Pierre Vanloot

Comparison between NIR, MIR, concatenated NIR and MIR analysis and hierarchical PLS model. Application to virgin olive oil analysis

This work investigates the potential use of simultaneously near-infrared (NIR) and mid-infrared (MIR) spectroscopies for the quantitative analysis of fatty acids and triacylglycerols and for identifying the Registered Designation of Origin (RDO) of extra virgin olive oils. The two spectral ranges were used separately using PLS and PLS-DA regressions. To combine both information, concatenated matrix was used at the first time, multiblock method using H-PLS models were constructed at the second time. The models were compared in terms of prediction errors. The results obtained with MIR spectroscopy are better than the ones obtained with NIR spectroscopy. The H-PLS methodology seems to be very interesting for quantitative analysis with the use of additional information in the NIR range, which is not present in the MIR one. For RDO identification by discriminant analysis, the use of multiblock method was less efficient.

Characterisation and authentication of A. senegal and A. seyal exudates by infrared spectroscopy and chemometrics

The authentication of Acacia gums samples requires usually the use of sophisticated and time consuming analytical techniques. There is a need for fast and simple analytical techniques for the objective of a quality control methodology. Commercial Acacia senegal and Acacia seyal gums present characteristic MIR spectra. Principal Component Analysis of the infrared spectra of gum exudates of trees allow to distinguish Acacia gums from another gum exudates (Combretum, Ghatti, Karaya, Tragacanth). Moreover, gums of A. senegal and A. seyal separate them and from other Acacia species (Acacia dealbata, Acacia karoo, Acacia nilotica, Acacia sieberiana). Chemometric treatments of A. senegal and A. seyal MIR spectra were assessed for the quantification of moisture content in Acacia gums, for the classification into the two species and for the adulteration detection and quantification. Results were quite satisfactory, the moisture content was estimated at 3.1%, adulteration was detected at 3.4% and quantified at 5.6%. The discrimination of the two species is done without any ambiguity.

Attempts to separate (―)-α-thujone, (+)-β-thujone epimers from camphor enantiomers by enantioselective HPLC with polarimetric detection

In a first step, 26 chiral stationary phases (CSPs) have been screened for the separation of (-)-α-thujone, (+)-β-thujone epimers and camphor enantiomers by LC. The separations were monitored by a polarimeter detector. None of these CSPs provided a noticeable resolution for camphor enantiomers. The three components of a test mixture were clearly baseline separated on Chiralpak AS-H, Chiralpak AZ-H and TCI-MBS (poly(N-alpha-(S)-methylbenzylmaleimide) coated on silica gel) in a mobile phase composed of hexane/2-PrOH (99:1 v/v). Interestingly, for a preparative application, the three CSPs produced different elution orders for the three constituents of the mixture. In a second step, it is shown that the use of online polarimetric detection constitutes an unprecedented method to reveal the occurrence and the relative content of thujone epimers and the chirality of the major camphor enantiomer in crude essential oils. A proof of concept is illustrated on crude essential oils from Rosmarinus tournefortii, Artemisia herba alba and A. arborescens, which grow in Morocco and have several traditional uses there. In a third step, pure (+)-β-thujone was quantitatively collected from A. arborescens crude oil by semi-preparative HPLC on Chiralpak AZ-H monitored by a polarimeter.

Analysis of the major chiral compounds of Artemisia herba-alba essential oils (EOs) using reconstructed vibrational circular dichroism (VCD) spectra: En route to a VCD chiral signature of EOs.

An unprecedented methodology was developed to simultaneously assign the relative percentages of the major chiral compounds and their prevailing enantiomeric form in crude essential oils (EOs). In a first step the infrared (IR) and vibrational circular dichroism (VCD) spectra of the crude essential oils were recorded and in a second step they were modelized as a linear weighted combination of the IR and VCD spectra of the individual spectra of pure enantiomer of the major chiral compounds present in the EOs. The VCD spectra of enantiomer of known enantiomeric excess shall be recorded if they are not yet available in a library of VCD spectra. For IR, the spectra of pure enantiomer or racemic mixture can be used. The full spectra modelizations were performed using a well known and powerful mathematical model (least square estimation: LSE) which resulted in a weighting of each contributing compound. For VCD modelization, the absolute value of each weighting represented the percentage of the associate compound while the attached sign addressed the correctness of the enantiomeric form used to build the model. As an example, a model built with the non-prevailing enantiomer will show a negative sign of the weighting value. For IR spectra modelization, the absolute value of each weighting represented the percentage of the compounds without of course accounting for the chirality of the prevailing enantiomers. Comparison of the weighting values issuing from IR and VCD spectra modelizations is a valuable source of information: if they are identical, the EOs are composed of nearly pure enantiomers, if they are different the chiral compounds of the EOs are not in an optically pure form. The method was applied on four samples of essential oil of Artemisia herba-alba in which the three major compounds namely (-)-α-thujone, (+)-β-thujone and (-)-camphor were found in different proportions as determined by GC-MS and chiral HPLC using polarimetric detector. In order to validate the methodology, the modelization of the VCD spectra was performed on purpose using the individual VCD spectra of (-)-α-thujone, (+)-β-thujone and (+)-camphor instead of (-)-camphor. During this work, the absolute configurations of (-)-α-thujone and (+)-β-thujone were confirmed by comparison of experimental and calculated VCD spectra as being (1S,4R,5R) and (1S,4S,5R) respectively.

Isolation of the major chiral compounds from Bubonium graveolens essential oil by HPLC and absolute configuration determination by VCD.

The chirality issues in the essential oils (EOs) of leaves and flowers from Bubonium graveolens were addressed by chiral high-performance liquid chromatography (HPLC) with polarimetric detection and vibrational circular dichroism (VCD). The chemical compositions of the crude oils of three samples were established by gas chromatography / mass spectrometry (GC/MS). The well-known cis-chrysanthenyl acetate (1), oxocyclonerolidol (2), and the recently disclosed cis-acetyloxychrysanthenyl acetate (3), the three major chiral compounds, were isolated by preparative HPLC. The naturally occurring oxocycloneroledol (2), mostly found in the leaf oil (49.4-55.6%), presents a (+) sign in the mobile phase during HPLC on a chiral stationary phase (CSP) with a Jasco polarimetric detection. The naturally occurring cis-chrysanthenyl acetate (1) and cis-acetyloxychrysanthenyl acetate (3), mostly found in the flower EO (35.9-74.9% and 10.0-34.3%, respectively), both present a (-) sign. HPLC on a CSP with polarimetric detection is an unprecedented approach to readily differentiate the flower and leaf EOs according to their chiral signature. The comparison of the experimental and calculated VCD spectra of pure isolated 1, 2, and 3 provided their absolute configuration as being (1S,5R,6S)-(-)-2,7,7-trimethylbicyclo[3.1.1]hept-2-en-6-yl acetate 1, (2R,6R)-(+)-6-ethenyl-2,6-dimethyl-2-(4-methylpent-3-en-1-yl)dihydro-2H-pyran-3(4H)-one) 2 and (1S,5R,6R,7S)-(-)-7-(acetyloxy)-2,6-dimethylbicyclo[3.1.1]hept-2-en-6-yl]methyl acetate 3. Compounds 1, 2, and 3 were already known in B. graveolens but this is the first report of the absolute configuration of (+)-2 and (-)-3. The VCD chiral signatures of the crude oils were also recorded.

Artemisia arborescens Essential Oil Composition, Enantiomeric Distribution, and Antimicrobial Activity from Different Wild Populations from the Mediterranean Area.

Aerial parts of Artemisia arborescens were collected from different sites of the Mediterranean area (southwestern Algeria and southern Italy) and the chemical composition of their essential oil (EO) extracted by hydrodistillation was studied by both gas chromatography (GC) equipped with an enantioselective capillary column and GC/mass spectrometry (GC/MS). The EOs obtained were tested against several Listeria monocytogenes strains. Using GC and GC/MS, 41 compounds were identified, accounting for 96.0 – 98.8% of the total EO. All EOs showed a similar terpene profile, which was rich in chamazulene, β‐thujone, and camphor. However, the concentration of such compounds varied among the EOs. A. arborescens EO inhibited up to 83.3% of the L. monocytogenes strains, but the inhibitory spectrum varied among the EOs, with those from Algeria showing a higher inhibition degree than the Italian EOs. Such effect likely depended on the ketone (β‐thujone + camphor) content of the EO. The differences in the EO composition support the hypothesis that A. arborescens has at least two different chemotypes: a β‐thujone and a chamazulene type. The EO inhibitory spectrum indicates the A. arborescens EO as a valuable option in the control of the food‐borne pathogens.

Modular assessment of the performance of embankment dams

The performance assessment of embankment dams in respect to main deterioration mechanisms is a major challenge for managers in charge of their security. One model developed knowledge-based models for assessing the performance of embankment dams. They have three hierarchical levels: performance of the analysed component (embankment, foundations), performance of the functions and status indicators. However, while these dams are frequently located upstream elements at risk, small dams are rarely equipped with drains and monitoring instruments, and woody vegetation is often present. These features were partially or not dealt with in the previous models. To cope with them, we propose to firstly, identify the specific status indicators linked with woody vegetation and secondly, adapt and complete the existing models. A modular approach is notably proposed in order to evaluate different types of dam. The approach was applied to a group of seven French and Italian dams and showed its relevance.