Ask the evidence.

1. Question

   What happens to a PMMA flame when airflow changes in microgravity?

2. Retrieved NASA evidence chosen by deterministic retrieval, before any model runs
   • BASS-II test B16test record · microgravity
   • BASS-II test B20test record · microgravity
   • BASS-II test B19test record · microgravity
   20 items were retrieved in total; 3 are cited below.
3. AI synthesis Saved OpenAI synthesis: MicroFire-Eval q036, 2026-10-03. The displayed summary is preserved from that run and rechecked as a cited claim. Not a live request.

   All three BASS-II runs used 2-cm, 100-µm-thick PMMA in opposed flow at about 16.4–16.5% O2 aboard the ISS. Outcomes differed: B16 quenched, B20 burned the entire sample, and B19 blew off; the reported fan “pot” settings were different across runs.

4. Claim extraction and verification each factual claim is checked separately

   1. observed✓ passed every check

      All three BASS-II runs used 2-cm, 100-µm-thick PMMA in opposed flow at about 16.4–16.5% O2 aboard the ISS. Outcomes differed: B16 quenched, B20 burned the entire sample, and B19 blew off; the reported fan “pot” settings were different across runs.

      • ✓Every citation exists in the retrieved evidence
      • ✓Every number appears in the cited NASA records
      • ✓Units match the cited evidence
      • ✓No microgravity result told as Moon or Mars
      • ✓No causal or safety claim the evidence does not make
      • ✓No past outcome turned into a prediction
      Show the evidence behind this claim

      BASS-II test B16BASS-II test B16; material 2-cm 100-µm-thick PMMA; opposed flow; oxygen 16.5 %; airflow as recorded "3, 5 pot, 0.4 pot" (cm/s; "pot" values are fan settings); thickness 0.1 mm; width 20 mm; outcome: Quenched as flow was reduced (coded by MicroFire Atlas from NASA's notes); NASA notes: "HCl 1.1 > 1.3; flame quenched after prolonged burn at 0.4 pot, one-sided flame; spread to top side and quenched immediately; did igniter wire affect top flame?"; source: bass2-summary Table A.1, PDF page 111; run in microgravity aboard the ISS

      BASS-II test B20BASS-II test B20; material 2-cm, 100-µm-thick PMMA; opposed flow; oxygen 16.5 %; airflow as recorded "5, 3, pot 0.5" (cm/s; "pot" values are fan settings); thickness 0.1 mm; width 20 mm; outcome: Burned the entire sample (coded by MicroFire Atlas from NASA's notes); NASA notes: "HCl 0.9 > 1.1; flame burned entire sample"; source: bass2-summary Table A.1, PDF page 112; run in microgravity aboard the ISS

      BASS-II test B19BASS-II test B19; material 2-cm, 100-µm-thick PMMA; opposed flow; oxygen 16.4 %; airflow as recorded "10, blowoff" (cm/s; "pot" values are fan settings); thickness 0.1 mm; width 20 mm; outcome: Blew off at high flow (coded by MicroFire Atlas from NASA's notes); NASA notes: "HCl 1.3 > 1.4; flame blew out at a pot of 6.0"; source: bass2-summary Table A.1, PDF page 112; run in microgravity aboard the ISS

   2. derived✓ passed every check

      The three tests used the same material dimensions and nearly identical oxygen levels (16.4–16.5 % O2).

      • ✓Every citation exists in the retrieved evidence
      • ✓Every number appears in the cited NASA records
      • ✓Units match the cited evidence
      • ✓No microgravity result told as Moon or Mars
      • ✓No causal or safety claim the evidence does not make
      • ✓No past outcome turned into a prediction
      Show the evidence behind this claim

      BASS-II test B16BASS-II test B16; material 2-cm 100-µm-thick PMMA; opposed flow; oxygen 16.5 %; airflow as recorded "3, 5 pot, 0.4 pot" (cm/s; "pot" values are fan settings); thickness 0.1 mm; width 20 mm; outcome: Quenched as flow was reduced (coded by MicroFire Atlas from NASA's notes); NASA notes: "HCl 1.1 > 1.3; flame quenched after prolonged burn at 0.4 pot, one-sided flame; spread to top side and quenched immediately; did igniter wire affect top flame?"; source: bass2-summary Table A.1, PDF page 111; run in microgravity aboard the ISS

      BASS-II test B19BASS-II test B19; material 2-cm, 100-µm-thick PMMA; opposed flow; oxygen 16.4 %; airflow as recorded "10, blowoff" (cm/s; "pot" values are fan settings); thickness 0.1 mm; width 20 mm; outcome: Blew off at high flow (coded by MicroFire Atlas from NASA's notes); NASA notes: "HCl 1.3 > 1.4; flame blew out at a pot of 6.0"; source: bass2-summary Table A.1, PDF page 112; run in microgravity aboard the ISS

      BASS-II test B20BASS-II test B20; material 2-cm, 100-µm-thick PMMA; opposed flow; oxygen 16.5 %; airflow as recorded "5, 3, pot 0.5" (cm/s; "pot" values are fan settings); thickness 0.1 mm; width 20 mm; outcome: Burned the entire sample (coded by MicroFire Atlas from NASA's notes); NASA notes: "HCl 0.9 > 1.1; flame burned entire sample"; source: bass2-summary Table A.1, PDF page 112; run in microgravity aboard the ISS

5. Final bounded answer

   Shown, because every claim passed. These checks test citations, numbers, units, gravity context and wording. They do not prove that every sentence is scientifically entailed by the evidence.

Near-matched runs show an association; they do not isolate airflow as the cause. Fan settings are not calibrated velocities.