Bread volatiles, identification

Sampling and Analysis. A frozen slice of bread was cut in pieces and stacked in an enlarged sample flask of an aroma isolation apparatus according to MacLeod and Ames (74). Volatile compounds were trapped on Tenax TA and afterwards thermally desorbed and cold trap injected in a Carlo Erba GC 6000 vega equipped with a Supelcowax 10 capillary column (60 m x 0.25 mm i.d.) and a flame ionisation detector. Similar GC conditions were used for GC-MS identification of volatile compounds by dr. M.A. Posthumus (Dept. Organic Chemistry, VG MM7070F mass spectrometer at 70 eV El, 75). 

The first comprehensive investigation of the volatiles of wheat bread was carried out by Mulders et al. (11. 16-18). After a qualitative analysis (16-18) which led to the identification of 90 compounds, the authors attempted to get an insight into the sensory relevance of the major components which were found in the headspace extract of white bread. A mixture of the main compounds identified was prepared in water to match the gas chromatogram obtained from the bread sample (11,). The odor of the synthetic mixture resembled that of the fermented dough but not that of wheat bread. 

Sizer at al. (20) observed that the compounds causing the pleasant odor which resembled bread crust occur in the basic volatile fraction of white bread. Identification experiments yielded the five pyrazines listed in Table II. A comparison of the odor threshold of each pyrazine in water to its concentration found in bread indicated that 2-ethyl-3-methylpyrazine and 2-methyl-6-propylpyrazine were present at concentrations above their odor thresholds. The authors described the odors of these two pyrazines as "butterscotch, nutty" and "burnt, butterscotch" notes (Table II). 

Crust volatiles were isolated immediately after baking by extraction with dichloromethane and sublimation in vacuo ( ). Application of aroma extract dilution analysis 6) to the acid-free crust extract led to the detection of 31 odorants. After separation and enrichment, these compounds were identified by comparison of the MS/EI, MS/Cl and retention data on two columns of different polarity to reference compounds. Aroma quality was also assessed. The results of the identification experiments (Table I) revealed that 2(E)-none-nal (No. 1), followed by 2(E),4(E)-decadienal (No. 2) and 3-methyl-butanal (No. 3) showed the highest FD-factors in the crust of the chemically leavened bread. Additionally l-octen-3-one, 2(Z)-nonenal, 2(E),4(E)-nonadienal and an unknown compound with a metallic odor contributed high FD-factors to the overall flavor (For a discussion of FD-factors, see Chapter by Schieberle and Grosch, this book). 

Schieberle P. and Grosch W. (1985) Identification of the volatile flavour compounds of wheat bread crust comparison with rye bread crust. Z. Lebensm. Unters. Forsch. 180, 474-8. 

Identification of volatile compounds in strawberries and sliced bread headspace was performed in full scan mode (m/z 30-550). Carvacrol and thymol were identified by a combination of the NIST mass spectral library and gas chromatographic retention times of standard compounds. The rest of volatiles were tentatively identified by their GC/MS spectra. In this sense, the compounds having 90% similarity with spectra in the NIST library were not taken into consideration. Chromatographic responses of detected volatile compoimds (peak area cormts) were monitored for comparative measurements of each compotmd in the studied samples. 

As mentioned in Section II, chromatograms obtained by FID detection and olfactory response are different. Aroma-active compounds usually do not correspond to the major volatile components in the food. As shown in Fig. 9, many important odorants of white bread crust were not visible in the gas chromatogram, for example, 2-acetyl-1-pyrroline (no. 11) (20). This can be explained by the low odor threshold of these compounds. Identification of such minor components (Fig. 10) is a challenging task. 

---