Concept Evidence Coverage Report

Book-anchor note: The primary Atlas anchor is the Victor C. Li 2019 book. The foundational-paper claim-evidence matrix summarized here is the supporting / verification layer beneath the book anchor; it verifies and deepens the book-anchored Atlas claims rather than serving as the primary evidence base.

Derived from 07_visualization/foundational_papers_claim_evidence_matrix.csv (70 data rows, 13/13 foundational papers, 15 distinct Atlas nodes).

Source scope: only the 13 foundational papers preserved under 00_sources/foundational_papers/. Non-foundational seed / book / handbook evidence is not counted here.

1. Row counts by Atlas node

Total data rows: 70.

2. Strongest evidence-backed nodes

Nodes with the widest paper coverage across the 13-paper set:

• strain_hardening_criteria — 12 rows spanning 10 papers (Li & Leung 1992, Lin & Li 1997, Li Wang Wu 2001, Li et al. 2002, Wang & Li 2004, Kanda & Li 2006, Yang 2008, Zhou 2012, Li & Li 2012, Tosun-Felekoglu 2013, Lu 2018). The most cross-validated node in the atlas.
• fiber_bridging_law — 9 rows spanning 7 papers (Li & Leung 1992, Lin & Li 1997, Redon 2001, Kanda & Li 2006, Yang 2008, Li & Li 2012, Lu 2018). Anchored by explicit constitutive derivations in Li & Leung 1992, Lin & Li 1997, and Yang 2008.
• flaw_design — 8 rows spanning 5 papers (Wang & Li 2004, Li & Wang 2006, Li & Li 2012, Tosun-Felekoglu 2013, Lu 2018). Covers both the LEFM a_c derivation and its experimental verification via artificial-flaw tailoring and muCT.
• interface_properties — 6 rows spanning 5 papers (Lin & Li 1997, Redon 2001, Li Wang Wu 2001, Li et al. 2002, Lu 2018). Reasonable spread across the interface-tailoring lineage.
• micromechanics_based_design — 6 rows spanning 5 papers.
• direct_tensile_test — 6 rows spanning 6 papers.

3. Weak / under-evidenced nodes

Nodes appearing in the matrix but backed by only one foundational paper each:

• two_way_debonding — 1 row, only Yang 2008.
• matrix_micro_spalling — 1 row, only Yang 2008.
• cook_gordon_effect — 1 row, only Yang 2008.
• pseudo_strain_hardening_index — 1 row, only Kanda & Li 2006.

These four nodes are single-paper concepts inside the foundational set. Their claims are traceable to a specific author, but they lack independent cross-paper corroboration inside the current 13-paper scope.

Additionally:

• crack_width_distribution — only 2 rows (Li Wang Wu 2001 for <100 um saturated crack width; Yang 2008 for ~60 um). The experimental crack-width literature is much wider than this and needs at least one more foundational or supplementary source in the atlas.
• surface_treatment — 3 rows all inside the PVA-oiling lineage (Redon 2001, Li Wang Wu 2001, Li et al. 2002). No independent surface-treatment evidence outside the PVA-oil route.

4. Nodes carried by seed / book evidence but with little foundational-paper evidence

Concept files present in 02_concepts/ and 05_experiments/ that do not appear (or barely appear) in this 13-paper matrix, meaning they are currently supported mainly by seed notes, book/handbook citations, or general narrative rather than by the foundational-paper set:

• 05_experiments/crack_width_distribution.md — only 2 foundational rows (see above), yet the concept description implies broader empirical support.
• 02_concepts/strain_hardening_criteria.md — the foundational Li & Leung 1992 paper is cited implicitly through strain_hardening_criteria and fiber_bridging_law, but no matrix row lands directly on steady_state_cracking. Consider adding a dedicated row from Li & Leung 1992 (steady-state flat crack propagation) and Yang 2008 (steady-state crack width control).
• 02_concepts/matrix_fracture_toughness.md — implicit in Kanda & Li 2006 first-crack theory and Lu 2018 K_tip parameters but not carried as its own row.
• 02_concepts/scale_linkage.md — although Li & Wang 2006 explicitly frames the scale-linking pipeline, the matrix files that under micromechanics_based_design. A dedicated scale_linkage row is missing.
• Nodes not appearing in the matrix at all (subset of 02_concepts/): constitutive_modeling, phenomenological_constitutive_model, multiscale_constitutive_model, damage_tolerance, SIMSS, and every durability-oriented node (chloride_diffusivity, freeze_thaw, permeability, abrasion_wear, cover_spalling, cyclic_loading, fatigue_behavior, long_term_strain_capacity, photocatalytic_function, phase_change_materials, functional_additive_effects, impact_resistance, applications, building_infrastructure, embodied_carbon, life_cycle_analysis, electrical_impedance_tomography). These are outside the foundational-paper scope and remain seed/book-supported until later evidence passes.

5. Verification-status snapshot

• verified_from_source_note: 62 rows
• needs_figure_table_verification: 5 rows
• needs_page_verification: 3 rows
• verified_from_full_text: 0 rows
• needs_doi_or_bibliographic_verification: 0 rows (DOIs already recorded in source notes / paper cards; independent verification is tracked in foundational_paper_integration_report.md).

All rows are grounded in the source notes; none rely on invented figure/page/equation numbers. Where an exact figure or page is not directly transcribed in the source note, the row is flagged needs_figure_table_verification or needs_page_verification.

6. Recommendations for next evidence upgrade

Priority 1 — cross-corroborate single-paper concepts

• two_way_debonding, matrix_micro_spalling, cook_gordon_effect: locate follow-up foundational or supplementary papers (Yang et al. later work; Ranade et al.; Zhang & Li) that independently invoke these mechanisms and add corroborating rows.
• pseudo_strain_hardening_index: add a Li 2003 monograph row (book, non-foundational) and one later applied ECC paper that quotes the index.

Priority 2 — deepen thin experimental concepts

• crack_width_distribution: add rows from foundational papers already in scope where crack-width is measured but not yet extracted (Li & Wang 2006 Fig. 12 crack pattern, Li & Li 2012 dogbone crack pattern). Consider promoting Lepech & Li crack-width papers to the atlas.
• direct_tensile_test: extract dogbone geometry / loading-rate rows for the four papers not yet represented (Redon 2001 [pullout, not applicable], Li 2002, Wang & Li 2004, Kanda & Li 2006, Yang 2008).

Priority 3 — bibliographic verification

• Run a bibliographic verification pass (DOI, journal volume/issue, page range) for the five DOIs still flagged as open in foundational_paper_integration_report.md (Lin 1997, Redon 2001, Li 2002, Kanda 2006, Yang 2008). When resolved, flip corresponding rows to verified_from_full_text where full-text pagination is checked, and add a needs_doi_or_bibliographic_verification row if any DOI cannot be recovered.

Priority 4 — figure/page verification against the extracted PDF text

• The 5 rows tagged needs_figure_table_verification (Kanda & Li 2006 numeric thresholds 1.3 / 2.7; Li & Wang 2006 Fig. 12; Zhou 2012 alpha bar-chart values; Li & Li 2012 Fig. 12) and 3 rows tagged needs_page_verification (Li & Leung 1992 delta_cu, direct_tensile_test framing; Lin & Li 1997 fracture-energy expression) can be resolved by grepping the corresponding .txt under 00_sources/foundational_papers/extracted_text/ for the specific tokens ("Fig. 12", "Eq. (2)", numerical constants).

Priority 5 — Atlas-graph promotion

• After the above, promote strongly-corroborated rows into the concept nodes' "Foundational evidence" sections without editing the concept files in this pass. Track in foundational_paper_graph_edge_promotion_plan.md.