Charakterisierung der Orchideenduftstoffe im Gebiet von Reserva Biologica Guaitil S.A. Costa Rica

The orchid population of the extremely species-rich flora of Reserva Biologica Guaitil SA, Costa Rica was analysed and compared in terms of scent components. This state-registered reserve was established by the author in 2013 and represents a private-level ecology conser-vation project. The distribution pattern of terpenes and other fragrances has been charac-terised within 55 genera and therein 133 varied species and examined for similarities. The main task was the question whether a chemotaxonomy by Volatile Organic Compound (VOC) measurement is possible in the orchids in Reserva Biologica Guaitil S.A./Costa Rica. The challenges of the project were the sampling in the cloud forest of Guaitil and the instru-mental analysis in the laboratory. Since the flowers were not allowed to be removed from their parent plant to remain as intact as possible, sampling on site was unavoidable. It had to be made sure that no stress or mechanical strain was exerted on the flower. Monolithic Ma-terial Sorption Extraction (MMSE) was chosen to collect the volatile compounds. In this me-thod, an absorber is placed over the flower which is supposed to be sampled for a certain period, and the volatile substances then accumulate on the monolith. In order to sample the scent of miniature orchids, a scentor suitable for field sampling was developed in numerous trials. It had to meet several requirements: it had to be handy and easy to attach near the flower, the size of the headspace had to correspond to the flower size and the material had to be inert so that no volatile substances (softeners) would diffuse out. Technically, the flow-through-variant and the recirculating-air-variant could be used. Most samples were taken with the recirculating-air-method, as satisfactory results were achieved here even with the smallest specimens and the environmental VOCs could be excluded. Thus, terpene concentrations of less than 10 ppm per flower could be measured. The samples were collected in the field, the orchid documented and analysed in the labora-tory of Reference Analytics GmbH in Austria. The analysis in the laboratory was performed by thermal desorption on a GCMS TRACE 1310 GC base module, TRACE 1300 GC SSL injector, AI 1310 GC autoinjector for liquid injection TriPlus 500 headspace from Thermo Fischer. Subsequent identification of the mass spectra was done using the NIST database, and verification of the results was done by reference to the AI values from the Robert Adams database. Before the start of the field study, the me-thodology was optimised regarding stationary phase and sample collection and then quali-fied. The evaluation of the more than 3000 measured values was carried out using JMP in nume-rous diagrams. The data source includes the chemical structures, GPS data, time of collection, GC chromato-grams and MS spectra and the image material for each sample. The identity of the orchids studied was confirmed by Lancester Garden of the University of Costa Rica, Diego Bogarin. Within the 55 genera, similar terpene patterns were checked as well as for all 133 species found. Most of the VOCs found are monoterpenes, followed by sesquiterpenes and to a lesser ex-tent terpenoids, i.e., terpene structures with functional groups. The ubiquitous components of air (alkanes, aldehydes, ketones) occurring in outdoor measurements originate from the - 39 - microbial degradation of higher molecular weight plant waxes (C 28 - C 35) and were consi-dered in the evaluation. For the characterisation of the fragrances, the VOCs were tested for possible regularities within genera, furthermore the analysis was carried out within the same species at different times of the day and year. A further investigation looked at the VOCs of the same species at the same time at different locations. The final analysis examined the mono- and sesquiterpene release on two different Tico-glossum species over a period of 24 hours at 120-minute intervals. A new scentor was deve-loped to minimise the mechanical stress on the flower during the 12 sampling sessions. The measurements were made qualitatively, the determined counts were used for the evalua-tion. The result of these investigations showed a very variable release of VOCs, no regularities between orchid species and the released fragrance could be determined depending on the time of fragrance release. The 24-hour odour analysis showed that at each time of the day, different substances are released, therefore a one-time measurement of the odour profile is only a snapshot of that moment and not representative for the characterisation of the spe-cies. The analysis of the available data suggested that a chemotaxonomic classification of orchids based on their VOCs measured in single samples does not provide valid results. Fragrance secretion shows no consistency in terms of chemical composition over time.