Laboratory Astrophysics

Laboratory astrophysics is the practice of making, on Earth, the states of matter that stars and planetary interiors are made of, and then measuring them. Three different activities travel under the name, and the brief is worth little if it does not separate them at the outset.

Established The first is direct measurement of numbers that astrophysical models consume. Opacities, equations of state, nuclear reaction cross sections, atomic transition rates. Here the laboratory is not an analogy for anything: a stellar structure code needs the Rosseland mean opacity of iron at roughly 2 million kelvin and an electron density near 10²³ per cubic centimetre, and an experiment either supplies that number with an error bar or it does not. This is the least glamorous part of the field and by a wide margin the most productive.

Frontier The second is the scaled analogue experiment, which generates the press releases. A laser deposits a few kilojoules into a millimetre target, and for nanoseconds the plasma is argued to be hydrodynamically similar to a supernova remnant expanding for a thousand years across ten parsecs. The warrant is Euler similarity: if viscosity, thermal conduction and radiative losses are negligible, ideal hydrodynamics has no intrinsic scale, and systems with matching dimensionless numbers evolve through the same shapes. The criteria were stated precisely in 1999 and say when the analogy holds and, equally, when it does not.

Established The third is code validation, and it quietly justifies the budgets. The radiation-hydrodynamics codes used for supernovae, accretion shocks and inertial fusion capsules are close cousins and sometimes the same code. An experiment no astrophysicist would call decisive can still be the only test a code will get in a regime it is asked to predict. That is a real product, and it is also why the field lives inside nuclear-weapons laboratories rather than astronomy departments.

Two neighbouring briefs own the adjacent questions: what the sky tells us about extreme objects belongs to black hole physics applications, and whether these facilities will ever make net electricity belongs to commercial fusion. This brief owns the terrestrial experiment and the institutions that ration it.

The claim the brief lands: laboratory astrophysics is a productive science bottlenecked not by ideas or physics but by shot rate. Single-shot statistics, unquantified systematics and the impossibility of a parameter scan all follow from the fact that its best facilities fire a few hundred times a year and were built to certify nuclear weapons. The fix is not exotic: it is the repetition-rate driver being developed, for unrelated commercial reasons, by inertial fusion energy programmes.

Established A note on sourcing. This brief was commissioned in September 2026 from the Institute’s research base. Reading-list entries without links are cited from the bibliographic record rather than re-fetched, and claims are dated no later than early 2026 unless carried by a linked source.

Established The National Ignition Facility is the largest high-energy-density instrument ever built and its performance is documented shot by shot. On 5 December 2022 it delivered 2.05 MJ of laser energy to a target and recovered 3.15 MJ of fusion yield, a target gain of about 1.5 and the first controlled ignition. The record since stands at 8.6 MJ from 2.08 MJ delivered, a target gain above four, enabled by a continuous-gradient-doped capsule, with ignition reported nine more times after the first. For this brief the relevant number is different: the same laser draws roughly 300 MJ of electricity to put 2 MJ on target, and the amplifier glass needs hours to cool between shots.

Established Shot rate, not energy, is the currency of the field. A megajoule-class facility fires a few hundred times a year across all missions, of which basic science receives a single-digit percentage. A 30-kilojoule ultraviolet facility of the OMEGA class fires on the order of a thousand times a year; a short-pulse petawatt system once a minute; a diode-pumped hundred-joule system ten times a second. At one shot per week you are performing a demonstration, at a thousand shots a year an experiment, and only above about a hertz a parameter scan with error bars. Almost all of the field’s famous images come from the first regime and almost all of its citable numbers from the others.

Established The strongest single result the field has produced is an opacity measurement, and it embarrassed the models. Bailey and co-workers at the Sandia Z machine measured iron transmission at electron temperatures around 2.1 million kelvin and densities near 3 × 10²² per cubic centimetre — conditions at the base of the solar convection zone — and published in 2015 a wavelength-resolved opacity above every standard model, the Rosseland mean about 7% higher than predicted at the hottest point. No stellar-structure model had been tested at those conditions before, because no facility could reach them.

Frontier That measurement landed inside the solar abundance problem and has not settled it. A downward revision of the Sun’s photospheric metal abundances in the mid-2000s broke the agreement between standard solar models and helioseismology: the convection-zone boundary depth and the surface helium fraction both came out wrong, and an opacity increase near 15% at the convection-zone base would repair it. The Z enhancement is in the right direction but concentrated somewhat hotter than the critical layer, its platform systematics have been contested in the literature, and a competing line of work revises the abundances back upward using new atomic data, repairing the agreement with no new opacity physics. Two candidate resolutions, no arbitration.

Established Turbulent dynamo and collisionless-shock particle acceleration have both been demonstrated, and both are N-of-a-few results. A 2018 OMEGA experiment drove colliding plasma flows through perforated grids, reached a magnetic Reynolds number of several hundred — above the fluctuation-dynamo threshold — and measured field amplification more than an order of magnitude above the Biermann-battery seed: the first laboratory confirmation of a mechanism assumed throughout galaxy-cluster modelling for decades. A 2020 experiment reported a non-thermal power-law electron tail in a laser-produced turbulent collisionless shock, with acceleration attributed to scattering off self-generated magnetic turbulence, bearing directly on the injection problem open since the 1970s. In each case the shot budget permitted the demonstration and not the scaling that the astrophysical application actually needs.

Established Planetary-interior equations of state are the field’s most durable product, and where two facilities have measured the same boundary they have disagreed. Ramp compression has taken iron above a terapascal, three times the pressure at Earth’s centre, constraining the cores of rocky planets several times Earth’s mass; shock-compressed water has been driven into a superionic phase now invoked to explain the non-dipolar magnetic fields of Uranus and Neptune. Against that, the insulator-to-metal transition in dense fluid hydrogen — the boundary setting the interior structure of Jupiter and every hydrogen-rich exoplanet — was located near 300 GPa by magnetically driven flyer plates at Z in 2015 and near 200 GPa by laser shock at NIF in 2018. The gap is attributed to different temperature paths and diagnostics rather than error by either group, and it has narrowed rather than closed. This brief reports the range, not an average, because the average is a number nobody measured.

Established Underground accelerators have changed the ages of the oldest stars. A sub-megavolt accelerator under a kilometre and a half of rock at Gran Sasso remeasured the slowest CNO-cycle reaction, nitrogen-14 proton capture, and found a cross section roughly half the accepted value; because that reaction sets the turn-off luminosity used to date globular clusters, the correction added on the order of a billion years to their inferred ages. A second facility under more than two kilometres of rock in Sichuan has extended the technique to helium burning. Canada contributes through a radioactive-beam recoil separator at TRIUMF, measuring proton-capture rates on short-lived nuclei no stable-beam facility can reach.

Frontier The diagnostic, not the driver, usually limits the result. A high-energy-density plasma exists for nanoseconds, is optically thick to its own emission, and is destroyed by the measurement; X-ray and proton radiography, Thomson scattering and streak cameras each return a line-integrated projection at one or a few instants. Multi-frame hybrid CMOS imagers with nanosecond gating have begun to replace film and streak tubes, converting one datum per shot into a short movie per shot — a larger change to the science than any plausible increase in laser energy.

Frontier Is the Z iron opacity enhancement a property of iron or of the platform? This is the field’s most consequential open question, because it is the one case where a laboratory measurement would directly rewrite a stellar model. Its authors have tested sample tamping, gradients, backlighter spectral structure and attenuation calibration, and the result survives. What it has not had is independent confirmation on a physically different platform. Until it does, the correct posture is that a well-controlled experiment reports something models do not predict, and that the field has not met its own replication standard.

Frontier Do laser-driven collisionless shocks reach the astrophysical regime or only its vicinity? Remnant shocks are high Mach number, weakly magnetized and last for centuries; laboratory shocks are moderate Mach number, run for nanoseconds, and are magnetized by fields the experiment generates itself. The magnetization, the ratio of shock width to ion inertial length, and the number of available scattering times are not all matched simultaneously in any experiment yet performed. Whether the observed acceleration is the astrophysical mechanism or a laboratory cousin is a contested reading of the same data.

Speculative Whether analogue-gravity experiments belong in this field at all. Sonic horizons in Bose-Einstein condensates and in flowing water produce spectra reported as analogues of Hawking radiation. The mathematics is exact for a wave equation on an effective metric; what it cannot do is test whether real black holes radiate, because the gravitational part of the problem has been replaced by a fluid. This brief treats analogue gravity as real condensed-matter physics with no evidential relation to astrophysical black holes, as quantum gravity also does.

Established Repetition rate is the binding constraint, and it is thermal. Flashlamp-pumped neodymium glass converts a few per cent of stored electrical energy into light and dumps the rest into the glass, which then cools by conduction. That single fact sets the few-shots-per-day ceiling on megajoule facilities and therefore the statistical power of every experiment run on them. Diode pumping raises efficiency by roughly an order of magnitude and permits active cooling, which is why hundred-joule diode-pumped systems already run at ten hertz.

Established Targets are hand-made and expensive. A precision cryogenic capsule assembly is machined, characterised and individually inspected, produced in batches of tens, with reported unit costs in the tens of thousands of dollars and up. At one shot per week that is tolerable; at one shot per second it is the entire economics of the facility, which is why target manufacture, injection and tracking is treated by energy programmes as co-equal with the laser.

Established Access is rationed by mission, not by merit. Basic-science allocation on the largest facilities is a single-digit percentage of shots, awarded through heavily oversubscribed proposal rounds. An experiment needing fifty shots for a scan with error bars is in practice not proposable, so the field self-selects toward claims publishable from three shots. This is the mechanism by which a shot-rate constraint becomes an epistemic one.

Frontier Magnetization is hard to buy. Many regimes of interest are set by the ratio of magnetic to thermal pressure, and reaching low plasma beta in a centimetre of hot plasma requires pulsed fields of tens of tesla delivered into the chamber without wrecking the diagnostics. Pulsed-coil platforms exist and remain an order of magnitude short of what several proposed experiments need. The magnets described in high temperature superconductors are a plausible route to steady fields in that range and have not been fielded on a high-energy-density platform.

Established The field waits on laser technology developed for fusion energy, not for science. Diode-pumped solid-state amplifiers, efficient frequency conversion and thermal management at kilowatt average power are the enabling stack, and every serious inertial fusion energy programme is funding them commercially, with public hub programmes assembled since 2023 making the same bet with public money. Laboratory astrophysics is a free rider on this development and should say so plainly.

Established It waits on detectors. Hybrid CMOS X-ray imagers with nanosecond gating, burst-mode framing and radiation-hard readout are the difference between a projection and a movie. Detector development is funded at a small fraction of driver development and delivers more science per dollar than any plausible laser upgrade.

Frontier It waits on atomic and nuclear data infrastructure. An opacity measurement is interpretable only against a model of which transitions exist at which strengths, so the measurement-model discrepancy is a statement about that model. The atomic-structure calculations, line-broadening treatments and databases behind it are maintained by a small community on soft money, and their uncertainties are usually not propagated into the stellar models that consume them.

Established It waits on code access. The best radiation-hydrodynamics codes are export-controlled weapons codes. The open community runs on a smaller set of public codes, and validation experiments are frequently designed against the open code and executed on a machine whose institutional purpose is the closed one. Open experiment, closed model is a structural feature of this field, not an accident.

This section ranks the tests that would most change the brief’s assessment, most decisive first.

Frontier The decisive experiment is an independent replication of the iron opacity enhancement on a physically different platform. If a second facility, using a different driver and a different diagnostic chain, reproduces the measured opacity at solar convection-zone conditions, then laboratory astrophysics has demonstrated its strongest possible claim: that a terrestrial measurement corrected a stellar model. If it does not, the field has found a platform systematic, which is a smaller but still valuable result, and the solar abundance problem reverts to being a question about abundances rather than opacity. Nothing else in this brief carries comparable weight, because nothing else is simultaneously a clean measurement, a live astrophysical controversy, and within reach of machines that already exist.

Frontier Second: a dynamo campaign with enough shots to measure a scaling rather than an occurrence. The 2018 result established that turbulent amplification happens above threshold. What is needed is the dependence of saturated field strength on magnetic Reynolds and Prandtl numbers across at least a decade in each — a campaign of order a hundred shots. At current repetition rates that is years of allocated beam time and has never been awarded. On a ten-hertz kilojoule driver it is a week.

Established Fourth: arbitration of the hydrogen insulator-to-metal transition by a third method. Two facilities disagree by roughly 100 GPa on a boundary that sets the interior structure of every hydrogen-rich planet. A third technique with an independently calibrated temperature diagnostic would resolve a discrepancy that currently propagates as a systematic uncertainty into exoplanet interior models.

Established Fifth, and already running: the inertial fusion energy programmes are a natural experiment on the field’s central constraint. Several private and public programmes are building repetition-rated drivers for reasons unrelated to astrophysics. If any delivers kilojoule pulses at hertz rates with open user access, this brief predicts that error bars in laboratory astrophysics improve sharply and immediately, because the queue of experiments that cannot currently be proposed is long and known. If repetition rate arrives and the science does not improve, the shot-rate diagnosis here is wrong.

Handwave The experiment most often proposed in popular coverage — making a black hole, or tearing the vacuum, in the laboratory — is not on this list. The field at which the vacuum breaks down into electron-positron pairs is some five to six orders of magnitude in intensity above the most powerful focused pulses yet achieved, and schemes for closing the gap rely on colliding a pulse with a relativistic electron beam so the field is enhanced in the electron rest frame. That is a real and funded direction. It is not a laboratory black hole and no facility roadmap claims it is.

Frontier A facility purpose-built for laboratory astrophysics has never been built, and its specification is not mysterious. One to ten kilojoules in shaped nanosecond pulses at one to ten hertz, a co-located short-pulse arm for backlighting and particle probing, a target injector, a multi-frame imaging suite and an integrated pulsed-field capability. Nothing on that list requires a physics advance; every element exists somewhere at demonstration scale.

Established Target handling at rate is the hardest engineering problem in the specification. Injecting a millimetre target, tracking it to micrometre accuracy, and firing a steered beam at it several times a second is a mechatronics problem the energy community has adopted as a first-order requirement and has not demonstrated in an integrated system at rate.

Frontier Pulsed-power and laser platforms should not be assumed interchangeable. A z-pinch delivers tens of megaamperes over a hundred nanoseconds into a centimetre-scale load and reaches large, relatively uniform, radiation-dominated volumes; a laser delivers a shorter, smaller, hotter, more sharply structured event. Opacity and photoionized-plasma work has favoured pulsed power for the volume, hydrodynamics has favoured lasers for the timing control. At least one privately funded z-pinch programme reports shots at rates measured in shots per minute, which if sustained opens the pulsed-power regime to the same statistical improvement.

Commercial fusion is the closest neighbour and the relationship runs one way. Fusion programmes build the drivers, targets, chambers and funding case; laboratory astrophysics uses the resulting machines on a minority allocation and returns validation data. Anyone modelling this field’s future is modelling fusion capital expenditure whether they intend to or not.

Fusion spacecraft is adjacent through pulsed-power physics rather than astrophysics: the pulsed fission-fusion and magneto-inertial concepts examined there depend on the same compression physics and inherit the same difficulty of measuring a plasma that exists for nanoseconds.

Black hole physics applications owns the observational side of the extreme-object question. The division of labour is clean: that brief reports what has been seen, this one what has been made and measured, and they meet only in the accretion and shock physics where laboratory measurements feed the models used to interpret observations.

High temperature superconductors are adjacent as an input: steady high fields in a diagnostic-accessible geometry are among the most requested and least available capabilities in high-energy-density experiment. Exotic materials for propulsion consumes the equation-of-state and ablation data this field produces as a by-product.

Nuclear astrophysics at underground accelerators and rare-isotope facilities is adjacent and is covered by no brief on this map. It is the branch with the highest ratio of settled result to expenditure, and its absence is a gap worth recording.

Established The field’s instruments belong to a weapons programme and its scientists to universities. The arrangement works better than that sentence suggests: basic-science allocations are real, proposal rounds are externally reviewed, and laboratory staff have consistently defended open publication. It nonetheless means scientific priorities are negotiated annually against a mission that outranks them, and that a change in national-security budgeting is a larger risk to laboratory astrophysics than any scientific development.

Established Networked access was the corrective, and it was deliberate policy. A 2018 national assessment found the United States had lost its leading share of high-intensity laser capability to Europe and Asia and recommended a coordinated network and new investment. The resulting user network now provides peer-reviewed open access to roughly ten petawatt-class facilities in the United States and Canada, and new high-power capability was funded at a university rather than a national laboratory. That is the most consequential institutional change in the field in two decades, and it was cheap: the network’s operating funding is a rounding error against one megajoule facility.

Established Canada participates through beam time rather than capital. Canadian high-intensity laser capability is concentrated in a single advanced laser source facility in Quebec that participates in the North American user network, and Canadian nuclear astrophysics runs through a rare-isotope facility in British Columbia. This is the standard posture for a mid-sized science economy, with a specific vulnerability: a country contributing users but not facilities has no leverage when allocation policy changes.

Speculative The private energy sector is about to become an institutional actor in a field it has no obligation to serve. If the rep-rated drivers are built by companies, open scientific access becomes a contractual question rather than a policy one. No standing mechanism exists for buying user time on a private high-energy-density facility, and the analogues from other fields — beamline buy-in, collaboration agreements, publication embargoes — each carry costs this community has not debated.

Established This is dual-use science in the most literal sense available. The facilities exist to maintain confidence in nuclear weapons without nuclear testing. Laboratory astrophysics is not a cover for that mission; it is a genuine secondary use, and the physics is the same physics. Anyone working in the field contributes at some margin to the credibility of a deterrent, and the honest framing is that this is a trade the community has largely accepted without debating in public.

Established Publication asymmetry is a real harm to the science. When the best model is classified and some data restricted, replication becomes structurally harder and the community loses the ordinary corrective of an outsider re-analysing the raw record. The field compensates with unusually detailed methods sections and open codes, and the compensation is partial.

Frontier Single-shot science strains the norms that make results trustworthy. A discipline whose modal published result rests on three shots analysed with a code the reader cannot run has weak error correction by construction. There is no evidence of systematic misconduct here and there is a clear structural vulnerability, and the appropriate response is the boring one: pre-registered shot allocations, published null results, and mandatory release of raw diagnostic records.

Established Laboratory astrophysics sets the error bars on statements about the rest of the universe. Stellar ages, and therefore the minimum age of the Galaxy, rest on nuclear cross sections measured underground. Exoplanet interior structure, and therefore judgements about which planets could have magnetic fields or plate tectonics, rests on equations of state measured by shock and ramp compression. Supernova light curves used as the cosmic distance ladder rest on opacities and mixing prescriptions calibrated in part against laboratory experiment. None of these are analogies; they are numerical inputs.

Frontier The civilizational argument is epistemic rather than economic. Astronomy cannot perform experiments on its objects, and laboratory astrophysics is the only mechanism by which any part of it becomes experimental. That is a structurally important thing for a civilisation to maintain and nearly impossible to justify on cost-benefit grounds in any single budget cycle, which is why it survives as a passenger on other missions.

Handwave Claims that this field is a route to exotic technologies do not survive an order-of-magnitude check. Vacuum energy extraction, artificial singularities and tabletop antimatter economies each fail on the arithmetic: the energy per shot is a few megajoules, comparable to a kilogram of high explosive, and field strengths remain far below the vacuum breakdown scale. The transformative content of this field is measured in corrected model parameters, which needs no inflation.

These horizons track the capability that governs the science — repetition rate with open access — rather than any particular discovery, because this brief’s argument is that discoveries here follow shot statistics.

• 10 yr: Frontier Kilojoule-class drivers at one to ten hertz enter service through inertial fusion energy programmes and at least one is opened to peer-reviewed external users; multi-frame hybrid CMOS imaging becomes standard; the opacity replication question is answered one way or the other. Expect the field’s first campaigns with genuine parameter scans and published error budgets.
• 25 yr: Speculative Laboratory measurement becomes a routine calibration input to stellar and planetary model pipelines rather than an occasional correction, with round-robin comparison between facilities treated as normal practice, and a dedicated high-repetition-rate facility for basic high-energy-density science exists somewhere — most plausibly as a secondary user programme attached to a fusion pilot plant.
• 50 yr: Speculative If inertial fusion energy becomes an industry, high-energy-density experiment becomes cheap the way synchrotron beam time became cheap and the constraint moves from shots to ideas. If it does not, the field continues roughly as today, dependent on weapons-programme facilities and vulnerable to their budgets.
• 100 / 250+ yr: Handwave Claims about laboratory access to vacuum breakdown, analogue horizons with evidential force, or engineered exotic states at astrophysical scale are not forecasts and this brief declines to date them. What can be said is that the gap between achievable and required field strength for genuine vacuum nonlinearity is measured in orders of magnitude, not in decades of incremental improvement.

• Depends on Commercial fusion, not for a scientific result but for the machines: every capability this brief says would transform the field is on a fusion programme’s critical path and on nobody else’s. Secondarily high temperature superconductors, the credible route to steady high fields in a diagnostic-accessible geometry. No pending physics result blocks this topic; the blockage is capability and access.
• Requires (not on this map) Diode-pumped amplifiers delivering kilojoules at ten hertz, the single capability that converts demonstrations into experiments. Multi-frame hybrid CMOS X-ray imagers that turn one datum per shot into a time series. Cryogenic target capsules made in thousands rather than tens, at a cost that survives a hertz-rate duty cycle. A similarity theory still valid when radiation transport and kinetic particle effects are not negligible, which is the regime where most of the interesting astrophysics lives and where Euler scaling is silent. An institutional guarantee of open, peer-reviewed shot time on facilities whose primary mission is something else. And capital willing to build a machine whose only product is knowledge.
• Enables Corrected opacities, equations of state and reaction rates feeding stellar, planetary and supernova models; validation data for the radiation-hydrodynamics codes that commercial fusion and fusion spacecraft both depend on; and the diagnostic toolchain any pulsed high-energy-density device will need.
• Adjacent Black hole physics applications on the observational side; exotic materials for propulsion, which consumes this field’s equation-of-state and ablation data; quantum gravity, which shares the analogue-gravity literature and treats it with the same caution; and artificial scientists, the plausible operating mode for a facility firing ten times a second.

Handwave “They made a supernova in the laboratory.” No experiment has produced a supernova, a star or a black hole. What has been produced is a plasma whose hydrodynamic evolution, under stated and checkable conditions, follows the same scale-free equations. The similarity is a licensed statement about one subsystem, usually the mixing of two fluids across an unstable interface, and says nothing about nuclear burning, neutrino transport or collapse. The claim is defensible in its narrow form and the narrow form is almost never the one reported.

Handwave “Ignition at a megajoule laser means fusion power is here.” Target gain above four is a genuine milestone and the wall-plug arithmetic is unforgiving: roughly 300 MJ drawn per shot, a best yield of 8.6 MJ, a few shots per week. The gap is three numbers — laser efficiency, repetition rate, target cost — and commercial fusion argues the point properly. For laboratory astrophysics the significance of ignition is different and real: it showed that integrated predictive modelling of a high-energy-density system can be made to work.

Frontier “The Sandia opacity result solved the solar abundance problem.” It did not, and the people who made the measurement have not claimed it did. It supplied a discrepancy in the right direction at conditions adjacent to those that matter. A competing resolution revises the abundances using improved atomic data and dispenses with new opacity physics. Both remain live, and this brief’s decisive experiment is designed to tell them apart.

Handwave “Petawatt lasers are approaching the point where they tear the vacuum apart.” The critical field for spontaneous pair production corresponds to an intensity five to six orders of magnitude above what the most powerful focused pulses achieve. The legitimate programme colliding an intense pulse with a relativistic electron beam exploits the Lorentz boost to the electron rest frame to probe strong-field quantum electrodynamics, which is a real and important goal and not the same as reaching the critical field in the laboratory frame. Facility names invoking higher power equivalents describe the boosted frame, and that distinction is routinely lost in coverage.

Handwave “Low-energy nuclear reactions are laboratory astrophysics too.” They are not, and the boundary is worth policing. Claims of nuclear reactions at ambient conditions have been examined repeatedly, including in a deliberately rigorous and well-funded modern attempt that reported no evidence, and they belong to a different literature with a different evidentiary record. Laboratory astrophysics reaches astrophysical conditions by force — megajoules, megaamperes, gigapascals — and its credibility rests on the conditions themselves being measured.