Flavor Release - ACS Publications - American Chemical Society

---

Chapter 6

Time-Resolved Headspace Analysis by Proton-Transfer-Reaction Mass-Spectrometry 1

2

1

C. Yeretzian , A. Jordan , H . Brevard , and W. Lindinger 1

2

N e s t l é Research Center, Vers-Chez-les-Blanc, 1000 Lausanne 26, Switzerland Institut für Ionenphysik, Leopold-Franzens-Universität, Technikerstr. 25, 6020 Innsbruck, Austria

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

2

A recently developed technique, Proton-Transfer-Reaction MassSpectrometry (PTR-MS), is reviewed based on applications on coffee. PTR-MS is a sensitive and fast method for on-line trace gas analysis. It consists of a specially designed chemical ionization cell, where headspace gas is continuously introduced and volatile organic compounds ionized by proton-transfer from HO. Protonated compounds are then mass analyzed in a quadrupole mass filter. First a description of the method will be given, with emphasis on the ionization mechanism. We then discuss a series of experiments that allow mass spectral intensities to be related to chemical compounds. Finally, two applications on coffee are discussed. +

3

Food products all along the food chain, from raw materials to final products, continuously emit volatile organic compounds (VOCs). These volatiles released from and hence surrounding the food, form the HeadSpace (HS). While VOCs represent only a small mass-fraction of the food product, they are related to important properties of the product itself, such as flavor, age or shelf-life, geographic or genetic origin, history of processing, safety, packaging, the presence of micro-organisms or others. Headspace-Gas of Food Products: A Rich Source of Information: Due to the substantial information that can be gained from a detailed knowledge of the chemical

58

© 2000 American Chemical Society

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

59

composition of the HS, several analytical techniques have been developed that sample and analyze the HS of food. Nearly all of them are based on gas chromatographic separation (1-3). Yet, in spite of advances made over the last twenty years, gas-chromatography (GC) based approaches are limited when it comes to monitor, in real time, the temporal evolution of HS profiles. The Added Value of Time-Intensity Profiles: The benefits, to the food industry, of having means to monitor fast on-line changes in volatile compositions are numerous. Here, we will mention just three: (i) Many food-processing steps involve fast transformations of foods, such as roasting, drying or cooking, with a marked trend towards ever shorter processing-times. Real-time monitoring of HS-gas can help to better control these processes, (ii) Flavor generation in thermal, enzymatic or fermentation processes is intrinsically dynamic. Time-resolved HS techniques can help to better understand the underlying chemistry and optimize process flavors, (iii) In situations of real-life sensory experiences of foods, the temporal evolution of flavor profiles within the oral or nasal cavity is believed to be a crucial attribute for the perceived quality of foods. In order to address experimentally these dynamic problems, one needs fast on-line analytical techniques. Over the last few years, several approaches have been implemented which provide temporal information on volatile compositions. Two main directions have dominated (Figure 1). Electronic Nose: One direction headed towards the development of a great variety of gas sensors. If grouped, such sensor arrays give fingerprint type timeintensity profiles of products (4). These techniques have been colloquially termed "Electronic Noses" (EN), and are in a stage of rapid development. Direct inlet mass spectrometry: Another group of scientists has taken the approach to directly inject HS-gas into an M S . The critical element of this approach is the ionization, and a series of ionization modes have been investigated. If volatile compounds are (close to) pure or have been separated prior to M S analysis (e.g. by GC), fragmentation can provide valuable structural information. Here the most appropriate mode of ionization is electron impact (EI) ionization. This "hard" ionization mode ionizes gases that have high ionization potentials, such as 0 N or C 0 , but reveals little chemical information i f used on complex mixtures of VOCs. Such mixtures are often investigated by chemical ionization methods, using "soft" ionization via ion-molecule reactions, where little or ideally no fragmentation occurs. One CI approach, by now well known in the field of flavour release, is atmospheric pressure chemical ionization (APCI) (5). Another particularly powerful CI method is Proton-Transfer-Reaction Mass-Spectrometry (PTR-MS). Besides CI, lasers are also used for soft ionization. Depending on the photon energy, two types of ionization processes are known: either resonant multiphoton ionization with ultraviolet (UV) photons, to selectively ionize V O C s out of a complex mixture, or vacuum ultraviolet ( V U V ) laser ionization, to ionize with a single laser photon, again with high sensitivity but less selectivity. Application of U V ionization to coffee roasting is documented in a series of publications (6-9). In a recent paper, we have outlined the advantages and limitations of both soft ionization approaches (10). Here we will discuss PTR-MS and present applications on coffee. 2>

2

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

2

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

60

Figure 1: Analytical techniques for time-resolved HS analysis. EN can be used as a low cost process-monitoring device, where chemical information is not mandatory. ΕΙ-MS adds sensitivity, speed and some chemical information. Yet, due to the hardionization mode, most chemical information is lost. PTR-MS is a sensitive onedimensional method which provides characteristic HS profiles (detailed fingerprints) and chemical information. Finally, REMPI-TOFMS combines selective ionization and mass separation and hence represents a two-dimensional method.

Proton Transfer Reaction Mass Spectrometry PTR-MS was introduced in 1993 by Lindinger and co-workers (77). Since then it has been steadily improved and applied to a variety of subjects. Medical and nutritional applications by means of breath analysis allow monitoring of metabolic processes in the human body (12-19). Environmental applications include investigations of V O C emissions from decaying bio-matter (20,21), or diurnal variations of VOCs in the troposphere. Finally, PTR-MS has been shown to be an ideal tool to measure Henry's law constants and their dependence on temperature and matrix (22). Detailed discussions of technical aspects of PTR-MS have been published in a series of review papers (23-25). Here, only a brief description will be given. A schematic drawing of the apparatus is shown in Figure. 2. A n electrical discharge in 1.5 mbar H 0 within the ion source generates nearly exclusively H 0 (>98%), and no mass filter is needed to pre-select H 0 reactant ions. H 0 ions are driven by a small electric field, through an orifice, into the drift tube, while HS gas is introduced coaxially to the orifice into the drift tube. The HS gas consists mainly of air or inert gas that acts as a buffer in the drift tube, and VOCs are present only in trace quantities. Using H 0 as reactant ions has many advantages. 2

+

+

3

3

+

3

+

3

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

61

H 0 has a proton affinity (PA) of 166.5 kcal/mol. Hence, VOCs with a P A exceeding 166.5 kcal/mol become ionized by proton transfer from H 0 : 2

+

3

+

H0 3

+

+ VOC —VOC

·H

+ # acetaldehyde, 59-»acetone, 61->acetic acid, 69->furan, 73^butanal, isobutanal, 2-butanone, 87-»2,3butanedione, 2-methylbutanal, 3-methylbutanal, 97->furfural, dimethylfuran, 101-»2,3-pentanedione, 111-^5-methylfurfural. A t some masses, more than just one compound is significantly contributing to the mass intensity.

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

69

40

50

60

70

80

90

100 110 120

130 140 150 160 170 180 190 2 0 0

mass [amu]

Figure 8: HS profiles of coffee brew plotted on linear (top frame) and logarithmic (bottom) intensity scales. Each type of scaling has its merits and drawbacks. Linear plots are often found in the literature on flavor release. In contrast, single logarithmic plots facilitate a comparison of intensities which are spread over a large range.

The most unique feature of PTR-MS is its capacity to monitor time-dependent variations of HS profiles on a scale of seconds (dynamic data). This is particularly important when we need to understand or even control fast processes, as they often occur in food processing. As an example of a dynamic process, we have reconstituted 2 grams of soluble coffee with 90 ml of hot water (70°C) and monitored the timeintensity evolution of 50 masses with a time resolution of about 5 seconds. For reasons of clarity we show in Figure 9 only seven masses. The assignment of these masses is analogous to those in Figure 8. For the first 40 seconds the HS profile of the dry soluble coffee powder is monitored. Then, we slowly add 90 ml of hot water,

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

70 which takes 15 seconds, and continue to monitor the HS profile of the liquid coffee. Immediately when adding water we observe a fast increase of HS intensities, reflecting largely two phenomen: (i) the increase in concentration of dissolved volatiles due to dissolution of the soluble powder and (ii) the permanent renewal of the liquid-gas interface for mass transfer due to turbulences during pouring of water. After about 15 seconds all the hot water was added and the turbulences slowly disappeared, reducing and ultimatly ending the renewal of the interface. Experimentally, we observe a maximum concentration of VOCs at about 30 seconds followed by a strong decrease. This decrease reflects a depletion of volatiles at the interface and a release that was controlled by the diffusion of volatiles to the interface region. A detailed discussion of the physical processes behind the observed timeintensity evolution upon reconstituting of soluble coffee will be given elsewhere.

0

50

100

150

200

250

300

Time [seconds] Figure 9: Ion-intensity profiles of seven selected masses upon reconstitution of 2 gram of soluble coffee with 90 ml of water at 70°C.

Conclusions In this contribution, a detailed description of Proton-Transfer-Reaction MassSpectrometry (PTR-MS) was given. It has the particular feature to provide within seconds quantitative data on headspace profiles, without any workup of the sample. We then discussed a series of specific experiments which each provided important guidance towards the chemical assignment of PTR-MS mass peaks. This included

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

71

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

energy dependent break-up patterns of pure compounds, bracketing of proton affinities and liquid-gas partition coefficients. We have chosen to address this question on one of the most challenging systems, namely the headspace of coffee. We would like to add that PTR-MS has many more fields of application. PTRM S profiles, taken as fingerprints, can be used to e.g. assess authenticity, monitor deviations in production or classify raw materials and products. Ultimately, we believe that PTR-MS holds promise to become an important tool for process control.

Acknowledgements: We would like to thank A . Hansel for fruitful discussions. This project was supported by the "Fonds zur Fôrderung der wissenschaftlichen Forschung" under project Ρ 12022.

References 1.

Wampler, T. P. In Techniques for Analyzing Food Aromas; Marsili, R., Ed.; Marcel Dekker, Inc., New York, 1997; 27-58. 2. Flavor Science: Sensible Principles and Techniques; Acree, T. E.; Teranishi R., Eds.; A C S Professional Reference Book, III Series, American Chemical Society, Washington, D C , 1993. 3. Techniques for Analyzing Food Aromas; Marsili, R., Ed.; Marcel Dekker, Inc., New York, N Y , 1997. 4. Sensors and Sensory Systems for an Electronic Nose; Gardner, J. W.; Bartlett, P. N., Eds.; Kluwer Academic Publishers: Dordrecht 1992. 5. Linforth, R. S. T.; Ingham, Κ. E; Taylor, A . J. In Flavour Science: Recent Developments; Taylor, A. J.; Mottram, D. S., Eds.; Special Publication No. 197, Royal Society of Chemistry, 1996, Thomas Graham House, Cambridge. 6. Zimmermann, R.; Dorfner, R.; Kettrup, Α.; Yeretzian, C. 18 International Scientific Colloquim on Coffee; Helsinki, Finland (ASIC 18), August 2-6, 1999; in press. 7. Dorfner, R.; Zimmermann, R.; Yeretzian, C.; Kettrup, A . Proceedings of the 9 Resonance Ionization Spectroscopy Symposium, June 1998, Manchester, U K , American Institute of Physics (AlP)-Conference Proceedings Series, AIP-Press New York, 309-312 (1998). 8. Zimmermann, R.; Heger, H . J.; Dorfner, R.; Yeretzian, C.; Kettrup, Α.; Boesl, U. Proceedings of the 1 International Convention on Food Ingredients: New Technologies, 15-17 September, 1997, Cuneo (Italy), 343-350. 9. Zimmermann, R.; Heger, H . J.; Yeretzian,C.;Nagel, H.; Boesl, U. Rapid Comm. Mass Spectrom. 1996, 10, 1975-1979. 10. Dorfner, R.; Zimmermann, R.; Kettrup, Α.; Yeretzian, C.; Jordan, Α.; Lindinger, W. Lebensmittelchemie. 1999, 53, 32-34. 11. Lindinger, W; Hirber, J.; Paretzke, H . Int. J. Mass Spectrom. Ion Processes 1993, 129, 79. th

th

st

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.

Downloaded by GEORGETOWN UNIV on August 24, 2015 | http://pubs.acs.org Publication Date: September 7, 2000 | doi: 10.1021/bk-2000-0763.ch006

72 12. Warneke, C.; Kuczynski, J.; Hansel, Α.; Jordan, Α.; Vogel, W.; Lindinger, W. Int. J. Mass Spectrom. Ion Processes 1996, 154, 61. 13. Taucher, J.; Hansel, Α.; Jordan, Α.; Lindinger, W. J. Agric. Food Chem. 1996, 44, 3778. 14. Taucher, J.; Hansel, Α.; Jordan, Α.; Fall, R.; Futrell, J. H . ; Lindinger, W. Rapid Comm. Mass Spectrom. 1997, 11, 1230. 15. Taucher, J.; Lagg, Α.; Hansel, Α.; Vogel, W.; Lindinger, W. Alcohol Clin. Exp. Res. 1995, 19, 1147. 16. Taucher, J.; Jordan, Α.; Hansel, Α.; Lindinger, W. Alcohol Clin. Exp. Res. 1997, 21, 939. 17. Jordan, Α.; Hansel, Α.; Holzinger, R.; Lindinger, W. Int. J. Mass Spectrom. Ion Processes 1995, 148, L1-L3. 18. Hansel, Α.; Jordan, Α.; Holzinger, R.; Paretzke, P.; Vogel, W.; Lindinger, W. Int. J. Mass Spectrom. Ion Processes 1995, 149/150, 605-619. 19. Prazeller, P.; Karl, T.; Jordan, Α.; Holzinger, R.; Hansel, Α.; Lindinger, W. Int. J. Mass Spectrom. Ion Processes 1998, 178, L1. 20. Warneke, C.; Karl, T.; Judmaier, H . ; Hansel, Α.; Jordan, Α.; Lindinger, W.; Crutzen, P. J. Global Biogeochem. Cycles 1999, 13, 9-17. 21. Fall, R.; Karl, T.; Hansel, Α.; Jordan, Α.; Lindinger, W.; submitted. 22. Karl, T.; Yeretzian, C.; Jordan, Α.; Hansel, Α.; Lindinger, W. Anal. Chem. submitted. 23. Lindinger, W.; Hansel, Α.; Jordan, A . Int. J. Mass Spectrom. Ion Processes 1998, 173, 191-241. 24. Lindinger, W.; Hansel, Α.; Jordan, A . Chem. Soc. Review 1998, 27, 347-354. 25. Hansel, Α.; Jordan, Α.; Holzinger, R.; Prazeller, P.; Vogel, W.; Lindinger, W. Int. J. Mass Spectrom. Ion Processes 1995, 149/150, 609-619. 26. Gioumousis, G.; Stevenson, D. P. J. Chem. Phys. 1993, 29, 294. 27. Su, T.; Chesnavish, W. J. J. Chem. Phys. 1982, 76, 5182. 28. Nijssen, L. M.; Visscher, C. Α.; Maarse, H . ; Willemsens L. C. In Volatile Compounds in Food; 7 edition, 1996, TNO Nutrition and Food Research Institute. 29. NIST Standard Reference Database Number 69 - August 1997 Release. 30. Leroi, J.-C.; Masson, J.-C.; Renon, H.; Fabries, J.-F.; Sannier, H . Ind. Eng. Chem., Process Des. Dev. 1997, 16, 139. 31. Mackay, D . ; Shui, W.; Sutherland, R . Environ. Sci. Technol. 1979, 11, 333. th

In Flavor Release; Roberts, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2000.