================================================================================ BLIND CIPHER SWEEP: CRYSTAL STRUCTURE SCORING Date: 2026-04-03 File scored: BLIND_SWEEP_PREDICTIONS_2026-04-03.json Reference data: CRC Handbook (97th ed), Pearson Crystal Data, ICSD, Wikipedia ================================================================================ SCORING CRITERIA: MATCH = predicted archetype matches known ambient-pressure structure exactly PARTIAL = coordination number correct but wrong variant/stacking MISS = wrong archetype entirely N/A = no stable crystal structure at STP, or too short-lived to measure NOTE: All structures refer to standard ambient conditions (298 K, 1 atm) unless the element is only solid at low T (in which case the standard solid-state structure is used). For elements that are gases/liquids at STP, the solid-state structure at low T is used as the comparison target where available. ================================================================================ ELEMENT-BY-ELEMENT SCORING ================================================================================ Z Sym Predicted Known Structure Score Notes --- ---- ---------------- --------------------- ------- -------------------------------- 1 H other/molecular HCP (molecular H2) PARTIAL Solid H2 forms HCP molecular crystal; "other/molecular" captures molecular nature but misses HCP packing 2 He FCC HCP PARTIAL He solidifies only under pressure; HCP at ~1.1 K, 26 atm. FCC has right CN=12 but wrong stacking 3 Li BCC BCC MATCH 4 Be HCP HCP MATCH 5 B other/icosahedral rhombohedral (alpha-B) MATCH alpha-B is icosahedral B12 clusters in rhombohedral lattice; "icosahedral" is correct local motif 6 C Diamond Diamond cubic MATCH Graphite more stable but diamond cubic is the archetype 7 N other/molecular cubic (alpha-N2) PARTIAL Solid N2 forms cubic Pa3 molecular crystal; "molecular" is correct but misses cubic packing 8 O other/molecular monoclinic (alpha-O2) PARTIAL Solid O2 is monoclinic molecular crystal; "molecular" correct, packing detail missed 9 F other/molecular monoclinic (alpha-F2) PARTIAL Solid F2 is monoclinic molecular crystal; same situation as O 10 Ne FCC FCC MATCH Solid Ne (24 K) is FCC 11 Na BCC BCC MATCH 12 Mg HCP HCP MATCH 13 Al FCC FCC MATCH 14 Si Diamond Diamond cubic MATCH 15 P other/molecular orthorhombic (black P) PARTIAL White P is molecular; black P (most stable) is orthorhombic layered; "molecular" is closest to white P 16 S other/molecular orthorhombic (S8) PARTIAL alpha-S is orthorhombic with S8 rings; molecular nature correct, packing missed 17 Cl other/molecular orthorhombic (Cl2) PARTIAL Solid Cl2 is orthorhombic molecular crystal 18 Ar FCC FCC MATCH Solid Ar (84 K) is FCC 19 K BCC BCC MATCH 20 Ca FCC FCC MATCH 21 Sc HCP HCP MATCH 22 Ti HCP HCP MATCH 23 V BCC BCC MATCH 24 Cr BCC BCC MATCH 25 Mn BCC complex cubic (A12) MISS alpha-Mn has 58-atom cubic unit cell (cI58, A12); not BCC 26 Fe BCC BCC MATCH 27 Co HCP HCP MATCH 28 Ni FCC FCC MATCH 29 Cu FCC FCC MATCH 30 Zn HCP HCP MATCH Distorted HCP (c/a=1.856 vs ideal 1.633) 31 Ga other/ortho orthorhombic (Cmca) MATCH Ga is orthorhombic with CN=7 (1+2+2+2); pred CN=8 is close 32 Ge Diamond Diamond cubic MATCH 33 As A7 A7 (rhombohedral) MATCH 34 Se other/chain trigonal (helical chain) MATCH Se forms helical chains in trigonal lattice; "chain" is correct motif 35 Br other/molecular orthorhombic (Br2) PARTIAL Solid Br2 is orthorhombic molecular; "molecular" correct 36 Kr FCC FCC MATCH Solid Kr (116 K) is FCC 37 Rb BCC BCC MATCH 38 Sr FCC FCC MATCH 39 Y HCP HCP MATCH 40 Zr HCP HCP MATCH 41 Nb BCC BCC MATCH 42 Mo BCC BCC MATCH 43 Tc BCC HCP MISS Tc is HCP; prediction was BCC 44 Ru BCC HCP MISS Ru is HCP; prediction was BCC 45 Rh FCC FCC MATCH 46 Pd FCC FCC MATCH 47 Ag FCC FCC MATCH 48 Cd HCP HCP MATCH 49 In FCC tetragonal (BCT, A6) MISS In is body-centered tetragonal (face-centered tetragonal distortion); not FCC. CN=12 is close but structure is distinct 50 Sn Diamond Diamond (alpha-Sn) MATCH alpha-Sn (grey tin, <13C) is diamond cubic; beta-Sn (white tin, room T) is BCT. Prediction matches alpha 51 Sb A7 A7 (rhombohedral) MATCH 52 Te other/chain trigonal (helical chain) MATCH Same structure type as Se; "chain" correct 53 I other/molecular orthorhombic (I2) PARTIAL Solid I2 is orthorhombic molecular; "molecular" correct 54 Xe FCC FCC MATCH Solid Xe (161 K) is FCC 55 Cs BCC BCC MATCH 56 Ba BCC BCC MATCH 57 La HCP DHCP (double-HCP) PARTIAL La is DHCP (Strukturbericht A3', ABAC stacking); CN=12 correct, stacking variant wrong 58 Ce HCP DHCP (double-HCP) PARTIAL gamma-Ce at RT is DHCP (some refs say FCC at RT); CN=12 correct 59 Pr HCP DHCP (double-HCP) PARTIAL Pr is DHCP; CN=12 correct 60 Nd HCP DHCP (double-HCP) PARTIAL Nd is DHCP; CN=12 correct 61 Pm HCP DHCP (double-HCP) PARTIAL Pm is DHCP; CN=12 correct 62 Sm A7 Sm-type (rhombohedral) MATCH Sm has unique 9R rhombohedral stacking (ABCBCACAB); A7-like rhombohedral classification is reasonable match 63 Eu BCC BCC MATCH 64 Gd HCP HCP MATCH 65 Tb HCP HCP MATCH 66 Dy HCP HCP MATCH 67 Ho HCP HCP MATCH 68 Er HCP HCP MATCH 69 Tm HCP HCP MATCH 70 Yb BCC FCC MISS Yb is FCC at ambient; BCC above ~270 K is a high-T phase but standard reference is FCC 71 Lu HCP HCP MATCH 72 Hf HCP HCP MATCH 73 Ta BCC BCC MATCH 74 W BCC BCC MATCH 75 Re HCP HCP MATCH 76 Os HCP HCP MATCH 77 Ir FCC FCC MATCH 78 Pt FCC FCC MATCH 79 Au FCC FCC MATCH 80 Hg HCP rhombohedral (alpha-Hg) MISS Hg solidifies at 234 K into rhombohedral (A10) structure, not HCP. CN=6 not 12 81 Tl HCP HCP MATCH 82 Pb FCC FCC MATCH 83 Bi A7 A7 (rhombohedral) MATCH 84 Po other/simple-cubic simple cubic (alpha-Po) MATCH Po is the only element with simple cubic structure at STP; nailed it 85 At other/metallic unknown N/A No bulk At has ever been produced; t1/2 ~8h max 86 Rn FCC FCC (predicted) N/A Rn is gas; too radioactive for bulk solid measurement. FCC expected by noble gas trend but unconfirmed experimentally 87 Fr BCC unknown N/A Most radioactive natural element; t1/2 = 22 min; no bulk solid ever produced 88 Ra BCC BCC MATCH Ra metal has been isolated; BCC structure confirmed 89 Ac HCP FCC MISS Ac is FCC (Fm3m); prediction was HCP 90 Th FCC FCC MATCH 91 Pa other/complex BCT (body-centered tet) PARTIAL Pa is BCT (I4/mmm); "complex" captures non-standard nature but misses specific structure 92 U other/complex orthorhombic (alpha-U) PARTIAL U has complex orthorhombic structure (Cmcm); "complex" is reasonable 93 Np other/complex orthorhombic (alpha-Np) PARTIAL Np has complex orthorhombic structure; "complex" captures this 94 Pu other/complex-6+ monoclinic (alpha-Pu) MATCH Pu has 6 allotropes; alpha-Pu is monoclinic with 16 atoms/cell; "complex-6+" correctly identifies extreme structural complexity 95 Am HCP DHCP (double-HCP) PARTIAL Am is DHCP; CN=12 correct, stacking variant wrong 96 Cm HCP DHCP (double-HCP) PARTIAL Cm is DHCP; CN=12 correct 97 Bk HCP DHCP (double-HCP) PARTIAL Bk is DHCP at RT; CN=12 correct 98 Cf HCP DHCP (double-HCP) PARTIAL Cf is DHCP; some refs say FCC at RT; CN=12 correct 99 Es HCP FCC MISS Es is FCC; prediction was HCP. Only microgram quantities measured 100 Fm HCP unknown N/A No bulk Fm; t1/2 = 100 days but only picogram quantities 101 Md HCP unknown N/A No bulk solid; t1/2 = 51.5 days; picogram quantities 102 No HCP unknown N/A No bulk solid; t1/2 = 58 min 103 Lr HCP unknown N/A No bulk solid; t1/2 = 27 s 104 Rf HCP unknown N/A Superheavy; t1/2 = 1.3 h max; no bulk solid 105 Db BCC unknown N/A Superheavy; t1/2 = 28 h max; no bulk solid 106 Sg BCC unknown N/A Superheavy; t1/2 = 14 s max; no bulk solid 107 Bh HCP unknown N/A Superheavy; no bulk solid 108 Hs HCP unknown N/A Superheavy; no bulk solid 109 Mt FCC unknown N/A Superheavy; no bulk solid 110 Ds FCC unknown N/A Superheavy; no bulk solid 111 Rg FCC unknown N/A Superheavy; no bulk solid 112 Cn HCP unknown N/A Superheavy; no bulk solid 113 Nh other unknown N/A Superheavy; no bulk solid 114 Fl other unknown N/A Superheavy; no bulk solid 115 Mc other unknown N/A Superheavy; no bulk solid 116 Lv other unknown N/A Superheavy; no bulk solid 117 Ts other unknown N/A Superheavy; no bulk solid 118 Og FCC unknown N/A Superheavy; no bulk solid. Predicted to possibly be a solid noble gas but unconfirmed ================================================================================ SUMMARY STATISTICS ================================================================================ Total elements: 118 Excluded (N/A): 22 (At, Rn, Fr, Fm-Lr, Rf-Og) Total scored: 96 MATCH: 66 / 96 = 68.8% PARTIAL: 22 / 96 = 22.9% MISS: 8 / 96 = 8.3% MATCH + PARTIAL: 88 / 96 = 91.7% ================================================================================ BREAKDOWN BY BLOCK ================================================================================ s-block (scored: H, He, Li, Be, Na, Mg, K, Ca, Rb, Sr, Cs, Ba, Ra = 13) H(P), He(P), Li(M), Be(M), Na(M), Mg(M), K(M), Ca(M), Rb(M), Sr(M), Cs(M), Ba(M), Ra(M) MATCH: 11 (Li, Be, Na, Mg, K, Ca, Rb, Sr, Cs, Ba, Ra) PARTIAL: 2 (H, He) MISS: 0 Score: 11/13 = 84.6% exact, 13/13 = 100.0% (M+P) p-block (scored: B, C, N, O, F, Ne, Al, Si, P, S, Cl, Ar, Ga, Ge, As, Se, Br, Kr, In, Sn, Sb, Te, I, Xe, Tl, Pb, Bi, Po = 28) B(M), C(M), N(P), O(P), F(P), Ne(M), Al(M), Si(M), P(P), S(P), Cl(P), Ar(M), Ga(M), Ge(M), As(M), Se(M), Br(P), Kr(M), In(MISS), Sn(M), Sb(M), Te(M), I(P), Xe(M), Tl(M), Pb(M), Bi(M), Po(M) MATCH: 19 (B, C, Ne, Al, Si, Ar, Ga, Ge, As, Se, Kr, Sn, Sb, Te, Xe, Tl, Pb, Bi, Po) PARTIAL: 8 (N, O, F, P, S, Cl, Br, I) MISS: 1 (In) Score: 19/28 = 67.9% exact, 27/28 = 96.4% (M+P) d-block (scored: Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, Re, Os, Ir, Pt, Au, Hg = 29) Sc(M), Ti(M), V(M), Cr(M), Mn(MISS), Fe(M), Co(M), Ni(M), Cu(M), Zn(M), Y(M), Zr(M), Nb(M), Mo(M), Tc(MISS), Ru(MISS), Rh(M), Pd(M), Ag(M), Cd(M), Hf(M), Ta(M), W(M), Re(M), Os(M), Ir(M), Pt(M), Au(M), Hg(MISS) MATCH: 25 PARTIAL: 0 MISS: 4 (Mn, Tc, Ru, Hg) Score: 25/29 = 86.2% exact, 25/29 = 86.2% (M+P) f-block (scored: La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Ac, Th, Pa, U, Np, Pu, Am, Cm, Bk, Cf, Es = 26) La(P), Ce(P), Pr(P), Nd(P), Pm(P), Sm(M), Eu(M), Gd(M), Tb(M), Dy(M), Ho(M), Er(M), Tm(M), Yb(MISS), Lu(M), Ac(MISS), Th(M), Pa(P), U(P), Np(P), Pu(M), Am(P), Cm(P), Bk(P), Cf(P), Es(MISS) MATCH: 11 (Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Lu, Th, Pu) PARTIAL: 12 (La, Ce, Pr, Nd, Pm, Pa, U, Np, Am, Cm, Bk, Cf) MISS: 3 (Yb, Ac, Es) Score: 11/26 = 42.3% exact, 23/26 = 88.5% (M+P) ================================================================================ BREAKDOWN BY PERIOD ================================================================================ Period 1 (H, He) -- 2 scored: H(P), He(P) 0 MATCH, 2 PARTIAL, 0 MISS Period 2 (Li-Ne) -- 8 scored: Li(M), Be(M), B(M), C(M), N(P), O(P), F(P), Ne(M) 5 MATCH, 3 PARTIAL, 0 MISS Period 3 (Na-Ar) -- 8 scored: Na(M), Mg(M), Al(M), Si(M), P(P), S(P), Cl(P), Ar(M) 5 MATCH, 3 PARTIAL, 0 MISS Period 4 (K-Kr) -- 18 scored: K(M), Ca(M), Sc(M), Ti(M), V(M), Cr(M), Mn(MISS), Fe(M), Co(M), Ni(M), Cu(M), Zn(M), Ga(M), Ge(M), As(M), Se(M), Br(P), Kr(M) 16 MATCH, 1 PARTIAL, 1 MISS Period 5 (Rb-Xe) -- 18 scored: Rb(M), Sr(M), Y(M), Zr(M), Nb(M), Mo(M), Tc(MISS), Ru(MISS), Rh(M), Pd(M), Ag(M), Cd(M), In(MISS), Sn(M), Sb(M), Te(M), I(P), Xe(M) 14 MATCH, 1 PARTIAL, 3 MISS NOTE: Period 5 has 3 of 8 total misses. The 4d mid-block BCC zone and In's BCT distortion are where the cipher most struggles. Period 6 (Cs-Rn) -- 30 scored (At, Rn excluded as N/A): Cs(M), Ba(M), La(P), Ce(P), Pr(P), Nd(P), Pm(P), Sm(M), Eu(M), Gd(M), Tb(M), Dy(M), Ho(M), Er(M), Tm(M), Yb(MISS), Lu(M), Hf(M), Ta(M), W(M), Re(M), Os(M), Ir(M), Pt(M), Au(M), Hg(MISS), Tl(M), Pb(M), Bi(M), Po(M) 23 MATCH, 5 PARTIAL, 2 MISS Period 7 (Fr-Og) -- 12 scored (Fr, Fm-Og excluded as N/A): Ra(M), Ac(MISS), Th(M), Pa(P), U(P), Np(P), Pu(M), Am(P), Cm(P), Bk(P), Cf(P), Es(MISS) 3 MATCH, 7 PARTIAL, 2 MISS ================================================================================ CLEAN PERIOD SUMMARY TABLE ================================================================================ Period Scored MATCH PARTIAL MISS Exact% M+P% ------ ------ ----- ------- ---- ------ ---- 1 2 0 2 0 0.0% 100.0% 2 8 5 3 0 62.5% 100.0% 3 8 5 3 0 62.5% 100.0% 4 18 16 1 1 88.9% 94.4% 5 18 14 1 3 77.8% 83.3% 6 30 23 5 2 76.7% 93.3% 7 12 3 7 2 25.0% 83.3% ------ ------ ----- ------- ---- ------ ---- TOTAL 96 66 22 8 68.8% 91.7% Cross-check: 66 + 22 + 8 = 96. Verified. ================================================================================ CLEAN BLOCK SUMMARY TABLE ================================================================================ Block Scored MATCH PARTIAL MISS Exact% M+P% ------ ------ ----- ------- ---- ------ ---- s-block 13 11 2 0 84.6% 100.0% p-block 28 19 8 1 67.9% 96.4% d-block 29 25 0 4 86.2% 86.2% f-block 26 11 12 3 42.3% 88.5% ------ ------ ----- ------- ---- ------ ---- TOTAL 96 66 22 8 68.8% 91.7% NOTE on f-block: The exact match rate is lower (42.3%) because many lanthanides/ actinides adopt DHCP (double-HCP) rather than standard HCP. The cipher predicted HCP for all of them, getting the CN=12 right but missing the ABAC vs ABAB stacking distinction. If DHCP counted as HCP-variant MATCH, f-block exact rate would be 80.8% (21/26) and overall would rise to 75.0% (72/96). ================================================================================ ALL MISSES (8 total) ================================================================================ Z Sym Predicted Known Analysis --- ---- ------------ --------------- ---------------------------------------- 25 Mn BCC complex cubic alpha-Mn is cI58 (A12), 58 atoms in cubic (A12, cI58) cell. Unique structure. Cipher's d-block BCC pattern breaks here. 43 Tc BCC HCP Tc is HCP. Cipher predicted BCC (following mid-d-block pattern from period 4). The 5d HCP-BCC-HCP-BCC pattern doesn't mirror 4d exactly. 44 Ru BCC HCP Ru is HCP. Same issue as Tc -- cipher mapped the 4d BCC zone too wide. Ru is group 8 like Fe(BCC)/Os(HCP). 49 In FCC BCT (A6) In is body-centered tetragonal (slightly distorted FCC, c/a=1.076). CN=12 in both. Very near-miss; this could be argued as PARTIAL but BCT is a distinct structure. 70 Yb BCC FCC Yb is FCC at ambient. Has BCC phase above ~270K but standard ref is FCC. Cipher assigned BCC (Eu-like divalent pattern). 80 Hg HCP rhombohedral alpha-Hg is rhombohedral (A10), similar (A10) to A7 but distinct. Not HCP. Anomalous element that breaks d-block trends. 89 Ac HCP FCC Ac is FCC (Fm3m). Cipher predicted HCP following early-actinide lanthanide homology, but Ac breaks that pattern. 99 Es HCP FCC Es is FCC based on limited measurements (microgram quantities). Cipher predicted HCP continuing the Cm-Bk-Cf trend. 50 Sn Diamond BCT (beta-Sn) NOTE: Scored as MATCH above because alpha-Sn IS diamond cubic. However, beta-Sn (metallic, RT stable above 13C) is BCT. This is an edge case that could be argued either way. Kept as MATCH since alpha-Sn is the ground-state thermodynamic phase. ================================================================================ DETAILED MISS ANALYSIS: PATTERNS ================================================================================ 1. MID d-BLOCK BCC OVERSHOOT (Mn, Tc, Ru): The cipher assigns BCC to the middle of the d-block following the V-Cr-Fe pattern. This works in period 4 and period 6 but fails in period 5 where Tc and Ru are HCP. Mn is a complete outlier with its unique A12 structure. Pattern: cipher assumes d-block symmetry that breaks for 4d elements. 2. RARE EARTH STACKING (Yb, Es): The cipher correctly identifies CN=12 metallic structures for f-block but cannot distinguish FCC from HCP in the divalent rare earths (Yb, Es). Both are FCC because of their anomalous +2 valence (half-filled or full f-shell). 3. ANOMALOUS STRUCTURES (Hg, In, Ac): Hg (rhombohedral), In (BCT), and Ac (FCC) are all elements that break their respective group trends. These represent genuine edge cases where local electronic effects override the geometric pattern. ================================================================================ EDGE CASES AND NOTES ================================================================================ - Sn (Z=50): Scored MATCH for alpha-Sn (diamond cubic). If scored against beta-Sn (BCT, the room-temperature metallic form), this would be a MISS. Conservative scoring would be MISS, which would make total 65/96 = 67.7%. - C (Z=6): Diamond is the predicted archetype. Graphite (hexagonal layered) is thermodynamically more stable at STP. Scored MATCH for diamond cubic since the prediction explicitly names "Diamond" archetype. - He (Z=2): Requires ~26 atm to solidify. HCP structure confirmed. Scored PARTIAL (FCC predicted, HCP actual, both CN=12). - Ce (Z=58): Complex phase diagram. gamma-Ce (DHCP) at RT, alpha-Ce (FCC) under pressure. Scored PARTIAL for HCP vs DHCP. - Sm (Z=62): Unique 9R rhombohedral stacking. Scored MATCH for A7 prediction because both are rhombohedral, though Sm-type and A7 are technically distinct Strukturbericht types. - Po (Z=84): Correctly predicted simple cubic -- impressive since Po is the ONLY element with this structure. ================================================================================ FINAL VERDICT ================================================================================ Strict Lenient (M+P) ------ ------------- Overall: 66/96 (68.8%) 88/96 (91.7%) If Sn counted as MISS: 65/96 (67.7%) 87/96 (90.6%) If DHCP counted as M: 72/96 (75.0%) 88/96 (91.7%) If both adjustments: 71/96 (74.0%) 87/96 (90.6%) The cipher correctly predicts the crystal structure archetype for ~69% of all elements exactly, and gets the coordination/topology family right for >91%. The 8 misses cluster in three identifiable failure modes: (1) Mid d-block BCC overshoot (3 misses: Mn, Tc, Ru) (2) Divalent f-element FCC/HCP confusion (2 misses: Yb, Es) (3) Anomalous/unique structures (3 misses: Hg, In, Ac) None of the misses are random -- they all occur at known electronic-structure anomaly points, which is itself data about where the geometric model needs refinement. ================================================================================ END OF SCORING ================================================================================