2-Step Multi-Volatile Method (2-Step MVM) for Characterization of Aroma Compounds in Bread

Importance of the topic

Analysis of volatile aroma compounds in fermented foods such as bread is crucial for quality control, product development and understanding the link between processing parameters and sensory properties. Headspace gas chromatography–mass spectrometry (HS-GC-MS) offers automation, reduced solvent use and minimal nonvolatile interference but often underrepresents hydrophilic or low-vapor-pressure compounds in ethanol-rich matrices.

Study objectives and overview

This study aimed to develop and validate a 2-Step Multi-Volatile Method (2-Step MVM) combining two dynamic headspace (DHS) sampling conditions to uniformly extract and enrich aroma compounds over a broad vapor pressure range while removing ethanol and water. The approach was demonstrated for bread samples made with butter or shortening and related lipid ingredients.

Methodology and instrumentation used

Instrumental setup

• 7890 B GC with 5977 MSD (Agilent) coupled to GERSTEL MPS with DHS option, Thermal Desorption Unit (TDU) and programmable temperature vaporization inlet (CIS4).

Sampling strategy

1. DHS step 1 at 25 °C, 650 mL headspace trapped on Carbopack B/Carbopack X to capture moderately volatile compounds (VP 1–20 kPa) while purging ethanol and water.
2. DHS step 2 at 80 °C, 3000 mL headspace trapped on Tenax TA to enrich low-volatility or hydrophilic compounds (VP <1 kPa).

GC-MS parameters

• DB-Wax column, helium carrier, oven 40 → 240 °C, scan m/z 29–300.
• Thermal desorption and PTV injection in splitless mode.

Sample preparation

• Bread made from wheat flour, yeast, sugar, water and either butter or palm oil shortening; crust and crumb analyzed separately.
• Butter and shortening tested before and after heating at 150 °C to profile heat-induced compounds.

Main results and discussion

Recovery and linearity tests in 5 % ethanol–water showed >79 % recovery (r2 > 0.996) for compounds with VP <21 kPa in step 1, except for 2,3-butanedione (27 %). Furan and propanal (VP >40 kPa) required the second step. Comparison of original and new step 1 trap conditions resolved ethanol overload and restored detection of key compounds such as 2,3-butanedione. Screening using AMDIS deconvolution and Aroma Office database identified 44–87 aroma candidates in fats and 60–65 in bread fractions. Principal component analysis differentiated butter vs shortening effects, fermentative vs baking markers, and crumb vs crust profiles.

Benefits and practical applications of the method

The 2-Step MVM delivers a comprehensive aroma profile in ethanol-rich samples without extensive sample cleanup. It is fully automatable, reduces matrix interferences and extends coverage to polar, low-volatility compounds. Applications include flavor quality control, comparative studies of formulations, process optimization and routine QA/QC in bakery and fermented food industries.

Future trends and applications

• Integration with chemometric tools and large-scale databases for rapid classification of product quality and authenticity.
• Application to other fermented matrices (dairy, beverages) and real-time monitoring of fermentation or baking processes.
• Coupling with two-dimensional GC or high-resolution MS for enhanced separation and compound identification.

Conclusion

The 2-Step MVM is a robust, efficient and versatile approach for headspace extraction of diverse aroma compounds in ethanol-rich, aqueous samples. It overcomes limitations of traditional HS methods, enabling reliable detection of hydrophilic and low-volatility analytes and offering valuable insights into formulation and processing effects on product aroma.

Reference

1. M. Markelov, J. P. Guzouski Jr., Anal. Chim. Acta 276 (1993) 235–245.
2. N. Ochiai, K. Sasamoto, A. Hoffmann, K. Okanoya, J. Chromatogr. A 1240 (2012) 59–68.
3. C. Devos, N. Ochiai, K. Sasamoto, F. David, P. Sandra, J. Chromatogr. A 1255 (2012) 207–215.
4. N. Ochiai, J. Tsunokawa, K. Sasamoto, A. Hoffmann, J. Chromatogr. A 1371 (2014) 65–73.
5. K. MacNamara, N. Ochiai, K. Sasamoto, A. Hoffmann, R. Shellie, GERSTEL AppNote 183, 2016.