start out paired in a state called singlet, with opposite spins. Each electron also feels the small magnets of nearby atomic nuclei, an effect called hyperfine coupling, and that coupling slowly rocks the pair back and forth between singlet and triplet, the state with parallel spins. An outside magnetic field changes the rhythm of that rocking. The pair can only recombine, by an electron jumping back, from the singlet state, so the field changes how many pairs recombine and how many go on to form other products. Klaus Schulten, Albert Weller and colleagues measured this field dependence in radical ion pairs in 1976 (Schulten et al. 1976). Centralized View

The inverted region buys time

This is where Marcus meets magnetism. For a weak field like Earth's, roughly 25 to 65 microtesla, to make a difference, the radical pair has to live long enough for its spins to rock, typically a microsecond or so in the reviews by Peter Hore and Henrik Mouritsen. The jump back that would destroy the pair is often extremely downhill, which puts it deep in the Marcus inverted region, where it is slow. Hore and Mouritsen describe this as part of the design of a working magnetic sensor: fast forward electron hops create the pair, and a slow, inverted back-jump keeps it alive long enough to feel the field (Hore & Mouritsen 2016). Centralized View for the physics of radical pairs in chemistry

Cryptochrome, the model system

The best-studied example is cryptochrome, a blue-light protein in the eye. In 2000, Thorsten Ritz, Salih Adem and Klaus Schulten proposed that light-made radical pairs in cryptochrome let birds sense the direction of Earth's field (Ritz et al. 2000). In 2021, Jingjing Xu, Peter Hore, Henrik Mouritsen and colleagues reported that purified cryptochrome 4 from the European robin, a night-migrating songbird, is magnetically sensitive in the lab, with a chain of tryptophan amino acids carrying the electron through the protein (Xu et al. 2021). Centralized View that the protein is magnetically sensitive; Middle Ground on how birds use it in the wild

Part 4: The Middle Ground: magnets and mitochondria

Reactive oxygen species as radical pairs

When the respiratory chain leaks an electron directly onto oxygen, it makes superoxide, a radical. Superoxide becomes hydrogen peroxide. Carlos Martino and Robert Usselman have argued that the pairs formed during this chemistry follow the same spin rules as cryptochrome. In 2011, Martino and Pablo Castello reported that changing the static magnetic field around cells altered their hydrogen peroxide output (Martino & Castello 2011). In 2014, Usselman and colleagues reported that a weak radiofrequency magnetic field, tuned to flip electron spins, shifted the balance between superoxide and hydrogen peroxide in cultured cells (Usselman et al. 2014). In 2016, the team reported that oscillating magnetic fields shifted this partitioning and changed measures of cellular energy use, and called it the quantum biology of reactive oxygen species (Usselman et al. 2016). Middle Ground

Hadi Zadeh-Haghighi and Christoph Simon at the University of Calgary reviewed magnetic field effects across biology, from circadian clocks to stem cells and oxidative stress, and argued that the radical pair mechanism may explain many of them (Zadeh-Haghighi & Simon 2022). Middle Ground

Taking the field away

Another approach is to remove Earth's field. Lanxiang Tian's group at the Chinese Academy of Sciences housed mice in a hypomagnetic field, a space shielded to a small fraction of Earth's field. They reported raised reactive oxygen species and cognitive deficits in the hippocampus (Tian et al. 2022), and later fewer active mitochondria with longer exposure, along with lower expression of PGC-1α, the main switch for building new mitochondria (Tian et al. 2024). Earlier, Jing-Peng Fu and colleagues reported lower viability and mitochondrial activity in mouse muscle cells kept in a hypomagnetic field for three days (Fu et al. 2016). Zhang and Tian describe reactive oxygen species as the likely messengers of the hypomagnetic response (Zhang & Tian 2020). Middle Ground

A note on magnesium

Magnesium enters the magnetic story through its isotopes. Of its three stable forms, only magnesium-25 has a magnetic nucleus. Anatoly Buchachenko and colleagues reported that an ATP-making enzyme, phosphoglycerate kinase, made ATP faster with magnesium-25 than with the non-magnetic isotopes, and proposed a spin-dependent radical pair step in phosphorylation (Buchachenko et al. 2005). A team led by Darragh Crotty and Michael Wilson in the UK re-examined this with creatine kinase and reported no isotope or field effect (Crotty et al. 2012). Middle Ground

Part 5: The Decentralized View: Dr. Jack Kruse

Cardiolipin as the magnetic lipid

Dr. Jack Kruse, a neurosurgeon who now runs longevity clinics in El Salvador and the U.S., puts magnetism at the base of mitochondrial biology. In his framework, cardiolipin is the most important magnetic lipid inside the organelle. Its four tails and cone shape force the membrane into tight folds at the cristae tips, and that geometry anchors the respiratory complexes at the precise angstrom spacing that efficient electron tunneling requires. He describes the voltage across the inner membrane as about thirty million volts per meter. Decentralized View

In a recent post, Kruse writes that without "a conical cardiolipin structure," outside forces "would warp the inner mitochondrial membrane (IMM) metric" and "plunge the respiratory complexes headfirst into an interstellar Marcus Inversion bottleneck." That bottleneck, he says, "spreads energy out in the wrong way and leads to reversal of electron and proton flow" (Kruse, Oct 2026 thread). Decentralized View

The Marcus inversion as a breaker

Kruse uses Marcus's curve as a switch. In his telling, a healthy, grounded membrane in a stable magnetic field keeps its liquid-crystal water and high dielectric constant, and electrons flow near the top of the curve without wasting energy as heat or reactive oxygen. When the geometry changes, through excess linoleic acid replacing stable tails in cardiolipin, oxidative damage, artificial electromagnetic fields or loss of grounding, he says the cristae flatten, the water's dielectric constant drops, and electron transfer enters a Marcus inversion regime, where a stronger driving force slows the reaction. He calls the Marcus inverted region a "quantum circuit breaker" that the cell uses when mitochondria change shape (Kruse, LinkedIn). Decentralized View

Earth's field as a driver

Kruse writes that this design makes "the mitochondrial clocks in the IMM very sensitive to planetary magnetic field changes," and that "those changes might be the key driver of evolution of all diseases on Earth" (Kruse thread). In interviews he puts it as "magnetism is the father of the electric current," and argues that light is turned into an electric current in the body to run circadian pathways (Circadian Biology Is Physics). He says cardiolipin supports chiral spin selectivity, a spin-filtering effect described in chemistry by Ron Naaman and David Waldeck (Naaman & Waldeck 2012), allowing clean electron flow when the field is stable and the body is grounded. Decentralized View

DHA and light

Kruse calls DHA a lipid that has stayed unchanged across more than six hundred million years of evolution, with six double bonds whose shared electron cloud lets it convert light into a direct current and back again. In his view, DHA sets the voltage in outer cell membranes and the retina, and cardiolipin turns that electrical and magnetic information into the geometry and proton current needed for ATP. He credits Michael Crawford's work on DHA as a starting point. Decentralized View

What Kruse recommends

Kruse's prescription follows from this model: seafood as the DHA source, ideally raw or lightly cooked; natural light cycles with morning and midday sun; grounding with bare skin on the earth; and less exposure to artificial electromagnetic fields. He argues that this combination restores the high-efficiency electron and proton flow the brain and heart were built to use. Decentralized View

The mitochondrial membrane in practice

Here is what the featured voices point to, each within their own framework. None of it is advice from this site.

• Know the engine. Mitchell, Kühlbrandt, Ikon and Ryan describe a membrane built from protein complexes, folds and cardiolipin, powered by a proton gradient.
• Short hops, tuned energy. Moser, Dutton and Page describe a chain spaced within about 14 Å, sitting near the Marcus peak.
• Spin matters at the margins. Hore, Mouritsen, Usselman, Martino, Zadeh-Haghighi and Simon describe radical pairs as a way weak fields could nudge chemistry, including reactive oxygen species.
• Earth's field as baseline. Tian's group reports that removing it affects mouse mitochondria and ROS.
• Seafood, sun, ground. Kruse pairs DHA-rich seafood with sunlight, light cycles, grounding and cold to protect cardiolipin and keep electron flow out of the inverted region.

Evidence map

This article presents viewpoints, attributed to their sources. It is not medical advice.

Sources

Dr. Jack Kruse (viewpoint sources):

• Kruse, X thread on cardiolipin, the IMM and the Marcus Inversion bottleneck (Oct 2026)
• Kruse, LinkedIn: the Marcus Inverted Region as a quantum circuit breaker
• Dr. Jack Kruse: Circadian Biology Is Physics: Light, Water & Magnetism (YouTube)
• Tensegrity #5: Your Magnetic Sense

Research:

• Mitchell, P. (1961). Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism. Nature 191, 144–148. https://doi.org/10.1038/191144a0
• Strauss, M., Hofhaus, G., Schröder, R. R. & Kühlbrandt, W. (2008). Dimer ribbons of ATP synthase shape the inner mitochondrial membrane. EMBO Journal 27, 1154–1160. https://doi.org/10.1038/emboj.2008.35
• Davies, K. M. et al. (2011). Macromolecular organization of ATP synthase and complex I in whole mitochondria. PNAS 108, 14121–14126. https://doi.org/10.1073/pnas.1103621108
• Ikon, N. & Ryan, R. O. (2017). Cardiolipin and mitochondrial cristae organization. BBA Biomembranes 1859, 1156–1163. https://doi.org/10.1016/j.bbamem.2017.03.013
• Gray, H. B. & Winkler, J. R. (2003). Electron tunneling through proteins. Quarterly Reviews of Biophysics 36(3), 341–372. https://doi.org/10.1017/S0033583503003913
• Moser, C. C. et al. (1992). Nature of biological electron transfer. Nature 355, 796–802. https://doi.org/10.1038/355796a0
• Page, C. C. et al. (1999). Natural engineering principles of electron tunnelling in biological oxidation–reduction. Nature 402, 47–52. https://doi.org/10.1038/46972
• Marcus, R. A. (1956). On the theory of oxidation-reduction reactions involving electron transfer. I. Journal of Chemical Physics 24, 966–978. https://doi.org/10.1063/1.1742723
• Marcus, R. A. & Sutin, N. (1985). Electron transfers in chemistry and biology. BBA Reviews on Bioenergetics 811, 265–322. https://doi.org/10.1016/0304-4173(85)90014-X
• Miller, J. R., Calcaterra, L. T. & Closs, G. L. (1984). Intramolecular long-distance electron transfer in radical anions. JACS 106, 3047–3049. https://doi.org/10.1021/ja00322a058
• Schulten, K. et al. (1976). Magnetic field dependence of the geminate recombination of radical ion pairs in polar solvents. Zeitschrift für Physikalische Chemie 101, 371–390. https://doi.org/10.1524/zpch.1976.101.1-6.371
• Ritz, T., Adem, S. & Schulten, K. (2000). A model for photoreceptor-based magnetoreception in birds. Biophysical Journal 78(2), 707–718. https://doi.org/10.1016/S0006-3495(00)76629-X
• Hore, P. J. & Mouritsen, H. (2016). The radical-pair mechanism of magnetoreception. Annual Review of Biophysics 45, 299–344. https://doi.org/10.1146/annurev-biophys-032116-094545
• Xu, J. et al. (2021). Magnetic sensitivity of cryptochrome 4 from a migratory songbird. Nature 594, 535–540. https://doi.org/10.1038/s41586-021-03618-9
• Martino, C. F. & Castello, P. R. (2011). Modulation of hydrogen peroxide production in cellular systems by low level magnetic fields. PLoS ONE 6, e22753. https://doi.org/10.1371/journal.pone.0022753
• Usselman, R. J. et al. (2014). Spin biochemistry modulates reactive oxygen species (ROS) production by radio frequency magnetic fields. PLoS ONE 9, e93065. https://doi.org/10.1371/journal.pone.0093065
• Usselman, R. J. et al. (2016). The quantum biology of reactive oxygen species partitioning impacts cellular bioenergetics. Scientific Reports 6, 38543. https://doi.org/10.1038/srep38543
• Zadeh-Haghighi, H. & Simon, C. (2022). Magnetic field effects in biology from the perspective of the radical pair mechanism. Journal of the Royal Society Interface 19, 20220325. https://doi.org/10.1098/rsif.2022.0325
• Fu, J.-P. et al. (2016). Decline of cell viability and mitochondrial activity in mouse skeletal muscle cell in a hypomagnetic field. Bioelectromagnetics 37, 212–222. https://doi.org/10.1002/bem.21968
• Zhang, B. & Tian, L. (2020). Reactive oxygen species: potential regulatory molecules in response to hypomagnetic field exposure. Bioelectromagnetics 41, 573–580. https://doi.org/10.1002/bem.22299
• Tian, L. et al. (2022). Hypomagnetic field induces the production of reactive oxygen species and cognitive deficits in mice hippocampus. International Journal of Molecular Sciences 23, 3622. https://doi.org/10.3390/ijms23073622
• Tian, L., Ren, J. & Luo, Y. (2024). The effects of different durations of exposure to hypomagnetic field on the number of active mitochondria and ROS levels in the mouse hippocampus. Biochemistry and Biophysics Reports 38, 101696. https://doi.org/10.1016/j.bbrep.2024.101696
• Buchachenko, A. L. et al. (2005). Magnetic isotope effect of magnesium in phosphoglycerate kinase phosphorylation. PNAS 102, 10793–10796. https://doi.org/10.1073/pnas.0504876102
• Crotty, D. et al. (2012). Reexamination of magnetic isotope and field effects on adenosine triphosphate production by creatine kinase. PNAS 109, 1437–1442. https://doi.org/10.1073/pnas.1117840108
• Naaman, R. & Waldeck, D. H. (2012). Chiral-induced spin selectivity effect. Journal of Physical Chemistry Letters 3, 2178–2187. https://doi.org/10.1021/jz300793y

← All articles