Solid State Fusion & Nuclear Physics

Why should fusion announce itself with neutrons?

When two deuterons fuse, the accounting is well known. A deuteron is a heavy hydrogen nucleus, one proton plus one neutron, and a deuterium-loaded experiment is swimming in them. Push two close enough and they react along one of a few set paths, called branches. About half the time you get tritium (hydrogen with two neutrons) plus a proton. About half the time you get helium-3 plus a free neutron. Only about once in a million reactions do the two deuterons fuse straight into helium-4 and dump their leftover energy, roughly 24 million electron-volts, as a single gamma ray.¹

The neutron branch is the giveaway. If something were running deuteron–deuteron fusion fast enough to noticeably warm a cup of water, the helium-3-plus-neutron path alone would spray the lab with neutrons. Neutrons are penetrating and easy to count, so a real fusion furnace cannot hide. It carries its own alarm.

Here is the snag. A subset of solid-state fusion (SSF) experiments report the opposite combination: measurable heat, helium-4 building up roughly in step with that heat, tritium above the natural background, and a neutron count far too low to pay for it all.² Two honest options follow. Either the measurements are flawed, or something is going on that the clean two-body picture from the textbook does not capture. You cannot settle which by deciding in advance who sounds more credible.

What is the actual question worth asking?

“Has cold fusion been proven?” is the wrong question, and the answer to it is no. A sharper question survives: what happens to nuclear reaction rates inside a metal packed with hydrogen, when the deuterium density climbs toward one hydrogen atom for every metal atom?

Most nuclear physics describes isolated nuclei, two particles meeting in a vacuum or a thin hot gas. There, the Coulomb barrier rules. Both nuclei are positively charged, they repel, and at low energy the only way through is quantum tunneling, whose probability falls off ferociously as the energy drops. That framework is among the best-tested in all of science, and nothing here challenges it on its own turf.

SSF asks a different question. Not about two nuclei alone, but about reactions happening inside a crowded place: deuterium dissolved in palladium or nickel, surrounded by a sea of mobile electrons and a lattice that can pass energy around in collective vibrations. The strongest skeptical argument is not hand-waving about vacuum. It is a careful upper limit on how fast such reactions could possibly go in that environment, and it lands far below what the heat measurements would need.¹⁰ A serious advocate has to meet that limit head-on. We will get to how.

We do know the barrier is not fixed in stone. Surrounding electrons can screen some of the repulsion between nuclei, and beam experiments firing deuterons into metals have measured screening effects of a few hundred electron-volts, several times larger than the simplest theory predicts and still unexplained.³ That cuts both ways. It shows the barrier genuinely responds to its electronic surroundings, and it caps how far that response goes. A few hundred eV does not come close to closing the gap of many orders of magnitude between vacuum reaction rates and the rates the heat claims imply.

How do you tell a nuclear event from a chemical one?

The single observation doing the most work in the field is the heat-helium correlation. In a set of palladium and heavy-water cells, cathodes that put out more excess power also accumulated more helium-4, and the ratio of energy released to helium produced sat near the value you would expect if the heat came from deuterons fusing to helium-4, around 24 MeV per atom.⁴ contested That number is exactly the kind of thing nuclear physics exists to check, because it predicts a specific energy-per-helium ratio and supplies the detection chain — accelerator mass spectrometry — able to measure helium-4 against the atmospheric helium that leaks into any apparatus.⁵ A hard background is a reason to build better instruments, not a reason to look away.

The rest of the toolkit follows the same logic. Tritium, when it shows up above background and tracks the operating conditions, is a nuclear fingerprint no chemical reaction can fake, with a clean 12.3-year radioactive decay you can detect.⁶ Neutron spectroscopy, which clocks neutrons by their flight time and energy, is the right tool to measure or bound the very neutron shortfall at the center of the puzzle.

And there is a real precedent for changing fusion rates by changing the electronic environment: muon-catalyzed fusion. Swap the electron in a hydrogen molecule for a muon, a particle about 200 times heavier, and the two nuclei sit roughly 200 times closer, so fusion proceeds at room temperature.⁷ Worth being careful about what that proves. It shows catalysis of fusion is possible in principle. It does not show a lattice can do the same trick, because the muon works through its mass, and electrons and lattice vibrations have no comparable lever.

Hasn’t someone just done the calculation?

The bound is rigorous, model-independent — and assumes thermodynamic equilibrium. A driven cathode is not in equilibrium.

Yes, and this is the part most worth slowing down for. In 1989 the physicists Anthony Leggett and Gordon Baym derived a rigorous upper bound on the deuteron–deuteron fusion rate in a metal. It is model-independent, written entirely in terms of measurable thermodynamic quantities, with no vacuum barrier assumed anywhere. Their bound sits orders of magnitude below the rate the excess-heat claims would require.¹⁰ So the complaint that “nobody ever did the calculation” is simply wrong. It was done, and it is good.

Notice the one assumption it rests on. The bound describes a system in thermodynamic equilibrium. A working electrolytic cathode is not in equilibrium. It is driven hard, with steep loading gradients across its surface, transient stress, cracking, and local chemical potentials that swing far from their bulk values. Whether an equilibrium bound still constrains such a driven system is, as far as the published record shows, an open question. No one has extended the bound to the non-equilibrium case, and no one has knocked it down on its own terms. That is the live disagreement, and it is far more precise than the usual shouting match.

The way to settle it is empirical: measure the products with instruments good enough to be believed, in experiments clean enough to repeat. A 2025 study shows what that standard looks like, with a sharp caveat about what it did and did not test. Deuterons were accelerated to 30 keV and fired into a palladium target — ordinary hot fusion — and electrochemically loading the target raised the fusion rate by 15 ± 2 percent.⁸ That is a clean demonstration of the screening effect. Right shape of result for the screening question, no evidence at all on the central anomaly. The lattice claims still need a measurement of their own to that standard.

Why would a nuclear physicist take this on?

Look honestly at the base rate first. When Google convened a multi-institution team in the late 2010s to re-examine these claims with modern methods, the headline result was negative. They did not reproduce excess heat, and they said so in Nature.⁹ The same paper did something more useful than render a verdict, though. It pointed to specific corners of the parameter space, especially the materials science of reaching and holding extreme hydrogen loading, where the older work was too crude to conclude anything either way. That is the real state of play: not a proven phenomenon, not a closed case, a poorly mapped region with a few stubborn, unexplained signals in it.

The traffic on this bridge runs both ways. Suppose the heat-helium-without-neutrons pattern holds up under careful scrutiny. Then the suppression of the usual branching becomes a genuine and deep nuclear physics problem, because branching ratios are supposed to be set by nuclear structure, something the surrounding electrons have no business touching. A confirmed environmental influence on branching would be a major result. The flip side is that such a large claim starts with a low prior and therefore demands a correspondingly high standard of proof. The size of the prize fixes the size of the bar.

So here is the thread for a sharp student to pull. Could you design the measurement that decides it — neutron spectroscopy, mass spectrometry for helium, calorimetry good enough to trust, on a cell loaded and driven the way the equilibrium bound is not built to describe? Either you confirm an anomaly that demands new theory, or you close the question on terms everyone can accept. The missing neutrons will not explain themselves.

D+D Fusion Branching Reference

• D+D → T + p (Q ≈ 4.03 MeV, ~50%)
• D+D → ³He + n (Q ≈ 3.27 MeV, ~50%)
• D+D → ⁴He + γ (Q = 23.8 MeV, radiative branch ~10⁻⁶–10⁻⁷)

Editorial note: This primer presents an accessible introduction to SSF’s relationship to nuclear physics. The underlying nuclear claims of SSF/LENR remain scientifically contested. Evidence claims are tiered as established, contested, or reported-but-unconfirmed as noted inline. Readers are directed to primary experimental literature for empirical evaluation.

Notes

1. The three D+D channels and their approximate branching: D+D → T + p (Q ≈ 4.03 MeV, ~50%); D+D → ³He + n (Q ≈ 3.27 MeV, ~50%); D+D → ⁴He + γ (Q = 23.8 MeV, radiative branch ~10⁻⁶–10⁻⁷). Standard evaluated nuclear data; see D. A. Brown et al., “ENDF/B-VIII.0: The 8th Major Release of the Nuclear Reaction Data Library,” Nuclear Data Sheets 148 (2018): 1–142. ↩
2. The neutron deficit relative to a deuteron–deuteron heat source is the defining anomaly of the field; overview in Edmund Storms, The Science of Low Energy Nuclear Reaction (Singapore: World Scientific, 2007). Tier C (single-community synthesis, replication limited); treated here as “reported.” ↩
3. Anomalously large electron-screening potentials for d(d,p)t in metallic hosts (order hundreds of eV, well above the adiabatic limit) are reported in F. Raiola et al., “Enhanced electron screening in d(d,p)t for deuterated metals,” European Physical Journal A 19 (2004): 283–287, and A. Huke et al., “Enhancement of deuteron-fusion reactions in metals,” Physical Review C 78 (2008): 015803. The effect is established and unexplained; its magnitude does not span the gap to calorimetric-rate fusion, and it bounds the screening argument as much as it supports it. ↩
4. M. H. Miles, B. F. Bush, et al., “Correlation of Excess Power and Helium Production during D₂O and H₂O Electrolysis Using Palladium Cathodes,” Journal of Electroanalytical Chemistry 346 (1993): 99–117. The energy-per-helium ratio is reported as roughly consistent (within about an order of magnitude) with 23.8 MeV per ⁴He. Tier C, contested: single-group result, criticized on calorimetry and contamination grounds; see note 5. ↩
5. U.S. Department of Energy, Report of the Review of Low Energy Nuclear Reactions (Washington, DC, December 2004). Reviewers split roughly evenly on the excess-heat evidence; opinion on the ⁴He evidence was divided, with several noting helium levels close enough to background to admit atmospheric contamination. ↩
6. Reports of weak, sporadic neutron emission and of tritium above background are summarized in Storms, The Science of Low Energy Nuclear Reaction; both classes of claim are preliminary and unevenly replicated, and are presented here as reported, not established. ↩
7. L. W. Alvarez et al., “Catalysis of Nuclear Reactions by μ Mesons,” Physical Review 105 (1957): 1127–1128; J. D. Jackson, “Catalysis of Nuclear Reactions between Hydrogen Isotopes by μ⁻-Mesons,” Physical Review 106 (1957): 330–339. ↩
8. Chen et al., “Electrochemical Loading Enhances Deuterium Fusion Rates in a Metal Target,” Nature 644 (2025): 640–645, doi:10.1038/s41586-025-09042-7. A 30-keV plasma-ion-implantation (beam-target) experiment: ordinary hot, neutron-producing D–D fusion, in which electrochemical loading produced a 15 ± 2 percent enhancement. Established but narrow: the measurement is not in dispute; it does not observe neutron-free heat, helium-4 accumulation, or altered branching, and has no bearing on the central pattern. ↩
9. C. P. Berlinguette et al., “Revisiting the Cold Case of Cold Fusion,” Nature 570 (2019): 45–51. The Google-convened effort did not reproduce excess heat but identified specific lines of inquiry, notably extreme-loading materials science, meriting further study. ↩
10. A. J. Leggett and G. Baym, “Exact Upper Bound on Barrier Penetration Probabilities in Many-Body Systems: Application to ‘Cold Fusion’,” Physical Review Letters 63 (1989): 191–194; and “Can Solid-State Effects Enhance the Cold-Fusion Rate?” Nature 340 (1989): 45–46. The bound is rigorous, model-independent, and assumes thermodynamic equilibrium; whether it constrains driven, far-from-equilibrium cathodes is, on the published record, unresolved — neither extended to the non-equilibrium case nor rebutted on its own terms. ↩

Introduction: Stripping the Puzzle to Its Core

Start with the cleanest version of the puzzle, the one stripped of three decades of argument. If two deuterons fuse, the books are unambiguous about where the energy goes. Roughly half the time you get helium-3 and a neutron; roughly half the time tritium and a proton; and only about once in a million events do you get helium-4, which then sheds its 23.8 MeV of excess binding as a single gamma ray.¹ The neutron branch is the floodlight here. Any process running deuteron–deuteron fusion at a rate high enough to warm a calorimeter should bathe the room in neutrons.

Yet a subset of solid-state fusion (SSF) experiments report the opposite arrangement: measurable heat, helium-4 accumulating in rough proportion to that heat, tritium above background, and a neutron flux orders of magnitude too small to balance the ledger.² Either the measurements are wrong, or something is happening that the vacuum two-body picture does not describe. Neither possibility can be settled by argument from authority.

For a nuclear physicist, the real question is not whether cold fusion has been proven. It has not. The question is what actually happens to nuclear reaction rates and branching ratios inside a metal lattice loaded with hydrogen to fractions approaching one-to-one. That question has not gone untouched. A rigorous, lattice-inclusive bound on the equilibrium tunneling rate does exist, and it falls far below what the heat claims would require.¹² The live question is narrower and more interesting than the usual standoff suggests. It turns on whether that equilibrium bound governs a driven, far-from-equilibrium cathode at all.

Section I — Open Questions

Conventional nuclear physics is, overwhelmingly, the physics of isolated nuclei: two bodies in vacuum, a dilute plasma at stellar temperature, or a beam on a target. The Coulomb barrier dominates everything at low energy, and the Gamow factor exponentially suppresses tunneling as the relative energy drops. This framework is among the best-tested in all of physics, and nothing in what follows disputes it on its home ground.

SSF makes a different kind of claim, not about isolated nuclei but about reaction rates in a specific, crowded, many-body environment: deuterium dissolved in palladium or nickel at extreme loading, surrounded by a sea of conduction electrons and a lattice that can carry energy in collective modes. Caricaturing the skeptical case as vacuum hand-waving would be easy and wrong. The strongest version has nothing to do with vacuum. It is a rigorous, model-independent bound on the equilibrium many-body tunneling rate, written in terms of the measurable thermodynamic affinities of deuterium and helium in the metal, that lands orders of magnitude below the rate the heat claims demand.¹² That bound is the genuine obstacle, and an honest advocate has to meet it head-on rather than pretend it does not exist.

We already know the barrier is not immutable. Electron screening measurably enhances low-energy fusion cross sections, and accelerator measurements of the d(d,p)t reaction in metallic hosts have repeatedly returned screening potentials of order hundreds of electron-volts, several times larger than the adiabatic-limit prediction and still unexplained.³ That anomaly is real and instructive in both directions. It shows the barrier genuinely responds to the electronic environment, and it bounds how far that response can be pushed, because an effect of tens to hundreds of eV does not begin to span the many orders of magnitude separating vacuum rates from calorimetric ones. The screened-barrier calculation is the standard tool here, and the natural first bridge from the home field to SSF.

Section II — What Can Nuclear Physics Actually Measure?

The most valuable thing nuclear physics offers SSF is a set of discriminating instruments: ways to tell a nuclear event from a chemical one, and a real signal from an artifact. Consider the heat-helium correlation, the observation that does the most work in the field. In a series of palladium and heavy-water cells, cathodes that produced more excess power produced more helium-4, and the ratio of energy to helium clustered near the value expected if the heat came from deuterons fusing to helium-4, on the order of 24 MeV per atom.⁴ That number is exactly the kind of thing nuclear physics exists to adjudicate. It sets the expected energy-per-helium ratio, and it supplies the detection chain — accelerator mass spectrometry, high-resolution mass spectrometry — capable of measuring helium-4 against the ever-present background of atmospheric helium leaking into any apparatus. The contamination worry is real, and it was a principal reason a 2004 Department of Energy review panel split on the helium evidence.⁵ But "the background is hard" is an argument for better instruments, not for looking away. Those instruments exist.

The same logic runs through the rest of the toolkit. Evaluated cross-section libraries such as ENDF/B give the quantitative baseline against which any claimed rate or product ratio must be compared.⁶ Tritium, when it appears above background and tracks operating conditions, is a nuclear tracer no chemical process can counterfeit. Nuclear physics defines its production channels, its 12.3-year decay signature, and the thresholds an experiment must clear to claim it. (The strength of specific tritium reports is contested; they are presented here as reported, not established. See note 7.) Neutron spectroscopy, using time-of-flight, energy-resolved detection, is the method needed to characterize or bound the very neutron deficit that defines the central anomaly.⁷ Nuclear activation analysis is the established, quantitative way to test reported transmutation products, the appearance of elements absent from the starting materials. That class of claim remains preliminary and poorly replicated, and it should be treated as a hypothesis to be tested rather than a result to be defended.

There is also a genuine mechanistic precedent worth stating carefully, because it is often overstated. Muon-catalyzed fusion is real, established nuclear physics: replace the electron in a hydrogen molecule with a muon, the internuclear spacing collapses by the muon-to-electron mass ratio, and fusion proceeds readily at room temperature.⁸ It proves that catalytic enhancement of fusion rates by altering the electronic environment is not a fantasy. It happens, and it was worked out theoretically in the 1950s. What it does not prove is that a lattice can do anything similar: the muon works because of its mass, and electrons and phonons have no such lever. The precedent establishes possibility in principle, not mechanism in fact, and that kind of precision strengthens the case rather than weakening it.

Finally, the formal machinery for sub-barrier reactions — the optical model, R-matrix theory, coupled-channels methods — was built for vacuum, but adapting it to a reaction embedded in a many-body host is a well-posed theoretical program. The tools developed for screening in dense stellar plasmas, in white-dwarf cores and neutron-star crusts, are the closest existing analogue, and porting them to the condensed-matter regime is work nuclear theorists are well equipped to do.

Section III — Back to Nuclear Physics

The traffic across this bridge runs both ways, and the return cargo is the part nuclear physicists tend to underweight. If the heat-helium-without-neutrons pattern survives rigorous scrutiny (a real if, presented as such), then the suppression of the standard deuteron–deuteron branching ratio in a dense medium becomes a new nuclear physics problem, and a deep one, rather than an embarrassment to be explained away. Branching ratios are supposed to be set by nuclear structure and the available phase space, quantities we do not expect the surrounding electrons to touch. A confirmed environmental influence on branching would rank among the more consequential results a low-energy nuclear program could produce. Precisely because the claim is that large, the prior against it is low and the evidentiary bar correspondingly high. The size of the prize is the measure of the proof it would demand.

The other returns are more modest and more certain. A loaded metal hydride is, in effect, a many-body tunneling laboratory at energies below where accelerator cross sections can be measured directly, the regime that matters most for stellar nucleosynthesis and is hardest to reach in the lab. Even null results there constrain screening models that astrophysics cares about. The controversy has already done concrete institutional work: in 2023 the U.S. Advanced Research Projects Agency–Energy committed roughly $10 million across eight teams, including groups at MIT, Stanford, and Lawrence Berkeley, to settle the low-energy-nuclear-reaction question with modern instrumentation, explicitly to either find the effect or close the book on it.⁹ Whatever those programs find, the detectors, the calorimetry standards, and the trained people transfer directly to the rest of nuclear measurement science.

Section IV — Has the Calculation Really Never Been Done?

The strongest objection is not that vacuum rates are negligible. It is a rigorous upper bound on the equilibrium many-body tunneling rate.

The honest opening for an advocate is to grant the bound and then ask the one question it leaves open: does it govern a driven, far-from-equilibrium cathode? Here is the argument worth making, because it is the one that survives contact with the literature. In 1989, Leggett and Baym derived a model-independent upper limit on the rate of deuteron–deuteron fusion in a metal lattice, assuming thermodynamic equilibrium and expressed entirely in terms of measurable thermodynamic quantities, with no vacuum barrier anywhere in it.¹² Their bound sits orders of magnitude below the rate the excess-heat claims would require. That is a serious, condensed-matter-aware result, and it is why the flat "nobody did the calculation" complaint is wrong. The calculation was done.

But notice exactly what it assumes. The bound is an equilibrium statement. A working electrolytic cathode is not in equilibrium: it is driven hard, with steep loading gradients across its surface, transient stress, fracture, and chemical potentials that swing far from their bulk equilibrium values. Whether the equilibrium bound still constrains such a system is, as far as the published record shows, genuinely open. No one has extended the bound to driven, non-equilibrium loading, and no one has rebutted it on its own terms. That is the live question, and it is a far more precise and defensible one than the circularity charge it replaces. The advocate's claim is not "the critics are reasoning in a circle." It is that the one rigorous bound we have rests on an assumption the experiments deliberately violate, and extending it is unfinished theoretical work.

The way to settle which side is right is, in the end, empirical: measure the products, with instruments good enough to be believed, in experiments controlled well enough to be repeated. That this is now tractable is the genuinely new element. High-purity germanium gamma spectroscopy, time-of-flight neutron detection, and accelerator mass spectrometry for helium are mature enough to make a definitive measurement of nuclear products against chemical and atmospheric backgrounds.

What that standard looks like in practice is worth a concrete example, with a sharp caveat. A 2025 study did not test the cold-fusion claim at all. It was a conventional hot experiment in which deuterons were accelerated to 30 keV and implanted into a palladium target, and the ordinary, neutron-producing beam-target fusion rate was measured by a pulse-shape-discriminating scintillator. Its clean, mainstream finding was that electrochemically loading the target raised that hot fusion rate by 15 ± 2 percent, a direct and modest demonstration of the very screening effect discussed in Section I.¹⁰ The result itself is not in dispute. What it does not show matters just as much: no neutron-free heat, no helium-4 accumulation, no altered branching. It produced neutrons, as hot D–D fusion must. It is the right shape of result for the screening question, and no evidence at all on the central anomaly. The lattice-anomaly claims still need a measurement of their own to that standard.

Section V — Why Take This On?

The base rate deserves a clear look. When Google convened a multi-institution team in the late 2010s to re-examine these claims with modern methods, the headline finding was negative: they did not reproduce excess heat, and they said so in Nature.¹¹ But the same paper did something more interesting than render a verdict. It identified specific, underexplored corners of the parameter space, particularly the materials science of achieving and sustaining extreme hydrogen loading, where the older work was simply not good enough to conclude anything. That is the honest state of play: not a proven phenomenon, not a closed case, but a poorly mapped region with a few persistent, unexplained signals in it.

The claimed phenomena — anomalous heat, helium-4 accumulation, tritium, altered branching — are nuclear physics claims, and they require nuclear physics methods to evaluate. Nuclear physics cannot referee a question while declining to run the experiment. Bring the detectors, the cross-section libraries, and the theoretical machinery to bear on well-characterized systems, and one of two things will happen: either an anomaly is confirmed that demands new theory, or the question closes on terms everyone can accept. Both outcomes advance the field, and neither leaves nuclear physics worse off. The missing neutrons will not explain themselves.

D–D Fusion Branching Reference

• D+D → T + p (Q ≈ 4.03 MeV, ~50%)
• D+D → ³He + n (Q ≈ 3.27 MeV, ~50%)
• D+D → ⁴He + γ (Q = 23.8 MeV, radiative branch ~10⁻⁶–10⁻⁷)

Editorial note: This article presents a scholarly synthesis of SSF's relationship to nuclear physics. The underlying nuclear claims of SSF/LENR remain scientifically contested. Evidence claims are tiered as established, contested, or reported-but-unconfirmed as noted inline. Readers are directed to primary experimental literature for empirical evaluation.

References & Footnotes

1. D+D branching: D+D → T + p (Q ≈ 4.03 MeV, ~50%); D+D → ³He + n (Q ≈ 3.27 MeV, ~50%); D+D → ⁴He + γ (Q = 23.8 MeV, radiative branch ~10⁻⁶–10⁻⁷). Standard evaluated nuclear data; see D. A. Brown et al., "ENDF/B-VIII.0: The 8th Major Release of the Nuclear Reaction Data Library," Nuclear Data Sheets 148 (2018): 1–142. ↩
2. The neutron deficit relative to a deuteron–deuteron heat source is the defining anomaly of the field; overview in Edmund Storms, The Science of Low Energy Nuclear Reaction (Singapore: World Scientific, 2007). Tier C (single-community synthesis, replication limited); treated here as "reported." ↩
3. Anomalously large electron-screening potentials for d(d,p)t in metallic hosts (order hundreds of eV, well above the adiabatic limit) are reported in F. Raiola et al., "Enhanced electron screening in d(d,p)t for deuterated metals," European Physical Journal A 19 (2004): 283–287, and A. Huke et al., "Enhancement of deuteron-fusion reactions in metals," Physical Review C 78 (2008): 015803. The effect is established and unexplained; its magnitude (tens to hundreds of eV) does not span the gap to calorimetric-rate fusion, and it bounds the screening argument as much as it supports it. ↩
4. M. H. Miles, B. F. Bush, et al., "Correlation of Excess Power and Helium Production during D₂O and H₂O Electrolysis Using Palladium Cathodes," Journal of Electroanalytical Chemistry 346 (1993): 99–117. The energy-per-helium ratio is reported as roughly consistent (within about an order of magnitude) with 23.8 MeV per ⁴He. Tier C, contested: single-group result, criticized on calorimetry and contamination grounds; see note 5. ↩
5. U.S. Department of Energy, Report of the Review of Low Energy Nuclear Reactions (Washington, DC, December 2004). Reviewers split roughly evenly on the excess-heat evidence; opinion on the ⁴He evidence was divided, with several noting helium levels close enough to background to admit atmospheric contamination. ↩
6. Brown et al., "ENDF/B-VIII.0," Nuclear Data Sheets 148 (2018): 1–142. ↩
7. Reports of weak, sporadic neutron emission and of tritium above background are summarized in Storms, The Science of Low Energy Nuclear Reaction; both classes of claim are preliminary and unevenly replicated, and are presented here as reported, not established. ↩
8. L. W. Alvarez et al., "Catalysis of Nuclear Reactions by μ Mesons," Physical Review 105 (1957): 1127–1128; J. D. Jackson, "Catalysis of Nuclear Reactions between Hydrogen Isotopes by μ⁻-Mesons," Physical Review 106 (1957): 330–339. ↩
9. Advanced Research Projects Agency–Energy (ARPA-E), "U.S. Department of Energy Announces $10 Million in Funding to Projects Studying Low-Energy Nuclear Reactions," February 2023. Eight teams, including groups at MIT, Stanford, and Lawrence Berkeley National Laboratory. ↩
10. Chen et al., "Electrochemical Loading Enhances Deuterium Fusion Rates in a Metal Target," Nature 644 (2025): 640–645, doi:10.1038/s41586-025-09042-7. A 30-keV plasma-ion-implantation (beam-target) experiment at UBC's "Thunderbird Reactor": ordinary hot, neutron-producing D–D fusion, in which electrochemical loading produced a 15 ± 2 percent enhancement. The measurement is a clean mainstream result and is not in dispute; only its relevance to the heat-helium-no-neutron pattern is contested. It does not observe neutron-free heat, helium-4 accumulation, or altered branching. ↩
11. C. P. Berlinguette et al., "Revisiting the Cold Case of Cold Fusion," Nature 570 (2019): 45–51. The Google-convened effort did not reproduce excess heat but identified specific lines of inquiry, notably extreme-loading materials science, meriting further study. ↩
12. A. J. Leggett and G. Baym, "Exact Upper Bound on Barrier Penetration Probabilities in Many-Body Systems: Application to 'Cold Fusion'," Physical Review Letters 63 (1989): 191–194; and "Can Solid-State Effects Enhance the Cold-Fusion Rate?" Nature 340 (1989): 45–46. Leggett and Baym derived a rigorous, model-independent upper bound on the equilibrium many-body D–D tunneling rate in a metal, expressed via the measurable thermodynamic affinities of D and He, with no vacuum-barrier assumption, and showed it falls orders of magnitude below the rate needed to explain the claimed heat. The open question is narrower: the bound assumes thermodynamic equilibrium, and whether it constrains driven, far-from-equilibrium cathodes (surfaces, cracks, transient loading gradients) is, on the published record, unresolved. ↩