QE (Step 4b): Popa-Simil, "Nano-Hetero Structure for direct energy conversion" (NSTI-Nanotech 2007)

QE run 2026-10-04. Inputs read: _text/manifest.json, _textprep.log, the full extracted text (4 pages), digest/ (1 file), deep/ (1 file), facts.json, claims.md, sites.md, all 9 research/ notes, physicist.md, mathematician.md, contrarian.md, arbiter.md, primer.md, example.py. I also read three of the Investigator's downloaded files myself, using the text layer only (pdftotext): US20100061503A1 paras 0105-0150, the Klassen et al. chapter (InTech 2013, book pp. 269-273) and the sibling Nanotech 2007 paper 3021 in full. I rendered p.2 of the source PDF at 150 dpi to spot-check Eq.1 and Fig.2 (scratch images in /tmp/qe-popa). Tags: [read] source text, [read: page image p.N], [read: research/files/...], [memory] with confidence, [calc] my arithmetic.

Verdict: MINOR

Coverage is complete and the central derivation (the e V / W bound) holds. The primer is good and example.py runs. The defects can be fixed by updating facts.json, claims.md, arbiter.md and primer.md: - p.3 of the paper was covered by the digest only, and the digest missed the author's own heating and "run cold" rationale. - The supporting theory in the Klassen chapter was never brought into the reviews. - A cryogenic energy balance bears on claim 5 and on the author's own "energetic balance" condition. It adds a "Refuted by calculation" part to claim 5 (cryogenic net-power reactor). The other headline verdicts are unchanged. - There are a few accuracy slips and retention gaps. None of this changes claims 1, 2, 4 or 6, so Steps 2-4 do not need to be redone.

1. Coverage

The source folder holds only this PDF (checked with ls). Nothing was skipped. Coverage: no defect.

2. Shortcuts

• p.3-4 rest on the digest alone (defect, moderate). Claims 3 ("runs cold"), 4 (mV modules, converter cascade) and 5 (Pu superconductors, Table 1) all come from p.3-4. Only p.2 was deep-read. The digest of p.3 left out a whole paragraph, the author's physical rationale for claim 3 (see section 3, item 1). It also misreported the voxel sizes (section 4, item 1). For a 4-page paper the right deep read was the whole paper. The reviewers did read the full extracted text, which softens this defect, but none of them picked up the missed paragraph.
• Investigator files were used through their notes, not read. The reviewers relied on the research notes for the Klassen chapter, the sibling 3021 paper and the patent; only patent paras 0116-0121 were read directly. This shortcut cost one important item (section 3, item 2).
• Unread primaries are flagged honestly. These are Anufrienko 2006/2008, Yershov 2011-2012, TMS 2008, Cleantech 2007, Physics Procedia 2011 and the USPTO file wrappers. Every later file says so. This is acceptable, because the documents are not in hand. They are Investigator targets, not deep-read targets.

3. Missed material (from the source text, the digest and the Investigator's files)

1. The author's own mechanism for "run cold" (p.3), missed everywhere. The text reads [read: p.3]:

   "The analysis of the heating process in nuclear fuel driven to the conclusion that the fast moving fission products are inducing in the insulator Uranium Dioxide ceramics electron showers of few micron range. The moving electrons interact with the lattice heating it. To prevent heating it is necessary to cut the electron showers circuits and drive them through conductive materials in circuits where the total resistance to be significantly smaller. The volume of the harvesting voxel is given by the conductivities of the two conductive layers, which have to create an internal resistance much smaller than the previous one of the ceramics."

So the author's model is that fuel heating is dissipation by electron-shower currents in resistive UO2, and that giving those currents a low-resistance path removes it. It is the stated physical basis of claim 3 and of the voxel sizing. - It does not appear in the digest, facts.json, claims.md, any review, the arbiter or the primer. A grep for "dioxide", "few micron" and "total resistance" found nothing. - It needs a reviewer's weighing. In standard physics, electron energy goes to the lattice through electron-phonon coupling within picoseconds and nanometres of the track [memory, high]. Ohmic dissipation of a macroscopic circuit current is a different thing. - It is also internally inconsistent with p.2, which gives the electron free path as "several nm". 2. Klassen et al.'s efficiency theory (InTech 2013, book pp. 269-272), missed by the research note and so by every review. This is the strongest supporting theory in the record, and it speaks directly to the arbiter's e V / W bound [read: research/files/InTech2012-Current-Research-Nuclear-Reactor-Technology.pdf, text layer]: - Their eqs. (4)-(8) define η = W_ext / W_abs. They show that η "approaches unity" when emitted electrons spend all their kinetic energy climbing the emitter-collector potential: "the electrons enter the collector with zero kinetic energy, thus eliminating heat release losses". - They note the opposite regime: when the electrons' kinetic energy "only slightly exceeds the work function ... the efficiency of the system will be always low". - They list four loss causes: planar geometry, emitter layers too thick, insulators too thick (with backflow through a radiation-conductive insulator), and a mismatch of electron energy. They propose aerogel or vacuum as the insulator. - The 10^-5 % Yoshida result in this passage is from gamma-ray pumping. The reviews present it as a generic irradiation result.

This does not overturn the arbiter. The bound uses the paper's own <10 mV cell voltage, and Klassen's ideal case needs a self-bias near the electron kinetic energy: tens of eV up to the 1.65 keV T_max, which is tens of volts to kV across nanometres. That is far beyond the ~1 V/nm breakdown the primer cites. The real design conflict is therefore stated by a supporting group itself. The insulator must be thin so that electrons cross it, but thick so that it can hold the voltage that makes efficiency high. The primer's "What the bound does not cover" paragraph should present this theory and the trade-off with its source. 3. The author's "energetic balance" condition (p.3-4), and the cryogenic cost nobody computed. The text says that "the electric harvested energy to be smaller than the energy consumed to remove the thermal effects" [read: p.3]. As printed this describes a net energy consumer, so it is probably a slip for "larger"; that reading is mine. Only the digest and facts.json mention it. Claim 5 and the conclusion ("cryogenic nuclear reactors ... with huge powers" [read: p.4]) need operation below Tc ≈ 18.5 K. - Carnot work to lift 1 J of heat from 15-18 K to 300 K is 15.7-19 J [calc]. - Fission-product beta energy alone (about 8 of 200 MeV, 4% [memory, high]) is deposited inside fuel-bearing structures of types (a) and (b). At Carnot that costs 0.6-0.76 of the total fission power [calc]. Real cryocoolers reach a fraction of Carnot [memory, medium]. - All unconverted fragment energy adds to this. Under the arbiter's bound that is nearly all 84%, which costs about 13-16 times the fission power [calc].

So a cryogenic in-core superconducting harvester is a net energy consumer unless almost all non-fragment energy is deposited outside the cold zone and conversion is near-perfect. This is the paper's own condition, worked through. The contrarian raised the Carnot cost qualitatively; no file quantifies it. 4. The patent offers a 1 V alternative. Para 0116 reads "harvested at 10 mV and 2.67 nA or at 1 V and 26.7 pA" [read: research/files/US20100061503A1.pdf]. The mathematician and arbiter quote only the 10 mV case. - At 1 V the e V / W bound is 3.3% at W = 30 eV, and the example still needs about 30 times more electrons than ionization frees [calc]. - The 1 V case shows the inventor contemplated volt-level (series) operation. It belongs beside the 10 mV case. - Para 0125 also gives the minimum structure as "20-40 microns for alpha and fission products". That fits the deep read's finding that the 600 nm stack stops only 9% of the ion's energy. 5. Table 1 arithmetic, never checked [read: p.3; calc]: - "Absorption decrease" is the Pu share of the compound's absorption: 558/611.68 = 91%, 1017.3/1068.23 = 95%, 289.6/340.53 = 85%, 18.5/69.43 = 27%. The table therefore shows that the "superconductor" is itself dominated by Pu absorption; the Pu is fuel or poison, not a passive conductor. - The Pu-238 "Amount" (611.68) does not equal its row sum (558 + 37.18 + 13.75 = 608.93); the other three columns add up. - The Ga entry 13.75 b = 5 × 2.75 b. The text says Ga is 2.5 b. - The text says Ga absorbs "over ten times" more than O and C. The real ratio is about 10^3 or more (O about 0.0002 b, C about 0.0035 b) [memory, high]. 6. Other content in the paper that the arbiter and primer omit (peripheral but pertinent): - The three harvesting geometries (a) fissile outside, (b) mixed in, (c) alpha tiles with no fissile material [read: p.3]. Type (c) is the one closest to existing alpha batteries. - The proposed fabrication route by binary nanoparticle superlattice self-assembly (Talapin, ref. 13) [read: p.3]. - The paper's history refs: Willson 1955 beta capacitor, the Polansky NERI 1999 "recently retested", DARPA LEAPS [read: p.1]. 7. The sibling paper 3021 (QE read in full). Its direct-conversion section is one paragraph. The ">90%" is asserted ("may even operate in cryogenic conditions, giving by quantum effects efficiencies higher than 90%") with no derivation, figure or data [read: research/files/Nanotech2007v4-3021.pdf, p.682]. Its baseline "200-500 w/cm3" [p.680] agrees with TMS 2008 and with the project paper's 0.2-0.5 kW/cm3. The arbiter's open question 2 can be answered: nothing new stands behind the 90%. A deep read of 3021 is not needed.

4. Accuracy spot checks

5. Retention

Nothing pertinent was dropped because a reviewer judged it negatively. The arbiter and primer keep the patents, the LENR turn, the Russian results, the FEC lineage and the absence-of-evidence analysis. The following pertinent items from the Investigator's findings are missing from both the arbiter and the primer. Restore them as omissions: - The inventor's own claim of confirmation. The Physics Procedia 2011 abstract says "some of the ideas being confirmed by collateral experiments" [web: https://www.sciencedirect.com/science/article/pii/S1875389211005943, per research note]. It belongs in the "Tested?" column of claims 1-2 as the inventor's characterization, with the experiments still unidentified. - The Romanian continuity lead. Popa-Simil was affiliated with the Institute of Atomic Physics (Magurele, 1996-97, per ScholarGPS) and the Muntele recoil work at the U-120 cyclotron in Bucharest. This is the same institute complex as Ursu's experimental FEC work (1978-88). The Investigator marks it "not established". - Cleantech 2007 "breakdown voltage of few milivolts" and the TechConnect 2015 / Granthaalayah 2021 restatements (the arbiter mentions them only in passing; the primer leaves them out). - The US Army 2007 items (Natick, "Active shield"; "Nuclear protective materials") as possible contractor records. They appear in the arbiter but not in the primer. - Forward patent citations by others (US10699820B2, WO2011011504A1 Colorado School of Mines, FR3018946A1, CN106340684A, JP2022551664A). They show uptake, not proof. - The Google legal-removal notice ("removed 17 result(s)") on the Russian-language query. The Investigator recorded it as observed, cause unknown. The arbiter's absence-of-evidence item 4 says "nothing found for or against". It should mention this notice and state that such notices are routine and their cause is unknown.

6. Process

• Source tags: present and honest throughout. [calc] is used consistently, and second-hand items are labelled as such. The "single source" wording follows SOURCE HONESTY ("documented in one published second-hand source").
• Web checks:
• Physicist used 1 of 6; mathematician used 1. The arbiter used 3 and covered W, PWR power density and PuRhGa5.
• The contrarian ran no web checks. Its claim 5 and claim 6 reasoning uses [memory, medium] items (PuCoGa5 Tc decline; highest research-reactor flux "a few ×10^15"). The mathematician and arbiter later web-checked the Tc decline. The flux comparison remains [memory, medium] in the primer ("above the flux of the highest-flux research reactors"); the physicist gives HFIR as about 1e15. This is a minor process defect. A web check of HFIR peak thermal flux would close it.
• Verdict scale: applied consistently, and split verdicts are stated clearly. One point to fix: claim 5's "cryogenic reactor" part should gain a calculation-based label (section 3, item 3).
• Primer: follows the STEP 4 order 1-9 with answers. Symbols and acronyms are defined (SRIM, FEC, LENR, CIci, W, T_c). There are worked numbers, and the primer adds no new verdicts.
• example.py: runs cleanly with the venv. Its output matches the primer (0.033%-1.0%; 24 V / 2.88 V; shortfall factor 3000; 5.56e15 n/cm²/s; 3.6 days; 4.84 THz).
• Small slips:
• The arbiter's item 5 Josephson line is garbled ("K_J = 483.6 GHz/mV x ..."), though the result is right.
• The physicist's LWR baseline is garbled (already corrected by the arbiter).
• claims.md claim 1 says "can convert ... at over 80%". The paper says "may be". The arbiter noted this, but claims.md was not updated.

7. Remediation

Deep reads (listed below and in the returned list, most important first): 1. Klassen et al. chapter, PDF pp. 273-283 (book pp. 264-274), Opus. Question: transcribe eqs. (4)-(8) and the efficiency definition. Is there an ideal-efficiency condition, and what self-bias voltage does it imply for fission-fragment delta electrons across a nm insulator? What exactly were refs [13, 14] (Anufrienko) and [49-51] (Yershov) measuring: gamma, neutron or U-235, and what does "efficiency" mean there? 2. The project paper, p.3-4, Sonnet. Question: the UO2 heating and resistance paragraph, the voxel sizes, the "energetic balance" sentence, Table 1 arithmetic, the PuCoGa5/PuCoRh5 naming, and the three structure types. 3. US20100061503A1, PDF pp. 26-36, Sonnet. Question: all efficiency, voltage and current statements (paras 0075-0140), the 1 V/26.7 pA alternative, series-connection voltages, FIG. 8/10/15 methodology, and any reported measurement. 4. Stepanov et al. 2018, pp. 1-4, Sonnet. Question: the measured quantities, any efficiency defined, and what the cited earlier Chernov MDM work measured. 5. US20130121449A1, PDF pp. 1-54 (description), Sonnet. Question: any measurement, and the stated mechanism of "nuclear entanglement and quantum non-localities". 6. PopaSimilLroadmaptof.pdf, pp. 1-81, Sonnet. Question: any CIci measurement or prototype, and which "collateral experiments" the author cites as confirmation.

Investigator (next run, not deep reads): - Download Physics Procedia 20 (2011) 270-292 (open access). - Find the primary Anufrienko 2008 paper (Nano- i mikrosistemnaya tekhnika No. 8) and the Yershov 2011/2012 papers. - Check USPTO file wrappers for 12/384,074, 13/297,243 and 11/603,812. - Search NM SOS for LAVM LLC, and SBIR/DTIC for the 2007 Army items.

facts.json: - Correct the voxel sizes to "mm3 to cm3; up to dm3". - Add the p.3 UO2 heating and resistance rationale and the "energetic balance" sentence (quote it as printed and note the probable slip). - Add the Table 1 arithmetic (Pu share; the Pu-238 sum mismatch; Ga 2.75 vs 2.5 b). - Change "PuCoRh5" to "PuCoRh5 as named; Tc 8.7 K matches PuRhGa5".

claims.md: - Claim 1: "the abstract says efficiency 'may be higher than 80%'". - Claim 3: add the author's stated mechanism (electron-shower currents through low-resistance conductors instead of the resistive UO2). - Claim 5: add "in a cryogenic reactor".

arbiter.md, add a "QE revision" section: - Claim 3: weigh the UO2 resistance-heating mechanism; it is inconsistent with p.2's nm free path. Expected verdict unchanged (Not testable as stated / weak form Plausible, untested). - Claim 1: record the Klassen ideal-efficiency theory and the thin-versus-voltage trade-off. Verdicts unchanged. Record the patent's 1 V alternative (3.3% bound; still about 30× short of charge). - Claim 5: add the cryogenic balance (Carnot 15.7-19 J per J at 15-18 K; beta heat alone costs 0.6-0.76 of fission power). Proposed: "cryogenic in-core superconducting harvester as a net power source: Refuted by calculation unless nearly all non-fragment energy is deposited outside the cold zone". Close the Rh question; it applies to PuRhGa5 too. - Answer open question 2: the 3021 ">90%" has no derivation. - Fix the garbled Josephson line. - Absence-of-evidence item 4: add the Google legal-removal notice, routine and of unknown cause.

primer.md: - Section 5, "What the bound does not cover": add the Klassen ideal-efficiency condition and the thin-insulator/voltage trade-off, with a short worked number (for example, 100 eV electrons need about 100 V across about 1-10 nm, which is 10-100 V/nm against about 1 V/nm breakdown). - Section 6: claim 5 row with the cryogenic balance; add to claims 1-2 "the inventor states some ideas 'confirmed by collateral experiments' (unidentified)"; add the 1 V patent example. - Section 7, "Genuinely open": the net-energy question for cryogenic operation. - Retention list: the Romanian continuity lead, Cleantech 2007, the Army items, forward citations, and the three structure types. - Optional: add a cryogenic Carnot step to example.py.

Recheck

Rechecked 2026-10-04 against current facts.json, claims.md, arbiter.md and primer.md (grep and spot reads of each fix).

No open defects in the content. The remediation deep-read and Investigator lists in section 7 are optional follow-ups, not blockers. Claim verdicts are unchanged except the intended addition to claim 5.

Final: PASS