Hand-considered hypotheses, 1101–1200
Written one at a time, in the order printed. Each entry has its own logical form, named above it, and the forms were rotated deliberately so that no two consecutive entries argue the same way. Subjects are drawn from solid-state and inorganic synthesis: ceramic densification, ordered intermetallics, framework solids, epitaxial films, hydrothermal and solvothermal crystallisation, and defect chemistry. Nothing here has been measured. Where a number appears it is a stated estimate, offered to make the claim testable, not as data. This is the batch numbered 1101 to 1200.
Oxide ceramics and sintering
Misattribution
H1101. Textbook activation energies for the densification of submicron oxides are extracted from shrinkage curves that begin while the powder is still losing chemically bound water, so the first-stage slope records dehydroxylation rather than grain-boundary diffusion. If that reading is right, a powder calcined to constant mass at 700 degrees keeps the final-stage slope and loses the low-temperature step entirely. Paired dilatometry on split batches, one calcined and one not, with matched particle-size distributions, separates the two.
Hidden term
H1102. Every measured sintering shrinkage contains a term from the dilatometer pushrod's own expansion, and for a sample shorter than about five millimetres that term is comparable to the shrinkage being reported (estimate, order of magnitude). The correction is not a constant, because it scales with the pushrod's thermal gradient, which moves when the furnace hot zone drifts. Two sample lengths in one run separate the contributions exactly, since only the sample term scales with length.
Identity
H1103. Two-step sintering and rate-controlled sintering are taught as distinct strategies, yet both hold a body in a temperature window where boundary diffusion outruns surface diffusion; the schedules differ and the mechanism does not. Any two-step schedule can therefore be rewritten as a rate-controlled one with the same density trajectory. Running both to one intermediate density and comparing pore-size distributions, which ought to be indistinguishable, is the check.
Bound
H1104. For a given powder, the density attainable at a chosen grain size is bounded by the competition of boundary against surface diffusion, and no heating profile pushes past that bound because the two fluxes share one driving force. The familiar density-grain-size map then becomes a property of the material rather than of the schedule. Three powders of equal composition but different surface areas will collapse onto one curve when density is plotted against grain size, or the map is a schedule artefact after all.
Threshold
H1105. Below a critical green density a compact cannot reach closure at any hold time, because the pore network stays connected to the atmosphere and the driving force for elimination vanishes once the pores are stable against it. For a typical oxide the threshold is estimated near 55 per cent of theoretical density, shifting with atmosphere. Mercury porosimetry before and after a long hold distinguishes a kinetic plateau from a true equilibrium.
Reversal
H1106. Volatile sintering aids help a compact densify only while they remain, and past the dosage at which they evaporate before the neck network is rigid the same additive coarsens the body instead. The reversal is set by the aid's vapour pressure against the heating rate, not by its chemistry. A dosage series fired in a closed and in an open crucible isolates it, since only the open one loses the aid.
Count
H1107. Varying temperature, time, atmosphere and heating rate in a full factorial still leaves at most four independent kinetic parameters for densification, so a fit reporting more is absorbing noise. Fitting one shrinkage curve at increasing parameter counts and watching the residuals stop improving is enough to see it. The count belongs to the model, not to the number of experiments run, which is why the largest sintering studies disagree most.
Artefact mimics effect
H1108. Crucible volatiles, not the additive, explain many reports of activated sintering: a trace of a low-melting oxide evaporating from the boat deposits on the compact's upper face and densifies it there first. The vertical asymmetry is the giveaway, appearing on the top before the bottom. Standing the sample on a sacrificial spacer of additive-free powder should remove the enhancement while leaving a genuine chemical effect untouched.
Scale argument
H1109. Once particle size falls below the scale at which surface curvature is set by faceting rather than by the particle radius, sintering stress stops scaling as one over radius and becomes a constant of the facet. For an oxide that crossover is estimated in the tens of nanometres. A log-log plot of shrinkage rate against particle size across the crossover settles it, since divergence and convergence are hard to confuse.
Inversion of standard practice
H1110. Reducing atmospheres are the standard prescription for oxides with easily reduced cations, but near a redox boundary a reducing gas stabilises a vacancy population that suppresses densification, and the correct practice is a mildly oxidising atmosphere that fixes the cation valence. A pO2 series at fixed temperature locates the density maximum instead of assuming it. If that maximum sits above the equilibrium pO2, the prescription is inverted.
Selection effect
H1111. Published sintering aids are the ones that worked, so the apparent consensus that a few per cent of a eutectic-forming oxide helps is conditioned on survival in the literature. The mechanism, a liquid that must wet the boundary and dissolve the host, predicts that many untried oxides serve equally well. A prospective trial on a random subset of candidates, reported whatever the outcome, would size the bias.
Unit trap
H1112. Per cent theoretical density is ambiguous whenever the reference cell differs from the cell the sample adopts, and a phase transition during sintering moves that reference by several per cent (estimate: two to four for a perovskite). Two laboratories can then dispute whether a body reached 95 per cent while measuring identical porosity. Reporting image-analysis porosity beside the density ratio, for one sample straddling the transition, resolves the disagreement.
Intermetallics and ordering
Misattribution
H1113. Nanocrystalline intermetallics owe their strength to order strengthening in the standard account, but below the antiphase-boundary spacing the dominant term is suppression of cross-slip at boundaries, which is not an order effect at all. The signature is that the strength anomaly at the ordering temperature fades as grain size falls. Ordered and disordered alloys matched in grain size would separate the two contributions.
Hidden term
H1114. Lattice parameters of hot-pressed intermetallics are usually reported without degassing, and residual hydrogen from the powder surface can expand the cell by as much as a solute does (estimate: up to a tenth of a per cent). An in-situ vacuum anneal at moderate temperature will shrink the cell with the composition unchanged. Measuring one specimen before and after that anneal is the whole experiment.
Identity
H1115. Environmental and intrinsic embrittlement of a nickel aluminide are one mode seen at two strain rates, the environmental version being the intrinsic one with hydrogen supplied faster than it can escape. If so, the ductility minimum should fall at the strain rate where diffusional supply matches mechanical demand. Testing in vacuum at that rate should reproduce the curve obtained in air.
Bound
H1116. Order strengthening saturates at the antiphase-boundary energy divided by the superlattice period, and alloying cannot push past it because the mechanism that strengthens also supplies the dislocation path that bypasses it. The ceiling is reachable with a composition series that varies ordering energy independently of defect structure. Strengths reported above it would indicate a different mechanism, not a better alloy.
Threshold
H1117. Titanium aluminide lamellae cannot form below a critical aluminium content, because the two phases' lattice mismatch exceeds what the interface can elastically accommodate and the boundary goes incoherent. That content is estimated at a few atomic per cent and moves with cooling rate. A composition series quenched at two rates places the lamellar-to-equiaxed transition at two different compositions.
Reversal
H1118. Boron refines grains in an intermetallic and improves ductility only until the borides themselves become crack initiators, after which toughness falls with further addition. Where it turns over is set by particle spacing relative to the plastic zone size, not by boron content alone. Measuring that spacing across a boron series, rather than assuming it scales, is what makes the reversal visible.
Count
H1119. Near a continuous ordering transition an intermetallic has at most two independent order parameters, so a diffraction study reporting three coexisting ordered phases is resolving domains rather than phases. Dark-field imaging of one region in two superlattice reflections shows whether the variants share a domain. Counting phases from peak multiplicity overstates the diagram, because twinning multiplies peaks without multiplying phases.
Artefact mimics effect
H1120. Compressing a brittle intermetallic between stiff platens produces a load that keeps rising after failure, and that rise is routinely read as plasticity. The effect is the cracked specimen constrained by the loading train, so it mimics ductility and vanishes when compliance is added. Two platen materials of different stiffness on identical specimens change only the artefact.
Scale argument
H1121. Lamellar intermetallics stop obeying inverse-spacing strengthening once the spacing approaches the dislocation core width, estimated at a few nanometres, because the interface can no longer emit a dislocation into the confined layer. The expected signature is a plateau in strength against spacing. Controlled cooling produces the spacing series, and the plateau is a length-scale count rather than a new deformation mode.
Inversion of standard practice
H1122. Vacuum annealing is the wrong final step for an intermetallic whose minority phase is volatile, since it strips that phase from the surface and leaves a soft brittle case over a sound core. Annealing under a partial pressure of the minority species is the inversion that preserves the surface. Hardness profiles after the two treatments, on one alloy, show whether such a case exists.
Selection effect
H1123. The correlation between ordering temperature and strength in intermetallics is conditioned on which alloys get tested hot, since high-ordering-temperature compositions are taken to high temperature because they are expected to be useful there. No intrinsic correlation can survive normalising the test temperature. Reanalysis stratified by homologous temperature, or a prospective run on a low-ordering alloy at high temperature, would settle it.
Thermodynamic bookkeeping
H1124. Mechanically alloyed intermetallics described as metastable are usually metastable only with respect to themselves at a larger grain size, because the enthalpy stored in boundaries of a ten-nanometre grain is estimated at a large fraction of the formation enthalpy. The phase is stable against the elements and unstable against coarsening. Annealing to coarsen without transforming is the test, and transformation should follow coarsening rather than accompany it.
Framework materials: zeolites and metal-organic frameworks
Misattribution
H1125. Zeolite rates are normalised to the silicon-to-aluminium ratio, yet the rate per acid site also changes with crystallite size because the probe's access is diffusion-limited, and normalising to external surface area removes most of the apparent trend. The mechanism is a Thiele modulus that varies with crystal size at fixed composition. One framework at three crystallite sizes, probed with a molecule too large to enter the pores, tests it cleanly.
Hidden term
H1126. Reported surface areas of metal-organic frameworks carry a term from residual solvent left by activation, and in pores below about a nanometre that term can exceed the framework's own contribution because the solvent holds volume the probe cannot reach. Ramping the activation temperature on one sample shows the area rising to a plateau. Where the rise stops is the activation temperature, not a property of the material.
Identity
H1127. Breathing in a flexible framework is one gate-opening transition passed at different rates, not two mechanisms, and the reported single-step and two-step variants differ only in how fast the guest was removed. Step the guest's chemical potential at several rates and the transition lands at the same guest content. If the transition composition is invariant while the number of steps changes, the mechanisms are one.
Bound
H1128. Maximum acid-site density in a zeolite is limited by the separation at which two sites stop behaving as isolated acids, since a paired site is weaker than either alone, and that limit arrives well before the framework runs out of aluminium. Roughly two tetrahedral sites apart is the estimate. Rates per site across a series that crosses the limit bend where pairing begins, not where aluminium runs out.
Threshold
H1129. A zeolite needs a threshold partial pressure of structure-directing agent to crystallise in the intended framework, because below it one template per cage is no longer guaranteed and the cage that selects the product is only partly occupied. Titrating the template and following the phase by diffraction locates the threshold. Above it, phase purity is insensitive to further template.
Reversal
H1130. Lengthening the linker in an isoreticular series raises porosity until the strut becomes flexible enough to interpenetrate, after which accessible pore volume falls with each added ring. The turning point belongs to the topology rather than to the metal. A homologous series with low-temperature gas uptake and independently confirmed topology shows the maximum, and a series that cannot interpenetrate must not show it at all.
Count
H1131. A powder pattern of a new framework supports no more independent phase fractions than it has resolved reflections, so a refinement reporting five phases from twenty reflections is underdetermined rather than impressive. Refining synthetic patterns of known mixtures and counting how many phases can be recovered gives the real number. That number is a property of the data, not of the sample.
Artefact mimics effect
H1132. Negative thermal expansion measured by diffraction in a hydrated framework is frequently the beam driving water off the sample, which changes the lattice without any phonon mechanism and mimics a real effect convincingly. A sealed capillary at controlled humidity and an open one, measured through the same temperature series, separate them. Only the open sample should contract.
Scale argument
H1133. Diffusion in a framework changes character when the pore narrows below about twice the guest's radius, passing from viscous flow to activated hopping, and this crossover is not the Knudsen one. A homologous guest series in a single framework puts the change at a pore size rather than a pressure. The activation energy steps at the crossover instead of varying smoothly through it.
Inversion of standard practice
H1134. Air calcination is the wrong activation for organic-templated frameworks containing redox-active metals, because direct calcination reduces the metal before the template leaves and collapses the structure; a slow inert ramp followed by mild oxidation is the inversion that works. Comparing the two schedules on one batch, with the metal oxidation state measured before and after, decides it. A structure surviving only the staged route proves the point.
Selection effect
H1135. Water stability of frameworks is reported for the survivors, since a preliminary exposure decides which samples reach the published test, and the literature's screening sets are conditioned on that survival. A random sample tested without preselection returns a far lower survival fraction. A prospective trial on an unselected library, with failures reported, is the only way to size the gap.
Unit trap
H1136. Surface areas of microporous frameworks are quoted without the pressure range used for the fit, and different defensible ranges give areas differing by ten to thirty per cent in one isotherm (estimate). Because the range is chosen after the isotherm is seen, the published number is a decision rather than a measurement. Publishing the full isotherm with areas from several ranges for one standard material would expose the spread.
Thin-film growth and epitaxy
Misattribution
H1137. Strain in a low-temperature epitaxial film is credited to lattice mismatch, but a large share of it can be a point-defect supersaturation that relaxes on annealing without any dislocation moving. Growing at two temperatures at identical mismatch leaves only the defect term different. The low-temperature film's peak shifts on annealing while its dislocation density stays put.
Hidden term
H1138. Growth rates from a magnetron carry the tooling factor's drift, which changes as the target erodes, and the rate can move by a factor of two across a target's life with every other setting fixed (estimate). A quartz monitor at the substrate plane rather than at the target exposes it. The drift then follows the erosion profile rather than the elapsed time.
Identity
H1139. Island coalescence and grain growth in a thin film are one event observed by two instruments: the microscope sees grains merge while the stress measurement records the same merger as a tensile jump. Synchronising the two on one growing film shows whether the events coincide. Coincidence in time, across several thicknesses, would collapse two literatures into one.
Bound
H1140. Epitaxial perfection cannot exceed the substrate's subsurface damage, because the damage's strain field propagates through the film and seeds dislocations there, so the deposition method sets the ceiling only when the substrate is better than the film. Growing identical films on substrates prepared to different depths of damage tests it. The best film should be no better than the best substrate.
Threshold
H1141. A misfitting film cannot relax below a critical thickness because the elastic energy available is less than the core energy of the dislocation it would have to create, so relaxation switches on abruptly rather than gradually. A thickness series at one mismatch shows the switch, and a second series at a different mismatch moves it predictably. Both series together distinguish core-energy control from strain-energy control.
Reversal
H1142. Raising substrate temperature improves crystallinity until interdiffusion turns the interface into a graded alloy that no longer templates the film, after which order falls with further heating. The optimum coincides with the onset of measurable interdiffusion, not with a fixed temperature. Depth profiling the interface across a temperature series locates both features in one experiment.
Rank argument
H1143. Substrate temperature, pressure, power and target distance do not supply four independent controls, because pressure and distance enter the deposition only through their product in the mean free path, so the sensitivity matrix has rank three. A design of experiments that estimates that rank exposes the redundancy. It also explains why studies using different distance conventions reach different optima for otherwise identical films.
Artefact mimics effect
H1144. Self-limited growth in atomic layer deposition is often precursor depletion within the dose pulse, which produces the same plateau in thickness against dose as true surface saturation. The artefact and the ideal are indistinguishable on that curve alone. Raising the pulse's partial pressure at fixed dose changes saturation but not depletion, so the plateau moves for one and not the other.
Scale argument
H1145. Resistivity of a metal film stops following surface scattering once the grain size rather than the thickness sets the mean free path, and for a typical metal the crossover is estimated at ten to twenty nanometres. A thickness series at fixed grain size and a grain-size series at fixed thickness separate the two controls. The thickness series shows the plateau and the grain-size series does not.
Inversion of standard practice
H1146. A single oxidising anneal is the wrong repair for a transparent conducting oxide grown oxygen-deficient, because it passivates the grain boundaries before the bulk is repaired and locks in the deficiency; a brief reducing step followed by slow oxidation is the inversion. Two schedules at matched thermal budget, compared on mobility rather than carrier density, show which repaired the bulk. Mobility rises only when the boundaries were not oxidised first.
Selection effect
H1147. Strain engineering of a functional oxide is reported for films that stayed coherent, while films that relaxed are filed as growth failures rather than as data, so the strain-property map is truncated at the strains that happen to be achievable. Reported enhancements shrink in a prospective series that includes every strain attempted. Publishing all attempted strains for one composition would measure the truncation.
Unit trap
H1148. X-ray reflectivity returns a mass thickness while profilometry returns a geometric one, and the two differ by the film's density deficit, by as much as a factor of 1.3 in a porous film (estimate), which is enough to change a strain calculation. Reporting both for one sample shows the gap. The gap tracks the refractive index deficit rather than the deposition method.
Hydrothermal and solvothermal routes
Misattribution
H1149. Crystal habit from a hydrothermal run is credited to the mineraliser when the autoclave's own thermal gradient is doing the selecting, dissolving the hot end and growing the cold one. A rotating autoclave averages the gradient away without changing the chemistry. If rotation makes the habit equant, the gradient was responsible and the mineraliser was not.
Hidden term
H1150. Hydrothermal yields omit the product that adheres to the liner wall and leaves with the washings, and in a small batch that fraction can reach a fifth of the solid (estimate). Weighing the liner before and after cleaning recovers it. The omission is why small-batch yields scatter far more than large-batch ones from the same recipe.
Identity
H1151. Ostwald ripening and dissolution-reprecipitation in a closed vessel are one process, both driven by size-dependent solubility and differing only in whether the shrinking crystals are called dissolving or consumed. A bimodal population tracked over time has the two modes converging at a rate set by that solubility. If the measured rate matches the solubility curve computed from the particle sizes, the two labels describe nothing.
Bound
H1152. A solvothermal yield is capped by the solvent's own decomposition at the reaction temperature, because past that point the solvent's reducing power shifts the product's oxidation state and the target phase stops forming. The cap is a property of the solvent, not of the precursor. A temperature series with the solvent's decomposition measured independently by pressure rise should show yield falling where decomposition begins.
Threshold
H1153. A hydrothermal reaction needs a minimum water activity even at high temperature, because dissolving the oxide requires a hydrated surface intermediate, and the threshold is set by water content rather than by total pressure. Water-in-alcohol mixtures held at fixed pressure test it: conversion collapses at a water fraction, not at a pressure. A threshold that follows mole fraction rather than fugacity rules out a purely mechanical explanation.
Reversal
H1154. Chelating agents enlarge solvothermal crystals by slowing nucleation, until the ligand itself becomes the sink that holds the metal and the product turns amorphous, with the turning point near a ligand-to-metal ratio of one. A titration series with crystallinity followed by diffraction shows the window closing. Past the turning point the solid comes out glassy rather than microcrystalline.
Mass-balance argument
H1155. Fixing the yield of a hydrothermal recipe removes one degree of freedom, so a recipe optimised for yield has one fewer independent variable than the search that produced it and will not reproduce at another yield. Re-optimising with the yield fixed shows the reduced dimension. This is a mass balance rather than a statistics claim, and it explains recipes that work once.
Artefact mimics effect
H1156. Supercritical behaviour inferred from pressure inside a sealed vessel is often the vessel's thermal expansion answering the heater, and every pressure-temperature curve then shows an apparent transition that is not the solvent's critical point. A vessel of known compressibility with an inert fill fraction separates them. The apparent transition moves with fill fraction if the vessel is responsible.
Scale argument
H1157. Induction times in hydrothermal crystallisation stop depending on vessel size once wall nucleation dominates, at a surface-to-volume ratio estimated in the range of a few reciprocal centimetres. The reason is that a wall supplies a fixed number of sites per unit area, so its contribution grows with area while the bulk contribution grows with volume. Vessels spanning a decade in size at one supersaturation decide it: the smallest show a plateau, the largest keep the bulk scaling.
Inversion of standard practice
H1158. Seeding is the wrong control for a system that grows by oriented attachment, because seeds supply the very attachment sites the seed was meant to pre-empt, and an inhibitor is the correct addition. Seeded and inhibited runs at matched supersaturation compare directly. The seeded run comes out broader in size distribution, inverting the usual expectation that seeding narrows it.
Selection effect
H1159. Hydrothermal routes are credited over calcination on compositions chosen because they are hydrothermally accessible, while calcination is demonstrated on refractory ones, so the comparison is conditioned on the pair chosen. A head-to-head trial on a random composition set, with both routes attempted on every member, is the test that has not been run. It either shows a general advantage or it shows that the advantage was the sampling.
Unit trap
H1160. Fill fraction is quoted by liner volume in one paper and by solvent mass in the next, and the two differ by the solvent's density, enough at typical conditions to move a synthesis from one product to another. A fill series with both conventions stated exposes it. Cross-laboratory reproducibility improves as soon as the convention is explicit.
Defect chemistry and non-stoichiometry
Misattribution
H1161. A conductivity maximum in a doped oxide is read as defect association when it is often the dopant's solubility limit, beyond which a second phase coats the boundaries and blocks them. Measuring the boundary composition directly, rather than inferring it, decides between them. A maximum that coincides with the first appearance of a second phase is solubility and not association.
Hidden term
H1162. Thermogravimetric non-stoichiometry includes a buoyancy term that changes as the sample's volume changes on reduction, and for a large sample it can be mistaken for a composition change of several tenths of a per cent (estimate). A blank run with an inert sample of matched volume supplies the correction. That blank also removes the apparent discontinuity at the reduction onset.
Identity
H1163. The two defect regimes reported for a perovskite, one at low oxygen partial pressure and one at high, are one regime viewed from two sides of the same neutrality condition, since the majority defect changes sign of charge without changing mechanism. Conductivity and Seebeck coefficient measured in one cell across the range should cross where the majority changes. The transition belongs at the crossing, not at either slope change.
Bound
H1164. Ionic conductivity cannot exceed the host's attempt frequency weighted by the best pathway's migration barrier, and doping raises the carrier concentration rather than lowering that barrier, so no dopant beats the bound. The bound is computable from the host's phonon spectrum. Best-doped and undoped host compared at one temperature, with carrier concentration measured independently, would show whether the barrier moved.
Threshold
H1165. Defects in a non-stoichiometric oxide order into a superstructure below a critical temperature set by their interaction energy rather than by their formation energy, and the ordering appears as an extra reflection on cooling. Conductivity activation energy above and below that temperature should differ. The change coincides with the reflection's appearance, which ties the transport to the ordering.
Reversal
H1166. Donor doping raises electronic conductivity until the dopant's charge is compensated by cation vacancies that scatter carriers more than the donors supply, after which conductivity falls with further doping. The crossover follows the compensation mechanism rather than the dopant's identity. Sweeping composition across the turn, with mobility separated from carrier density, shows the two terms trading places.
Conservation argument
H1167. Site balances plus electroneutrality leave at most three independent defect concentrations in a ternary oxide of fixed cation ratio, so a model invoking more is fitting correlated noise. Count the constraints first and the number of free concentrations follows arithmetically, with no fitting needed. That is why published defect models for one material can match the same data while disagreeing about which defect dominates.
Artefact mimics effect
H1168. Colossal dielectric constants in doped oxides are frequently electrode contact resistance in series with the sample, which produces a large apparent capacitance at low frequency and mimics a real polarisation. Two sample thicknesses in one cell at fixed electrode area change only the artefact. A constant that scales with thickness is a property of the sample; one that does not is a contact.
Scale argument
H1169. When the space-charge layer at a boundary grows thicker than the grain, the boundary no longer depletes a finite region but shifts the whole grain's Fermi level, and the conductivity's dependence on grain size inverts. The crossover is estimated in the tens of nanometres for a typical oxide. A grain-size series with the boundary potential measured independently puts the inversion at that size.
Inversion of standard practice
H1170. Quenching is the wrong way to preserve a high-temperature defect state when the defects associate faster than heat can be removed, because the quench freezes a configuration that never existed at temperature. Equilibrating at temperature and measuring in situ is the inversion. A quenched and an in-situ conductivity at one temperature differ if the quench created its own state.
Selection effect
H1171. Reports of vacancy ordering in oxides are conditioned on samples cooled slowly enough to order, so phase diagrams built from them lean toward the ordered side and the disordered state is underrepresented. The prediction is that ordering vanishes above a temperature rarely reached in synthesis. Cooling at several rates and extrapolating to infinite time would locate it.
Thermodynamic bookkeeping
H1172. A reduced oxygen content measured in a sealed cell cannot be lower than the cell's getter allows, so a reported thermodynamic limit can be the apparatus's own floor. The same measurement with and without an internal getter moves the apparent limit. Bookkeeping of this kind is what separates a materials limit from a chamber limit.
Solid-state reaction kinetics and precursors
Misattribution
H1173. The rate-limiting step of a solid-state reaction is inferred from the temperature dependence of conversion, but sintering of the compact produces the same signature by shrinking the interfacial area as the reaction proceeds. Running the reaction diluted in an inert matrix tests it. If sintering is responsible, dilution removes the drift in apparent activation energy.
Hidden term
H1174. Reaction onset temperatures from calorimetry include the instrument's thermal lag, which scales with heating rate and sample mass, and for a ten-milligram sample at twenty kelvin per minute the lag is estimated at several kelvin. Three masses at one heating rate give the correction by extrapolation. The mass-independent onset is the number worth reporting, and the only one that travels between laboratories.
Identity
H1175. Mechanochemical activation and a nanoscale precursor are two names for raising the contact area between reactants, so a milled mixture and a nanoparticulate one follow identical reaction curves once normalised to measured interfacial area. Gas adsorption supplies that normalisation. Identical curves would show that the two routes differ only in how the area was made, not in what the area does.
Bound
H1176. Completeness of a solid-state reaction is capped by the diffusion distance between particle centres, and for micrometre particles the cap leaves a few per cent unreacted no matter how long the anneal, because the last fraction must diffuse through product. Residual reactant against annealing time plateaus rather than decaying. Where the curve flattens is the bound, and that height scales with particle size.
Threshold
H1177. A volatile reactant imposes a minimum partial pressure below which the target phase cannot form at any temperature, set by volatilisation competing with product formation and calculable from the two equilibrium constants. A sealed ampoule with a reservoir at a controlled temperature supplies that pressure independently of the furnace. Yield should vanish at a reservoir temperature, not at a furnace temperature.
Reversal
H1178. Grinding a precursor mixture accelerates the reaction by refining particles until the energy dose is high enough that grinding itself creates an amorphous phase, which later crystallises into the wrong polymorph. The dose, measured as specific energy rather than as time, marks the turn. A milling series with the product phase identified shows the wrong polymorph only past it.
Count
H1179. One non-isothermal curve supports two independent kinetic parameters at most, because the temperature programme and the conversion trace together supply only two constraints, so the three-parameter models in common use are not identifiable from such data. Fitting simulated curves of known parameters and counting recoveries demonstrates it. Most fits fail to recover the third parameter at all.
Artefact mimics effect
H1180. Self-propagating combustion in a pressed pellet is sometimes thermal runaway in the die, and the recorded temperature profile cannot tell the two apart because both show a steep rise moving away from the ignition point. Wave velocity against pellet thickness separates them, since a true wave's velocity is thickness-independent while runaway's is not. Vary thickness at fixed composition and the artefact is the only thing that cares.
Scale argument
H1181. Once the product layer approaches the particle radius, parabolic kinetics give way to linear kinetics because the diffusion field stops being planar, and the crossover is estimated at a layer thickness near a tenth of the radius. A particle-size series with layer thickness measured by electron microscopy shows where the exponent changes. The change is geometry, not a change of mechanism.
Inversion of standard practice
H1182. Regrinding between furnace steps is standard, but for a reaction whose intermediate is metastable at room temperature the correct practice is to grind cold, because room-temperature regrinding decomposes what the furnace just made. Room-temperature and liquid-nitrogen regrinding compared step by step, with the intermediate's fraction measured between steps, shows the loss. It appears only in the warm route.
High pressure, molten salts and fluxes
Misattribution
H1183. Recovered high-pressure polymorphs are credited to pressure when the deviatoric stress of a solid medium is what stabilises them, the phase surviving because the matrix clamps it rather than because it is favoured at ambient. Repeating it hydrostatically at the same nominal pressure separates the two, since a hydrostatic medium cannot clamp. Whichever medium returns the polymorph identifies the stabiliser.
Hidden term
H1184. Molten-salt working temperatures carry the salt's absorbed water, which lowers the eutectic and shifts the useful window by tens of kelvin (estimate) while leaving no trace in the salt's specification. Karl Fischer titration before and after drying, paired with the same synthesis both ways, shows the shift. The window moves with the measured water content rather than with the nominal salt.
Identity
H1185. Flux growth and molten-salt synthesis are one process at two product sizes, both dissolving solute in a molten medium and precipitating on cooling, with the flux simply the salt in excess. At matched supersaturation and cooling rate the habits come out identical. Any difference should reduce to the cooling profile, which is checkable by imposing the same profile on both.
Bound
H1186. Pressure in a multi-anvil cell is capped by the anvil material's yield strength rather than by the press's force, so a larger press does not guarantee a higher pressure. An internal standard in two cells of different truncation measures where the cap sits. Pressure should saturate at the load where the anvil deforms, and the saturation should follow the anvil's strength.
Threshold
H1187. A pore resists closure until the applied pressure exceeds the capillary pressure holding it open, so a pressing schedule below that pressure cannot remove porosity at any dwell, however long, while one above it closes readily. Threshold pressure runs inversely with pore radius, a few megapascals for a micron pore in an oxide (estimate). Varying pore size at fixed pressure moves the transition, and the transition pressure tracks the inverse radius rather than the material's hardness.
Reversal
H1188. A molten salt accelerates a solid-state reaction by providing a transport medium until the salt fraction isolates the reactants in separate pockets, after which the rate falls because they never meet. Percolation sets where the rate turns over, near the salt's threshold fraction. Imaging the microstructure across a salt-fraction series shows the pockets forming at the drop in rate.
Selection effect
H1189. High-pressure phase diagrams are assembled from products that survive decompression, so they are biased toward phases with high kinetic barriers and against those that revert on unloading. In-situ diffraction at pressure, with no unloading step, is the only way to see the difference. Running both on one composition and counting the phases missing from the quench diagram measures the bias directly.
Unit trap
H1190. Nominal multi-anvil pressure is a calibration performed once and then trusted, while anvil wear drifts it, and a factor of two between nominal and actual is possible late in an anvil's life (estimate). An internal standard in every run, rather than once per campaign, exposes it. The discrepancy correlates with cumulative load history rather than with the sample.
Characterisation traps and cross-cutting
Misattribution
H1191. Transition temperatures from calorimetry are assigned to the sample when the pan's conductivity sets the onset, since latent heat is partly conducted away before the sensor sees it. Extrapolating to zero heating rate in two different pans is the standard correction, rarely done. The two pans should converge only after extrapolation, and the gap between them is the pan's contribution.
Derived quantity
H1192. A crystallite size refined from a nanocrystalline pattern is a derived quantity whose error bar is correlated with the instrument profile, and below about twenty nanometres that correlation can shift the refined size by tens of per cent (estimate). Refining the same data against a profile measured on a standard under identical conditions tests it. The refined size moves while the fit quality barely changes.
Identity
H1193. Amorphous and nanocrystalline states are told apart by diffraction, yet a fine enough nanocrystalline assemblage produces the same broad halo, so the two are one state seen at different coherence lengths. Heating the amorphous sample until the halo narrows without a discrete reflection appearing tests it. A halo that does not move in position indicates the same local order throughout.
Bound
H1194. Minority-phase detection by laboratory diffraction is bounded by the majority's background rather than by counting statistics, so longer counting stops helping once the background dominates. Counting a fixed mixture for increasing times and plotting signal against background shows where the plateau begins. The plateau belongs to the phase pair, not to the instrument alone, and it moves when the majority phase changes.
Threshold
H1195. A dopant does not enter the bulk below a critical loading because surface sites cost less energy than bulk sites, so the dopant segregates until the surface is saturated and only then substitutes. The threshold marks surface saturation, not the end of solubility. Depth profiling by XPS at several loadings puts the onset of substitution at a loading rather than at zero.
Path dependence
H1196. An oxidising anneal heals defects until surface oxidation outruns bulk diffusion, after which the same anneal creates a new defect population by oxidising the skin while the core stays reduced. The condition depends on the path taken to the anneal, not only on its endpoint. Two thicknesses tested at one temperature show it, since only the thin sample reverses.
Count
H1197. A quenched phase-diagram section has one independent tie line per two-phase field, so a study reporting several distinct two-phase regions between the same end members is more likely resolving kinetics than thermodynamics. Equilibrating one composition for two durations and checking that phase fractions stop changing tests it. Counting regions without counting tie lines overstates the diagram.
Artefact mimics effect
H1198. Negative expansion in a sintered body measured along one axis is often microcracking opening on cooling, which looks like contraction in that direction and expansion perpendicular to it. The artefact is anisotropic and disappears in a hot-pressed sample of the same composition. Measuring both directions on one specimen separates it from a genuine negative expansion.
Scale argument
H1199. Once the boundary's own width becomes a non-negligible fraction of the grain, at roughly ten to twenty nanometres, the boundary composition stops behaving as a monolayer and the material's effective stoichiometry shifts (estimate). Nanocrystalline and coarse samples of identical nominal composition then differ in lattice parameter by an amount that scales with the boundary width, measurable at the tenth-of-a-percent level in a careful scan (estimate).
Inversion of standard practice
H1200. Reporting a synthesis as phase pure on the strength of a laboratory pattern is the wrong claim for nanocrystalline products, where the minority phase often controls the property and sits below the detection limit. The inversion is to state the detection limit with the claim and to seed known amounts of a second phase as a check. The pattern changes only above the seeded fraction that equals the limit.
Forms used, counted
Across the hundred entries the labels fall as follows, counting each entry once: Bound (9); Identity (9); Misattribution (9); Threshold (9); Artefact mimics effect (8); Hidden term (8); Inversion of standard practice (8); Reversal (8); Scale argument (8); Selection effect (7); Count (5); Unit trap (5); Thermodynamic bookkeeping (2); Conservation argument (1); Derived quantity (1); Mass-balance argument (1); Path dependence (1); Rank argument (1). The count is printed so that a reader can see which forms were leaned on and which were used sparingly, rather than taking variety on trust.
Status
One hundred entries, each written to stand on its own and each with a test attached. None has been attempted. The point of the list is not that any particular one is right; it is that each is stated in a form that a single experiment could kill, which is the only property a hypothesis needs before it is worth an afternoon.
If one of these is worth doing first, it is H1138 — that a magnetron's rate drift is the tooling factor rather than the plasma, checkable by moving the monitor to the substrate plane and watching the drift follow target erosion instead of elapsed time.
Incoprea Research · hand-considered hypotheses 1101–1200 · 3 November 2026 · ~/Desktop/Nov3/