reference

Assumptions, Validity and Limitations

A formal enumeration of the core assumptions behind the P(t) framework, each with its validity condition and its breakdown or limiting case, closing with a domain-of-validity summary.

Last updated: 2026-07-26

Every quantitative framework rests on assumptions, and its authority is bounded by them. The Schubart Master Formula is a general accounting identity for continuous ambient energy conversion in open systems — a bookkeeping structure, not a device model — and its structure carries a definite set of premises. This chapter states each premise explicitly, gives the condition under which it holds, and identifies the limiting case in which it breaks down. The purpose is not to defend the framework but to fence it: to mark clearly where the identity applies and where a user must not trust it. The canonical object under discussion is

P(t) = \iint \Phi(E,\theta,t)\,\sigma(E)\,\Delta p\,v_{\text{ph}}\,\eta_{\text{conv}}\;dE\,d\theta ,

whose factor-by-factor construction is set out in the mathematical architecture. What follows treats that construction as given and interrogates the conditions that make it meaningful.

Throughout, we distinguish two categories of statement. Some assumptions are structural — they define what the formula means and cannot be relaxed without writing a different formula. Others are modelling assumptions — the model assumes them for tractability, and they can in principle be relaxed at the cost of added complexity. We flag which is which, because the two fail in very different ways: a structural assumption failing means the identity does not apply at all, whereas a modelling assumption failing means a correction term is needed.

Assumption A — linearity and independence of the five factors

Statement. The integrand is a product of five factors — flux density, cross-section, momentum transfer, phonon velocity, and conversion efficiency — each entering to the first power. This multiplicative form assumes the factors are statistically independent and that none of them saturates or depends on the others within the integration window.

Validity condition. The product form is valid when the five factors are separately determined by distinct physical mechanisms, so that fixing one does not shift the distribution of another, and when each factor operates in a regime linear in its own argument. Formally, the joint expectation must factorise, $\langle \Phi\,\sigma\,\Delta p\,v_{\text{ph}}\,\eta\rangle \approx \langle\Phi\rangle\langle\sigma\rangle\langle\Delta p\rangle\langle v_{\text{ph}}\rangle\langle\eta\rangle$ , which holds when the factors are uncorrelated.

Limiting case. The form breaks down where any factor saturates or couples to another. If $\eta_{\text{conv}}$ falls as the incident flux $\Phi$ rises — a saturation of the converting medium — then $\Phi$ and $\eta$ are anti-correlated and their product overstates the output; a cross term $\langle\Phi\,\eta\rangle - \langle\Phi\rangle\langle\eta\rangle$ must be subtracted. Likewise, if $\sigma(E)$ itself depends on the incident intensity (a nonlinear cross-section), the first-power assumption fails. This is a modelling assumption, and its failure is signalled by non-vanishing covariances, discussed quantitatively in the uncertainty analysis.

Assumption B — channel additivity for multi-channel flux

Statement. When several distinct energy channels contribute to the ambient drive, the model assumes their fluxes add: the effective flux is a sum, $\Phi_{\text{eff}} = \sum_c \Phi_c$ , with each channel entering the same integrand structure.

Validity condition. Additivity holds when the channels are incoherent — when they carry independent phases and their contributions to the converting medium do not interfere. This is the ordinary situation for statistically independent ambient sources, whose intensities (not amplitudes) add.

Limiting case. Additivity fails when channels are mutually coherent. Coherent channels add at the amplitude level, producing interference cross-terms $2\,\mathrm{Re}\langle A_c A_{c'}^{*}\rangle$ that can enhance or suppress the total beyond a simple sum — the regime treated in coherent enhancement. In that regime $\Phi_{\text{eff}} = \sum_c\Phi_c$ is replaced by a full coherent superposition, and channel additivity is no longer a valid reduction. This too is a modelling assumption.

Assumption C — effective, ensemble-averaged parameters

Statement. The symbols $\sigma$ , $\Delta p$ , $v_{\text{ph}}$ , and $\eta_{\text{conv}}$ are taken to be effective quantities — spectrum-averaged or ensemble-averaged representatives of underlying distributions — rather than microscopically resolved functions.

Validity condition. An effective average is representative when the averaging window is wide enough to sample the distribution fully but narrow enough that the distribution does not shift across the window. That is, the parameter’s distribution must be approximately stationary over the range of $E$ and $\theta$ contributing to the integral, so that a single averaged value stands in for the whole.

Limiting case. The reduction fails when a parameter varies sharply within the window — for example a cross-section with a narrow resonance in $E$ . Then the average $\langle\sigma\rangle$ misrepresents the integral $\int\sigma(E)\,\mathrm{d}E$ , because the resonance contributes disproportionately and the mean value does not capture it. In such cases the effective parameter must be replaced by its resolved profile, and the representative-slice reduction described in the methods chapter is invalid. Whether the window is representative is an empirical question, not something the formula can guarantee.

Assumption D — steady external drive and open-system exchange

Statement. The system is open and driven: it exchanges energy with an external ambient reservoir that maintains it away from thermal equilibrium. A non-zero net momentum transfer $\Delta p$ is available because the drive is asymmetric.

Validity condition. This holds away from equilibrium, where a persistent gradient or flux breaks the symmetry between forward and reverse processes, as developed in open thermodynamic systems. The net momentum transfer $\Delta p$ is the symmetry-breaking factor; it is finite only when detailed balance is violated by the external drive.

Limiting case. At detailed balance — a closed system relaxed to equilibrium, or a driven system whose forward and reverse rates have equalised — the net momentum transfer vanishes, $\Delta p \to 0$ , and with it $P(t) \to 0$ . This is the most important limiting case in the whole framework: it enforces the Second Law by construction. No arrangement of the other four factors can produce net continuous output at equilibrium, because the factor that would multiply them is identically zero. This is a structural assumption: an equilibrium system is simply outside the framework’s domain.

Assumption E — boundedness of the conversion efficiency

Statement. The conversion efficiency is bounded, $0 \le \eta_{\text{conv}} \le 1$ .

Validity condition. Boundedness is a thermodynamic admissibility condition, valid always: $\eta_{\text{conv}}$ is the fraction of intercepted power actually converted, and a fraction cannot exceed unity or fall below zero. Any evaluation that returns $\eta_{\text{conv}} > 1$ has mis-defined the factor — typically by omitting a normalising reference power — and must be rejected.

Limiting case. The boundary $\eta_{\text{conv}} = 1$ is the reversible, loss-free limit; it is an upper bound never attained in practice, and treating it as achievable is a modelling error, not a limiting case of the physics. The lower boundary $\eta_{\text{conv}} = 0$ recovers the trivial no-conversion case. Because $\eta_{\text{conv}}$ multiplies the whole integrand, it can never rescue an output that the other factors set to zero, but it can and does cap the maximum. This is a structural constraint that guarantees thermodynamic admissibility of any number the formula returns.

Assumption F — separability of the (E, θ) integral

Statement. The double integral over energy and angle is treated, where reduced, as separable: $\iint g(E,\theta)\,\mathrm{d}E\,\mathrm{d}\theta \approx \big(\int g_E\,\mathrm{d}E\big)\big(\int g_\theta\,\mathrm{d}\theta\big)$ , so that spectral and angular structure factorise.

Validity condition. Separability holds when the integrand’s energy dependence and angular dependence are independent — when the flux and cross-section factorise as $\Phi(E,\theta) \approx \Phi_E(E)\,\Phi_\theta(\theta)$ over the domain of interest, as discussed in the spectral-angular structure.

Limiting case. Separability fails when energy and angle are correlated — for instance when the incident spectrum hardens toward grazing angles, so that the peak of $\Phi_E$ shifts with $\theta$ . Then the integral does not factor, and the product of one-dimensional integrals misestimates the true two-dimensional integral. In that case the full double integral must be evaluated numerically without factorising, and the representative-slice shortcut does not apply. This is a modelling assumption adopted for tractability.

Domain of validity — summary

Collecting these, the Schubart Master Formula is on solid ground when, and only when, the following hold together: the five factors are independent and unsaturated (A); contributing channels are incoherent so their fluxes add (B); the effective parameters are averaged over representative windows (C); the system is open and driven away from equilibrium, so that $\Delta p \neq 0$ (D); the efficiency respects $0\le\eta_{\text{conv}}\le1$ (E); and the spectral and angular dependences are separable or else integrated jointly (F). Of these, (D) and (E) are structural and non-negotiable — they place the framework firmly on the correct side of the Second Law — while (A), (B), (C), and (F) are modelling assumptions that can each be relaxed at the price of an explicit correction term.

The honest reading is therefore this: the formula is an exact accounting identity under its structural assumptions and a controlled approximation under its modelling assumptions. Outside these conditions it does not merely lose accuracy; it stops being the right object. The open questions about which regimes real ambient systems occupy — and how to measure the symmetry-breaking $\Delta p$ that assumption (D) requires — are collected in the scientific questions, and the primary literature grounding these premises is gathered in the references.