Jasmine Hertzog,1 Vincent Carré,1 Anthony Dufour,2 Frédéric Aubriet1

Semi-targeted analysis of complex matrices by ESI FT-ICR MS or how an experimental bias may be used as an analytical tool

Jasmine Hertzog,1 Vincent Carré,1 Anthony Dufour,2 Frédéric Aubriet1

1LCP-A2MC, FR 2843 Institut Jean Barriol de Chimie et Physique Moléculaires et Biomoléculaires, FR 3624 Réseau National de Spectrométrie de Masse FT-ICR à très haut champ, Université de Lorraine, ICPM, 1 boulevard Arago, 57078 Metz Cedex 03, France

2LRGP, CNRS, Université de Lorraine, ENSIC, 1, Rue Grandville, 54000 Nancy, France

Supporting information

Paragraph § 1 - Calibration and data post-acquisition treatment of FT ICR MS features

The mass spectrumwas first internally calibrated with a well-known peak list (CxHyOz compounds). A list of peaks, whose signal-to-noise ratio is greater than 4.5,was generated and transferred to the Composer software (Sierra Analytics, Modesto, CA) for calculation and assignment. The following assignment criteria were used: C1–100H1–100N0–4O0-20S0–1Cl0-3+, with a 3 ppm tolerance error and a double bound equivalent (DBE) ranging from -0.5 to 40 (a DBE of -0.5 corresponds to a saturated compound cationized by NH4+). Then, the recalibration of the mass spectrum was conducted with signals assigned with an error lower than 1 ppm, by considering the following equation

with m/zthe mass-to-charge ratio, m (amu) the mass, z the charge of the ion, f the measured frequency (Hz), and A and B the constraints.

For the non-assigned signals or the assigned signals with a mass error greater than 1 ppm, a manual assignment was conducted by using Omega 8 Elemental Composition software (Varian-IonSpec Inc.) with the previously reported search criteria.

The correct assignment of the peaks and the efficiency of the calibration were checked by the graphical representation of the mass error vs.m/z, represented below, and the value of the calculated RMSmass error, which was equal to 0.41 for the considered set of data.

Representation of the mass error vs. mass-to-charge ratio of the compounds assigned in the oak bio-oil with 3-chloroaniline by (+) ESI FT-ICR MS.

Figure S1: (+) ESI LIT MS of vanillin in (a) methanol, (b) methanol/NH4OH (1% v/v), (c) methanol/aniline (1% v/v) and, (d) methanol/3-chloroaniline (1% v/v).

Figure S2: Alternativeproposed mechanisms of [M+H]+ vanillin MSn fragmentation

Figure S3:13C NMR spectrum of the vanillin in MeOD.

Figure S4: 13C NMR spectrum of the vanillin-NH4OH imine in MeOD

Figure S5: 13C NMR spectrum of the vanillin-aniline Schiff base in MeOD

Figure S6:(+) ESI LIT MS of cinnamaldehyde in (a) methanol, (b) methanol/NH4OH (1% v/v), (c) methanol/aniline (1% v/v) and, (d) methanol/3-chloroaniline (1% v/v).

Figure S7: Proposed mechanisms of[M+H]+cinnamaldehydeMSn fragmentation.

Figure S8: Proposed mechanisms of[M+H]+cinnamaldehyde+NH4OH imineMSn fragmentation.

Figure S9: Proposed mechanisms of [M+H]+cinnamaldehyde+aniline Schiff baseMSn fragmentation. With X=H for aniline and X=Cl for 3-chloroaniline.

Figure S10:(+) ESI LIT MS of butyrophenone in (a) methanol, (b) methanol/NH4OH (1% v/v), (c) methanol/aniline (1% v/v) and, (d) methanol/3-chloroaniline (1% v/v).

Figure S11: Proposed mechanisms of [M+H]+butyrophenoneMSn fragmentation.

Figure S12: Proposed mechanisms of [M+H]+butyrophenone+NH4OH imineMSn fragmentation.

Figure S13: Proposed mechanisms of [M+H]+butyrophenone+aniline Schiff baseMSn fragmentation.

Figure S14:(+) ESI LIT MS of trihydroxyacetophenone in (a) methanol, (b) methanol/NH4OH (1% v/v), (c) methanol/aniline (1% v/v) and, (d) methanol/3-chloroaniline (1% v/v).

Figure S15: Proposed mechanisms of [M+H]+trihydroxyacetophenoneMSn fragmentation.

Figure S16: Proposed mechanisms of [M+H]+trihydroxyacetophenone+NH4OH imineMSn fragmentation.

Figure S17: Proposed mechanisms of [M+H]+trihydroxyacetophenone+aniline Schiff baseMSn fragmentation.

Figure S18: Venn diagram of the CxHyOz (blue), CxHyOz*(green), and CxHyOz** (red) compound formulae assigned by (+) ESI FT-ICR MS analysis of oak bio-oil.

Figure S19: Reaction mechanisms proposed for the successive addition of two molecules of 3-chloronaniline on formic acid to form a twice-derivatized compound.