GC/MS consisted of a Clarus 500 GC (Perkin Elmer), equipped with a Zebron ZB-FFAP column (Phenomenex), and a quadrupole Clarus 500 MS (Perkin Elmer). The 15-m long column had an inner diameter of 320 μm. The elapsed time during measurement was 32 min. We applied a temperature gradient from 40 to 230°, increasing in a stepwise manner by 6° per minute. The spectra of the detected molecules were compared to the NIST database.
The gaseous samples (A3_1gas-, A3_2gas-, and A3_3gas-) were directly injected into the gas chromatograph. Hydrophobic substances were extracted from 1 ml liquid sample that included the oil-like layer in 1 ml of an 1:1:1-mixture of dethylether, n-pentane and cyclohexane (A3_1liquid+, A3_2liquid+, and A3_3liquid-).
By comparing the surface area of the peaks in the chromatogram, we determined the concentrations of substances in the liquid phase relative to a reference (Appendix: Tab. 6, 7, and 8). We found the solvent cyclohexane most suitable to use as a reference. The estimated concentrations of the different molecules detected in the broth reached up to the millimolar range.
The gaseous samples (A3_1gas-, A3_2gas-, and A3_3gas-) were directly injected into the gas chromatograph. Hydrophobic substances were extracted from 1 ml liquid sample that included the oil-like layer in 1 ml of an 1:1:1-mixture of dethylether, n-pentane and cyclohexane (A3_1liquid+, A3_2liquid+, and A3_3liquid-).
By comparing the surface area of the peaks in the chromatogram, we determined the concentrations of substances in the liquid phase relative to a reference (Appendix: Tab. 6, 7, and 8). We found the solvent cyclohexane most suitable to use as a reference. The estimated concentrations of the different molecules detected in the broth reached up to the millimolar range.